Foreign Relations of the United States, 1969–1976, Volume E–11, Part 2, Documents on South America, 1973–1976

138. Action Memorandum From the Coordinator for Human Rights and Humanitarian Affairs (Wilson) and the Legal Adviser of the Department of State (Leigh) to Secretary of State Kissinger¹

Washington, October 25, 1976.

Harkin Amendment: Brazil IDB Loan

The Problem

On Wednesday at 10:00 a.m., we must state our position in the Inter-American Development Bank (IDB) on a Brazilian application for an $8 million loan to finance exports.

Background/Analysis

Unless the loan would “directly benefit . . . needy people”, the Harkin Amendment requires the US to vote “no” if Brazil is found to be engaging in a “consistent pattern of gross violations of internationally-recognized human rights”. As this loan does not meet the terms of the exception, we must make a judgment about the present human rights situation in Brazil. L and D/HA feel that the Harkin Amendment compels the US to vote against this loan because of the continuing serious human rights situation in Brazil. H and EB feel that improvements over the last 9 months make it possible to vote for the loan. H feels a positive vote must be accompanied by discussions with the Brazilians at a level no lower than the Foreign Minister describing the reasons for our vote. ARA and EB believe that a “consistent pattern of gross violations” does not currently exist in Brazil.

L and D/HA’s View:

This paper presents a fundamental question: whether we fight against the inflexibilities of the Harkin Amendment by distorting the language and intent of the Amendment so as to narrow the scope of its current application, or whether we apply the human rights criteria in a straightforward way, thus laying the foundation for a later approach to Congress requesting it to modify the Harkin Amendment. L believes that the former course is doomed to failure because the Congress will progressively tighten the legislative strictures. Admittedly the latter course is politically painful. The former course is likely to put the State Department on the defensive; the latter may put the Congress on the defensive. In the long run we believe the latter course is preferable.

Whereas the legislative history of the Harkin Amendment indicates that Congress intended to offer great leeway on the interpretation of the criterion, direct benefit for “needy people,” the same is not true for the other criterion, “consistent pattern of gross violations of internationally recognized human rights.” With respect to the upcoming loan, it is agreed that there was no possibility of using the needy people exception. Thus, we are forced to face the human rights criterion. For reasons stated below, we do not believe that this criterion can be interpreted in such a way as to allow the U.S. representative to vote for the loan.

Brazil has had a long history of serious human rights violations under the military governments which have come to power following the 1964 coup. While L and D/HA recognize that the Geisel administration has recently undertaken to curb such abuses, they perceive no real improvement. Thus they are not prepared to state that there is evidence of a break in what has indisputably been a “consistent pattern” of human rights violations since 1964. The correction of isolated incidents of abuses which ARA notes is not, in their view, sufficient to change a generalized pattern.

Large numbers of persons remain in prison in Brazil without having been charged or brought to trial. Due process of law has not been restored in many cases for persons detained for alleged security or political offenses. The exercise of vigilante justice by “death squads” and the recent activities of right-wing terrorist groups have not been curbed by a military security apparatus or government which should have the capability of doing so. Incidents of gross abuse of human rights still occur in the prisons. Moreover, the Government of Brazil has still not demonstrated any intention of restoring to full force the human rights guarantees included in the 1967 Constitution. These rights have been effectively suspended by executive decree in cases held to involve national security. Police and other action pursuant to the decrees are immune from judicial review under the Constitution. The Government’s failure to support its recent more promising actions with necessary changes in the legal regime suggests that the improvements which ARA cites are only superficial.

ARA’s argument for not finding Brazil to be engaged in a “consistent pattern” of gross human rights violations is based on the assertion that many other Latin American countries exhibit similar tendencies. The question is not, however, how many governments may on occasion fail to respect fully the human rights of their citizens, but rather the frequency and the manner in which any one government may violate those rights. Even if numerous other countries had human rights postures as weak as Brazil’s, that would not affect the question of invoking the Harkin Amendment against Brazil. The standard to be applied is not a country-to-country comparison. The Harkin Amendment forces us to evaluate the situation in each country in terms of certain “internationally-recognized” values, and L and D/HA have concluded that the Government of Brazil has still not taken sufficient corrective measures to end the pattern of abuses in existence since 1964. Contrary to ARA’s suggestion, L and D/HA do not believe that their decision that a “no” vote is legally required for Brazil at this time will logically result in similar recommendations for any more than four or five countries eligible for IDB loans.

ARA and EB’s view:

There are clear indications that President Geisel has made efforts to improve the human rights situation. In January he removed the Army General commanding a district in which flagrant violations had occurred; a considerable number of other personnel changes followed in that command and elsewhere in the security structure. Reports of torture of political prisoners have virtually ceased since that time. Political arrests have decreased significantly since January and where these occur, there is an increasing observance of formal legal procedures through the military tribunal system. A relaxation of censorship has allowed a broad public debate of political arrests and human rights questions. It is far from self-evident, contrary to the apparent views of L and D/HA, that failure to curb police death squads and a lunatic rightist group represents governmental encouragement of or acquiescence in a “consistent pattern of gross violations”—any more than the FBI’s failure to suppress Cuban exile terrorists justifies Castro’s charges that the USG is implicated in their acts. In short, the current situation—which is what we must deal with for both legal and policy purposes—is characterized by sufficiently positive features so as to make a finding of a “consistent pattern” of gross abuse unreasonable.

Further ARA Views:

Neither the law nor our own procedures yet provide us with anything like precise criteria for defining what constitutes a consistent pattern of gross violations. Admittedly, Brazil is not a democratic state. But the case for finding it a gross violator seems based more on impressions and general reputation than on the comparative measurements that are about all we can rationally use in the absence of precise tests.

Currently there is considerably less political violence in Brazil than, for example, in Colombia or Mexico—both countries that have so far escaped opprobrium. Brazil has a far more open and equitable society than does, say, Peru or Guatemala where large Indian populations live in virtually feudal conditions. Brazilians in the vast majority live freer of fear and with more opportunity to express themselves than do the populations of at least 10 other countries in Latin America.

The charges against Brazil focus currently on denial of due process in security cases; the use of torture and the failure to implement constitutional guarantees of human rights. Due process in security cases is denied by all Hispanic-American governments (and Haiti and Jamaica) except for Costa Rica and occasionally Venezuela. The test is the number of such cases. In terms of its size, Brazil’s number is currently small. All police and security forces in Latin America outside the Commonwealth Caribbean, including those of Venezuela and Costa Rica, use torture to extract information. The test here is pervasiveness and bestiality. The evidence is that the use of torture, particularly in its more horrible forms, has been declining in Brazil. As for effective constitutional guarantees, they can reasonably be said to prevail only in Venezuela, Colombia, Costa Rica, Barbados and Trinidad—five out of 24.

We conclude that if Brazil is to be found guilty of a consistent pattern of gross violations, the same judgment will have to be applied to

• Mexico
• Guatemala
• Nicaragua
• El Salvador
• Panama
• Dominican Republic
• Haiti
• Argentina
• Bolivia
• Paraguay
• Uruguay
• Chile
• Guyana, and probably to
• Peru and possibly to
• Jamaica.

Options:

Option I—Vote against the loan.

Pro:—In the judgment of L and D/HA the Harkin Amendment requires a “no” vote because of what they see as a continuing pattern of human rights violations which have not yet been sufficiently corrected by the Government of Brazil.

—A negative vote would satisfy the Congress that the Department was carrying out the intent of the Harkin Amendment and quiet voices criticizing us for disobeying the law.

Con:—Our refusal to support GOB export programs, especially when taken on sensitive human rights grounds, will have an extremely adverse impact on our relationship with Brazil, calling into question the credibility of our recent efforts to strengthen bilateral ties.

—It may draw charges we seek to embarrass Geisel before the politically important November municipal elections, and thus strengthen the hardline military elements opposed to Geisel’s liberalization policies, and undermine his ability to continue progress in the human rights field.

—Other members of the IDB may again accuse the US of introducing a political issue into a financial undertaking as they did after an earlier US “no” vote on a Chile loan, but this is inherent in the Harkin Amendment itself.

Option II—Vote for the loan but instruct Ambassador Crimmins to discuss the question with the Brazilians at the highest levels, but in any case no lower than the Foreign Minister, explaining that our yes vote was based on our recognition of Brazil’s efforts to improve the human rights situation and our hope that improvements will continue. We would discuss Congressional opposition to human rights violations in Brazil and express our hope that continuing improvements will make possible “yes” votes on future loans. H supports this option.

Pro:—This would be a positive approach aimed at support of Geisel’s human rights improvements.

—We might be able to convince all but the Congressional hard liners that we are taking positive steps to encourage observance of human rights in Brazil.

Con:—We can expect an outcry from human rights advocates in Congress and elsewhere that we are ignoring the law. We could try to blunt this by pointing to the ameliorating action taken by the GOB and stating our position that our vote, coupled with discussion with the GOB, represents a positive attempt to encourage further improvements in human rights situation in Brazil.

—President Geisel does not receive Ambassadors on substance and a representation, however phrased, at the presidential or ministerial level would produce strong resentment and charges of interference in Brazilian internal affairs. Indeed, formal representations at any level will run head-on into Brazil’s very real nationalism and risk a closing of hardline ranks—precisely the effect we are trying to avoid.

Option III—Vote for the loan, without making any representations as described above.

Pro:—This would be consistent with our expression of support for Brazilian development goals and would be in keeping with the recent commitments to closer bilateral ties.

—It would avoid any possible counterproductive effects on the political/human rights liberalization process in Brazil which could flow from a negative US position on the basis of Harkin Amendment provisions.

—It would reflect the difficulty we have in reaching a clearcut decision as to the existence of a “consistent pattern” and would be consistent with ARA’s views that no such pattern exists in Brazil.

Con:—We can expect even stronger charges from human rights advocates in Congress and elsewhere that we are violating the law.

—The vote will be used by Harkin Amendment supporters to demonstrate the need for even stronger Harkin-type amendments which the Executive cannot circumvent.

Recommendations:

D/HA and L recommend that you accept Option One and authorize us to vote against this loan with appropriate explanations to the Brazilians and the Congress.

Alternatively, H recommends that you accept Option Two and authorize us to vote for the loan but instruct our Ambassador to explain our vote to the Brazilians at the highest levels.

Alternatively, ARA and EB recommend that you accept Option Three and authorize us to vote for the loan without representations.

Congressional Concerns:

The decision in this case will affect our ability to amend the Harkin language and to prevent similar or stronger language from being added to other legislation, such as authorization for the World Bank and the Asian Development Bank. Our posture thus far on Harkin affected loans has drawn some criticism, especially on the Argentine vote, but our prior consultation and representations to the GOA worked to our advantage.

We are particularly disturbed by a growing mood among some members who have been strong supporters of multilateral aid to institutions and programs to sharply limit the funds available to international lenders (IDB, World Bank, African Development Fund), their rationale being that only a cutoff of funds to these institutions will be effective in view of the Administration’s unwillingness to find any country to be a gross violator of human rights. H believes the human rights activists in Congress will view a “yes” vote on the loan to Brazil as further evidence of the Administration’s intractibility. Unlike the situation in Argentina, where a new government faced a chaotic situation, thus making viable the argument that a consistent pattern of violations of human rights had not been established, the assumption is widespread that Brazil has been a long-term violator of human rights. To point to positive trends in Brazil without being able to say conclusively that a new pattern of adherence to internationally recognized guarantees of human rights has developed will not be convincing in the Congress. Therefore, to defend a “yes” vote, the Department must point to the improvements and changes in the human rights area brought about by the GOB and also argue that we are using our vote to urge the GOB to continue further in this positive direction.

1. Summary: Wilson and Leigh reviewed for Kissinger three options regarding the U.S. Government’s vote on an IDB loan for Brazil: voting against the loan, voting for it contingent on improvements in the human rights area, and voting for it without making any representations.

   Source: National Archives, RG 59, Central Foreign Policy File, P760168–1248. Confidential. Sent through Robinson. Drafted by Hill and Shlaudeman on October 24. Swift, Boeker, and Austin concurred. A typewritten note at the bottom of the last page of the memorandum indicates that Robinson directed EB to vote yes on the loan and that he later instructed Wilson and Shlaudeman to “arrange a low-key approach to Brazilians,” which would note that the United States voted yes because of Brazilian progress in human rights and that U.S. cooperation would continue “as long as this progress is sustained.” Attached but not published is the Department’s undated Human Rights Report on Brazil.