Naval Postgraduate School
Monterey, CA 93943-5100
(408) 656-2863 (phone)
(408) 656-2949 (fax)
Prepared for delivery at the 1995 meeting of the
Latin American Studies Association,
The Sheraton Washington,
September 28-30, 1995
[Initial Draft - Please do not cite without author's consent]
A. Major Questions
The wave of democratization that swept South America since the early 1980s has been well-documented and analyzed. That democratization has affected civil-military relations in the region, but with great variations. At one extreme, Paraguay in mid-1995 is a case in which Army General Lino César Oviedo exercises significant control over the civilian president, Juan Carlos Wasmosy, calling into question whether that country can be considered a democracy. At the other extreme, Argentina's civilian leaders have effectively managed to control their military, albeit with severe challenges and even coup attempts along the way. In dramatic fashion, Argentine civilian leaders have cut the size and budget of the military, dismantled a ballistic missile program (Condor II), established a civilian-led Ministry of Defense, and introduced an all-volunteer military force.
The purpose of this paper is to assess the political influence of the Brazilian military in mid-1995. Specifically, does Brazil's military play a tutelary role? Tutelary is defined as "protecting; watching over; guardian."  Military tutelage refers to a state between democracy and dictatorship, in which the military acts as the guardian. In this state, the implicit or explicit threat of a coup by the military places severe limits on civilians and their scope of action., undermining democracy. According to some observers, Brazil's military play a tutelary role, acting autonomously, with virtually no civilian control over their activities.
The paper focuses on three issues related to Brazil's security: (1) defense expenditures; (2) nuclear programs; and (3) defense industries. The issues were chosen because they were all important to the Brazilian military when they ceded power to civilians in March 1985. Carlos H. Acuña and William C. Smith, in assessing the political influence of the military in Argentina, Brazil, and Chile, argue that the manufacture of armaments and the military budget are the two most important issues (in that order) by which to gauge the political influence of the military. Many other issues could be examined, such as the size, roles and missions of the armed forces, the existence (or lack thereof) of a Ministry of Defense, etc.
The major argument of the paper is that despite the fact that the Brazilian military have retained significant prerogatives, their political influence has continued to decline since 1985. While Brazil's democracy falls short of any ideal type, it would be a mistake to consider Brazil a tutelary democracy today. Certainly Brazil generally meets Robert Dahl's eight institutional guarantees required for a democracy: (1) freedom to form and join organizations; (2) freedom of expression; (3) right to vote; (4) eligibility for public office; (5) right of political leaders to compete for support/votes; (6) alternative sources of information; (7) free and fair elections; and (8) institutions for making government policies depend on votes and other expressions of preference. While formal civilian control of the military is unquestionably desirable for consolidating democracy, such control does not guarantee democracy. Nor does the lack of such control necessarily lead to dictatorship or tutelage by the military.
If it can be demonstrated that Brazil's military exercises declining political influence, the obvious question is "why"? Various reasons are posited for the decline in the political influence of Brazil's military: (a) the process of democratization; (b) the neo-liberal economic model and the restructuring of the Brazilian state; and (c) external conditions.
The questions raised in this paper are important because their answers may help explain a seeming contradiction: a military with significant prerogatives yet limited political influence. Furthermore, the answers can yield insights that provide an initial agenda for improving civilian control of the military and further consolidating democracy in Brazil. These answers could apply to other countries, such as Chile and Argentina.
The analysis is based on extensive research and interviews conducted in Brazil, Chile, and Argentina between August 1993 and May 1994, and in Brazil and Argentina in August and September 1995. Interviewees included civilian policy-makers, military officers, academics, and journalists.
B. Historical Background
The armed forces of Brazil have played an active political role ever since they helped overthrow the empire in 1889, and ushered in the Old Republic (1889-1930). From 1930 until 1964, they asserted their "moderating power" and intervened frequently in the political process. In 1964 the Brazilian military ousted the civilian president and ruled for 21 years.
The military regime was guided by a national security doctrine with two major elements. The first was a broad definition of security, which included not only the defense against external aggression, but the internal defense against insurgencies and Communism. Utilizing repressive measures, the military successfully countered domestic insurgencies from 1967 through 1973. The second component of the national security doctrine was economic development. Under the military, the scope of the state in the economy grew considerably with the deepening of Brazil's industrial base. Growth rates exceeded 11 percent a year between 1968 and 1973, helping to legitimize military rule.
The armed forces left office in March 1985, but even under civilian rule they have continued to assert themselves politically. In mid-1995, the armed forces retain many of the prerogatives they possessed upon leaving office. There is little formal civilian control of the military, except at the presidential level. In many ways, the military continues to be a "state within the state." 
Despite such an activist political role, the political influence of the armed forces has diminished since leaving office, as their relationship to the state has been gradually re-structured. Congress and civilian ministries have become more involved and influential in broadly-defined security issues. The military have been forced to compete with civilians for extremely limited resources, and have been unable to halt a continual decline in their share of government expenditures. The 1988 Constitution, while preserving the external and internal roles of the armed forces, actually changed the manner by which the military could exercise its moderating power, placing them under presidential authority.
C. Theory and Brief Review of Literature
No analysis of Brazil's national security policy is complete without some understanding of the military's role. In Rethinking Military Politics: Brazil and the Southern Cone, Alfred Stepan argued that although the Brazilian military ceded power to José Sarney (president, 1985-1990), it retained significant prerogatives. Zaverucha, a Brazilian political scientist, updated Stepan's model, and stated that on nine important items, the armed forces in Brazil had preserved their prerogatives: (1) peak intelligence agencies were controlled by a military chain of command, with weak civilian review boards; (2) military officers on active service participated in the cabinet; (3) there was no Defense Ministry; (4) there was a lack of legislative routine of detailed hearing on defense matters; (5) the police were under overall direct command of the military, and most local police chiefs were active-duty military officers; (6) the military played a major role in setting the standards for promotion; (7) military personnel were unlikely to face judgment by civilian courts; (8) there was a potential for military autonomy during internal unrest; and (9) there was military control over specific areas of economic activity (aviation, aeronautics industry, navigation, etc.).
Zaverucha's conclusion was that from 1985 to at least 1993,
"Civil-military relations did not advance, and practically nothing substantial was done to curtail military prerogatives. Indeed the military continues openly challenging the constituted civilian authorities. Leaders of the democratic transition opted out of confronting the military over its scope for autonomy. In those rare cases when the Executive and Legislature tried to impose their will the military threatened to disrupt the transition and so the civilians simply resigned themselves to the continuation of military prerogatives." 
A contrasting opinion was offered by Wendy A. Hunter, who in a brilliant article argues that "Over time, democratically elected politicians have successfully contested the power of the military over a broad range of issues and narrowed its sphere of influence."  Hunter's theoretical approach (rational actor model) is different from that of Stepan, Zaverucha, and others (game theory; historical institutionalism), as are her conclusions.
The conclusions reached in this paper suggest that while it may be important to analyze military prerogatives, such analysis does not allow for the broad generalization that civil-military relations have not advanced (Zaverucha). A more important notion in politics is that of influence, which is broader than prerogatives. Prerogatives may affect influence, but they are only a first step in determining influence. Influence is broader in that it includes resources, capacity, and even the will to exercise prerogatives and utilize resources.
For example, while the military may have control over a specific sector of the economy (Zaverucha's prerogative #9) such as the space industry, if the military are denied resources to develop or sustain that industry, their prerogative is essentially hollow. In Brazil, this has been the case. If the military are given resources to support that industry, but fail to do so, again their prerogative is meaningless.
As a separate example, Brazil has never had a Ministry of Defense (MOD - prerogative #3). President Fernando Henrique Cardoso (inaugurated 1 January 1995) stated that he would create a Ministry of Defense, but current efforts are stalled in the Armed Forces General Staff (Estado Maior das Forças Armadas - EMFA), which has been tasked with the initial planning. In a recent interview with the author, Brazil's Minister of the Secretariat for Strategic Matters (Secretaria de Assuntos Estratégicos - SAE), Ronaldo Sardenberg, predicted that the MOD would be created by the end of 1996.
In contrast, Chile has had a Ministry of Defense for decades, which is currently headed by a civilian, Edmundo Perez Yoma. Civilians staff the undersecretary positions, and are increasingly being brought in at lower levels. A strict prerogatives approach would suggest that Chile, therefore, has greater civilian control over its military than Brazil, a dubious claim at best. Clearly, context is important. Chile's MOD operates in an environment where the Chilean military enjoy substantial political influence, especially on matters relating to their direct institutional interests (promotion and retirement, budget, amnesty for human rights abuses, etc.).
A third and final example is that of human rights abuses (prerogative #7). Under the Brazilian military regime, an estimated 136 people "disappeared" and many more were tortured by the military. Under the Amnesty Law of 11 August 1979, Brazil's military could not be held accountable for any human rights abuses committed prior to that time. This suggests that the Brazilian military continues to exercise influence on an important issue.
On 28 August 1995, the 16th anniversary of the Amnesty Law, President Cardoso signed a proposed law (projeto de lei) that recognizes the death of the 136 political prisoners under the military regime, and calls for indemnifying the families of those killed and tortured. Cardoso took responsibility, in the name of the State and as commander in chief, for the deaths and injuries. The payments would range between R$ 100,000 and R$150,000 (US$ 105,000 to US$ 160,000) per family; the precise amount would be established by a Congressional panel. Brazil's Congress will have 45 days in which to vote on the measure, and passage is expected in both the Chamber of Deputies and the Senate. 
The proposed law was opposed in a signed article by Army General Murillo Neves Tavares Silva, commander of the 7th Military Region. The general was due to retire, but resigned instead, in protest. The Minister of the Navy, Admiral Mauro César Rodrigues Pereira, also criticized the payments, claiming that "there are many disappeared ones who are still alive," but he later backed down from his opposition. Despite such high-profile resistance to the president's indemnity bill, that opposition did not resonate within the armed forces, among both active duty and retired officers. Indeed, the Army Minister, General Zenildo de Lucena, remained silent. Several days earlier, he had taken the president to the Vila Militar in Rio de Janeiro to explain to officers that the indemnities would not affect the Amnesty Law. 
Brazil's military are therefore still protected by the 1979 Amnesty Law, and it would be easy to argue that they are playing a tutelary role on matters relating to human rights abuses. However, it could also be argued that President Cardoso's proposed law is a statesman-like gesture, one that seeks to heal past wounds and avoids ressentiment politics. The signing ceremony was attended by the Chief of the Military Cabinet, General Alberto Cardoso, who literally embraced the widow of former Congressional Deputy, Rubens Paiva, one of the "disappeared." That embrace captured the spirit of Cardoso's proposal.II. MAJOR ISSUES
The following section assesses the level of military influence on three important issues: defense expenditures, nuclear programs, and defense industries. As mentioned earlier, each issue is important to the Brazilian military, and serves as a test of the hypothesis that the political influence of the military has declined. This is followed by a section which seeks general explanations for the findings.
A. Defense Expenditures
In determining the level of military influence in Brazil since 1985, few issues are more important than that of defense expenditures. One would expect an influential military to preserve, if not expand the level of defense expenditures. Instead, Brazil's military have been unable to counter deep cuts in their budget - whether measured in absolute or relative terms. This suggests a lack of influence in this critical issue-area.
Perhaps the most dramatic decline in military indices has been in the total number of armed forces, from 496,000 in 1985 to 296,000 in 1993. Otherwise, much of the data on Brazil's military expenditures from 1985 to the present are contradictory, but a common pattern is one of increasing expenditures between 1985 and 1990, and decreasing expenditures between 1990 and the present. According to the U.S. Arms Control and Disarmament Agency, Brazil's military expenditures in constant 1993 dollars fell from $8.8 billion in 1990 to $5.8 billion in 1993. This corroborates a common argument that the military were able to retain some influence under President José Sarney (1985-1990), before facing the budget ax of President Fernando Collor de Mello (1990-1992). Sarney, who became president under highly unusual conditions, lacked broad backing in Congress, and sought support from the military, although the level of that support is often exaggerated in the literature.
Methodologically, it is difficult to assess Brazil's military expenditures. Figures attributed to Brazilian defense expenditures generally understate their true value. The accuracy of most data is complicated by the following: high rates of inflation since the 1950s; the secrecy surrounding the funding of various military-related projects; personnel costs that are sometimes hidden in other budgets; and the common practice of mixing the accounts of the national treasury, the Central Bank, and the Banco do Brasil.
Despite the methodological problems noted above, there is consensus that Brazil is now among the countries of the world with the lowest levels of military expenditures (in relation to Gross Domestic Product or the budget). In terms of military expenditures as a share of central government expenditures, in 1993 Brazil ranked 133rd out of 166 countries, according to ACDA. Brazil ranked 140th in terms of military expenditures as a percentage of GNP. Within South America, no country ranked lower. 
Wendy Hunter presents evidence that the military's share of the national budget has declined, from over 20 percent in 1985, to under 15 percent in 1992. According to Veja, in 1993 Brazil's defense expenditures as a share of all government expenditures were at the lowest level since independence in 1822 (the highest was in 1864 and 1865, at the early stages of the Paraguayan War, when defense expenditures accounted for 49.6 percent of all government expenditures).
There is general agreement that the levels of Brazil's defense expenditures in relation to gross domestic product have dropped from an average of 2 percent in the 1960s, to 1.5 percent in the 1970s, to less than .5 percent in the 1980s and early 1990s. In 1993, according to some accounts, that rate was lower than .5 percent.
The armed forces in Brazil have had their minor triumphs on budget issues. In 1993, they received $600 million to install, over eight years, a system of radar tracking in the Amazon. In early 1994, the Itamar Franco administration (December 1992-31 December 1994) announced that it would cut $22 billion in the federal budget, dividing the cuts equally across ministries. The military ministers reacted quickly, and went directly to the president to criticize the proposed cuts. They succeeded in reducing at least $300 million of their proposed cuts. Such funding, however, cannot compensate for the larger declines in the military budget.
B. Nuclear Programs
Under Brazil's military regime, nuclear programs were controlled almost exclusively by the military. In 1977, Brazil's military regime signed a major agreement with Germany to acquire German nuclear ("jet nozzle") technology. In the following year, Brazil's military began a secret program to develop an atomic device.
Under President José Sarney, the first of several significant nuclear accords were signed with Argentina. Sarney's successor, President Collor de Mello, revealed the secret bomb project in 1990, thereby embarrassing the military. Later, Brazil's Congress undertook a special investigation of the entire nuclear program.
President Collor signed various nuclear-related agreements with Argentina, culminating in December 1991 with the quadripartite comprehensive safeguards agreement between Brazil, Argentina, the International Atomic Energy Agency (IAEA), and the Brazilian-Argentine Agency for Accountability and Control of Nuclear Material (ABACC). The quadripartite agreement permits the inspection of all the nuclear installations in Argentina and Brazil by the IAEA. In March 1994, Brazil's Senate finally ratified the quadripartite treaty after many delays. The Senate's ratification came only after significant arm-twisting by then-Minister of Foreign Relations, Celso Amorim.
Brazil has now ratified the Treaty of Tlatelolco, as has Argentina. On the other hand Argentina has signed the Non-Proliferation Treaty, but Brazil has not. At least one expert, Paulo Wrobel, predicts that Brazil will sign the NPT by the end of 1996.
The nuclear accords point to the increasing influence of Brazil's civilians - especially those within the Ministry of Foreign Relations (often referred to as Itamaraty) and SAE. Itamaraty was at the forefront of all nuclear negotiations, and has now developed a cadre of specialists in the area of nuclear and space technology, which includes some of the brightest of the Itamaraty officials. It is no coincidence, therefore, that Ronaldo Sardenberg, a seasoned diplomat, was selected by President Cardoso to head SAE, which is in charge of all Brazilian nuclear programs. The armed forces, while obviously interested actors, have been displaced by SAE and Itamaraty as the dominant actors on political decisions relating to nuclear matters.
C. Defense Industries
In the early 1980s, Brazil emerged as one of the leading armaments exporters in the developing world. From 1985 to 1989, it was the eleventh-largest exporter of arms. Brazil exported arms to at least 42 countries, in all regions of the world. By far the largest regional market was the Middle East, to which Brazil sold approximately 50 percent of its arms from 1977 through 1988. According to an estimate by the Stockholm International Peace Research Institute (SIPRI), nearly 40 percent of all Brazilian arms transfers from 1985 to 1989 went to Iraq.
After 1988, Brazil's arms industry nearly collapsed, due to the termination of the Iran-Iraq war, a reduction in global demand for armaments, and the decline in state support for the industry. In early 1990, Brazil's two major manufacturers, Engesa and Avibrás, filed for a concordata - the equivalent of chapter-11 bankruptcy. By mid-1995 it appears that Brazil's arms industry has virtually disappeared, and it is highly unlikely that it will return to the robust form of the mid-1980s. Avibrás has paid off much of its debt, and is the most viable of the three large companies. Today, it is involved in the production of primarily civilian products, and seemingly is a successful case of conversion. Engesa has been dismembered, with some of its companies sold to private interests, and with ordnance-related firms taken over by the state, and integrated with Imbel (Indústria de Material Bélico - Industry of War Materiél). Embraer was privatized at the end of 1994, although there were many within the Aeronautics Ministry who resisted that measure. Embraer's former management has come under severe criticism, and its failure to win the U.S. JPATS contract for 711 Super Tucano trainers for the U.S. Air Force and Navy places the company in a precarious financial state. 
Despite considerable pressure by Brazil's armed forces to maintain a vibrant arms industry, that industry has virtually collapsed. Under President José Sarney, considerable state support continued to be provided to the industry. Under President Collor de Mello, however, much of that support was withdrawn. President Fernando Henrique Cardoso has yet to provide much state support for the industry, given his economic policies of austerity and privatization.
The issue-area approach utilized above seeks to capture the dynamics of military influence. The approach complements the notion of prerogatives, which serves as a useful first step in analysis. The conclusion reached in the three issue areas is that the political influence of Brazil's military is indeed declining. But why?
Uni-causal explanations should be eschewed in examining military political influence. There are a wide variety of factors that affect the level of that influence, including those of an historical, institutional, and cultural nature. The following section focuses on three factors that have influenced the Brazilian military's political influence: political process, economic model, and external conditions.
A. Political Process
The most important factor affecting the military's political influence are domestic and political in nature. As Wendy Hunter has so adeptly demonstrated, the very process of democratization and competition affect the level of military influence. Brazil's military are still major actors on many issues, but they are forced to share power and compete for scarce resources with civilians.
In terms of political society, Congress is increasingly challenging the armed forces, as is the president and the executive branch. Itamaraty and SAE wield significant power on many issues relevant to the military (including peacekeeping operations). Itamaraty is represented in each of the ministries, giving it access to a broad spectrum of domestic and foreign policies.
Civil society is playing a broader, but modest role, such as on the issue of human rights. Philip Oxhorn has addressed the topic of the popular sectors in Chile, and concluded that civil society was strengthened under military rule, and that "politics and society have been changed in fundamental ways. The ways in which the state and, in particular, political parties relate to civil society are very different from what they had been in the past."  The same applies to Brazil. The press, for example, has played a major role in Brazilian affairs, from the impeachment of President Collor, to criticism for the nuclear-powered submarine program.
B. Economic Model and the Restructuring of the Brazilian State
Perhaps the most overlooked explanation for change in the political influence of Brazil's military is that of a neo-liberal economic model. The statist policies perpetuated by the military regime rewarded those groups closely aligned to the state - including the military as an institution. The neo-liberal economic policies imposed by the Fernando de Collor administration were particularly harsh on the military. Even since Collor's ouster in December 1992, the federal government has been less willing to reward the military through larger budget deficits or special taxes - common practices until 1990.
As finance minister and president, Fernando Henrique Cardoso has sought to streamline Brazil's fiscal and monetary policies, with the neo-liberal real plan, named after the currency introduced in July 1994. Such efforts will almost certainly continue to affect the military's budget. It is no surprise, therefore, that some Brazilian military officers preferred the resurgent nationalism and statism of Luís Inácio Lula da Silva, Cardoso's main challenger for the presidency on 3 October 1994.
Most of the literature ignores the economic model argument, which today manifests itself most clearly in Argentina. The Left, quick to embrace statism and the import substitution industrialization model, created and perpetuated a model that benefited, ironically, the military as an institution. One of the additional ironies of late 20th century history in Latin America is that neo-liberal economic policies implemented by civilians have tended to reduce military influence.
C. External Conditions
Brazil's security and foreign policy interests have been affected by numerous changes in the early 1990s: the transformation from bi-polarity toward multi-polarity in the international system; greater integration with Argentina, Uruguay, and Paraguay (Mercosul), as well as with other countries in South America; and a liberalizing economic model that has forced an opening of Brazil's markets. Closely related to the neo-liberal economic model, therefore, is a changing international system that changes the rewards and costs of various courses of action.
For example, in order to maintain cordial diplomatic relations with Germany, Brazil's ministry of foreign relations pressured its own Congress to ratify the quadri-partite nuclear agreement. Increased integration with Argentina has reduced the perception of threat to Brazil's south, and allowed for the movement of some army units in the southern state of Rio Grande do Sul to the north of Brazil. Brazil's desire to play a larger role on the international scene drives the country to participate in peacekeeping operations around the world. Those peacekeeping missions receive substantial funding and control from the civilian-led Ministry of Foreign Relations.
As a parallel example, Brazil announced in early 1994 that it would adhere to Missile Technology Control Regime (MTCR) guidelines, and even seek membership in the regime. This monumental shift in Brazilian policy was resisted by the Aeronautics Ministry. As a virtual prerequisite for receiving sensitive technologies from the United States, Brazilian authorities created a civilian-led space agency to control space programs with dual-use applications. It is too early to tell whether that civilian oversight will be effective. Again, Itamaraty played the major role in negotiating Brazil's new space policy.
The demands placed on a country like Brazil or Argentina in joining a regimes like the NPT and the MTCR are great, and they help force the development of a bureaucracy that is knowledgeable in these areas. In my interviews with officials in the Argentine and Brazilian foreign ministries, those associated with space and nuclear matters have expressed incredulity at the number of issues that they are being forced to confront as members of the NPT and MTCR. They are developing an expertise that no one in the Argentine bureaucracy possessed even five years ago. This seems to underscore the fact that the process for providing effective civilian oversight of the military is a gradual one, and that the interaction between external and domestic variables can affect that oversight.
The political influence of Brazil's military is waning, and it can no longer be stated that Brazil is a tutelary democracy. Although they continue to enjoy considerable prerogatives, Brazil's military have been unable to effectively defend their budget, nuclear programs, and defense industries. Indeed, the military's budget has declined to its lowest level ever; their nuclear programs have been placed under international monitoring; and the defense industries, once coddled by the state, have either been dismembered (Engesa), privatized (Embraer), or allowed to operate without significant state support (Avibrás).
The confluence of various factors have limited the military's influence. Those factors include (a) the very process of democratization and the political competition in that process; (b) a neo-liberal economic model that has shrunk the scope of the state; and (c) a changing international environment that further removes perceptions of threats and rewards civilian actors such as those in the ministry of foreign relations, SAE, Congress, and even in the private sector.
The flaws in Brazil's democracy have less to do with the role of the military, and more to do with other shortcomings, many of which are institutional in nature: weak political parties, a recalcitrant Congress, etc. Brazil's military have little disposition to become "military as government," involved in the complex problems facing the nation. This is a lesson that they have learned after 21 years of dictatorship. The military are content to focus on the many issues facing the "military as institution." 
Finally, a caveat. The reduction in military influence has been achieved in an unintended manner. Brazil does not have a clearly-articulated defense policy, and internal security is in a state of utter chaos. There is still a dire need for civil society and political society to take responsibility for defining the role and missions of the military. For example, in the presidential, legislative, and gubernatorial campaigns in 1994, there was virtually no debate concerning national security policy. None of the political parties, with the exception of the Workers' Party, articulated a clear position on security matters. While a number of civilians are experts in defense matters, their influence is negligible. More ominously, there is virtually no tradition of Congressional oversight of the military, and only a minuscule defense-related bureaucracy. Civil society continues to demonstrate a complete lack of interest on issues related to defense. The modest attempts by the armed forces to re-evaluate their role, structure, doctrine, strategy, and tactics have been conducted in a vacuum. The creation of a Ministry of Defense has been recognized by many as a necessary condition for establishing effective civilian control of the armed forces, but no concrete step has been taken to make such a reality.
It is here that the "prerogatives approach" (Stepan, Zaverucha) can help identify an agenda for action. Zaverucha's list, cited earlier, could serve as a point of departure. An agenda for action would include the creation of a Ministry of Defense as a necessary but insufficient step for civilian control of armed forces. It would also include the revamping of the intelligence agencies so that they are under de facto civilian control. Again, the purpose of this exercise would be for political society to explicitly address the issue of security, and to provide for effective civilian oversight of the military. A first step could be taken with the current revision of the Constitution.