APPROVED FOR RELEASE 1994
CIA HISTORICAL REVIEW PROGRAM
18 SEPT 95
A newcomer to intelligence takes an uninhibited look at the community's finished product.
Janet Hill Merkle
The question of the extent to which the U.S. intelligence assessment of foreign situations is biased by already established government policy toward them is. a delicate one and in all its ramifications too complex to be broached by a junior trainee like the present writer. But any student with access to the materials can sample one aspect of it by separating off a particular fairly clear situation and examining the community's finished reports on it for signs that their objectivity has been impaired by the policy makers' views. This is what I have done, taking as sample the National Estimates, articles in CIA's Current Intelligence Weekly, and State's INR publications concerned with the situation in Portuguese Angola over a period of about two years.
Here the established U.S. policy, first publicly declared by Ambassador Stevenson in the United Nations in March 1961, is one of support for Angolan self-determination and of opposition to Portugal's resolve to keep the colony, which was legally declared a "province" in 1951. Evidence that the finished intelligence reports had been affected by this policy was found in their phrasing and emphasis, in their omission of facts reported from the field (by the U.S. and British attaches, the American consul in Luanda, and the clandestine services) which could be cited in favor of the opposing Portuguese policy, and in their measurement of Portuguese performance against standards set up by the U.S. policy. In these respects the National Estimates showed the least anti-Portuguese bias, the INR publications the most.
Although the four estimates between 1959 and 1962 which treated the subject of Portugal's overseas territories seem to be for the most part objective, they do contain a few manifestations of bias. In an NIE of 21 July 1959, it is said that
Portuguese policy is a curious mixture of indifference to the lot of the native, half-hearted efforts to elevate him from savagery, repression of all dissident voices, and cheerful assertion that in fact no problems exist.
Hyperbole and ridicule of this kind are clearly inconsistent with objectivity. It is possible, however, since the estimate antedates the public declaration of U.S. policy, that this is an instance of personal rather than policy bias.
An NIE of 11 April 1961 estimates that Salazar may take some measures designed to give the impression of liberalizing the colonial regime.
This statement implies, first, that no measures of reform had theretofore been taken, and second, that any reforms in the future would be made only in order to influence world opinion. But reports from the field show that some reform measures had already been taken and that currently schools for Africans are being built rapidly and public health facilities greatly expanded and improved. It seems clear that the Portuguese have concluded, whether reluctantly or not, that reforms must be made if they are to stay in Angola; and they are determined to stay. Given their lack of resources and the conservatism of the government at home and in Angola, it is not surprising that the reforms are neither sweeping nor rapid. But it is unrealistic to assume that what measures are being taken are designed only to impress international opinion. The Portuguese have never been terribly concerned by adverse public opinion before, and it is unlikely that they would now base their policy on it.
Several passages in the estimates also leave an exaggerated impression of the "rigid, harsh, and penurious" conditions under which the average Angolan lives. Conditions in Angola are far from utopian for the African, but the field reports supply evidence that they are not so bad as generally believed. This evidence is not presented in the NIE's. On the other hand, it was only in an NIE, of all the finished reports, that a reference was found to the "unusual cruelty on both sides" in the rebellion.
Many of the estimates' conclusions were the same as those which have been reached by U.S. policy makers-that the Portuguese are likely to have continuing troubles in Angola, for example, and that reform will have to be considerable if the situation is not to become explosive. One cannot say whether this is because policy influenced intelligence, because intelligence influenced policy, as it should, or because the evidence led both independently to the same conclusions.
Current Intelligence Weeklies
Examining seventeen articles in the Weekly from May 1960 to April 1962 covering the Angolan situation, I found no evidence of a lack of objectivity prior to the U.S. declaration of policy, but beginning in April 1961 there was a prejudicial omission of mitigating material contained in the field reports. In these articles there are several references to "brutal repression" on the part of the Portuguese armed services and civilians. According to reports from State and Army personnel on the scene, the attacks of the African terrorists have been equally brutal. For example, one State despatch said that Africans were "killing white families, mulatto families and native Africans who had not joined their movement with equal and impartial brutality." Reports of African brutality have also appeared in the New York Times. This the Weeklies do not mention anywhere, leaving the impression that there was no provocation whatever for the Portuguese reprisals.
There is also considerable discrepancy between the articles and field reports with regard to the extent of Portuguese brutality. In the panicky month following the uprising, according to the latter, there were indeed indiscriminate acts of cruelty and reprisal on the part of the Portuguese authorities and civilians in Angola, and some groups of innocent Africans were killed or driven from their homes in both official and vigilante-type actions. The reports go on to say, however, that since the Portuguese army moved into Angola in force there have been only isolated instances of such reprisals. The army officers in the north, feeling that the natives in that area had some reason for revolt, have instituted a policy of "psychological rehabilitation." They are laying out new villages where they can protect the natives, assisting in the construction of homes and schools, and encouraging rebels and refugees to return to their homes with no punishment. The civilian Portuguese often regard all Africans as rebels or potential rebels, but the army discourages this view and is trying to avoid indiscriminate acts of violence. The Weekly articles do not mention this effort of the Portuguese army to deal with the situation; they make no distinction between military and civilian actions. They also do not mention the statements in field reports that Portuguese retaliation and cruelty have been greatly exaggerated.
Although the INR publications carry a caveat that they do not necessarily reflect Department of State policy, the two Research Memoranda and the one longer Intelligence Report covering the rebellion in Angola do seem to have been written in support of policy. One of the Research Memoranda begins by setting up the standard,
The US had hoped these reforms would set the stage for (1) a marked improvement in the status of Africans, and (2) eventual self-determination in the provinces.
and then proceeds to measure Portuguese performance against this U.S, "hope," reporting for example that
. . . the Portuguese seem to have little understanding of, or inclination toward, the positive programs needed to prepare either the African for full participation in modern political or economic life or the overseas provinces for ultimate self-determination.
and concluding that
The rigid attitude of the present government offers no hope that the principle of self-determination will be accepted in the near future.
Thus Portuguese policy is judged in the light of what the U.S. policy maker thinks should be done in Angola. Moreover, the publications openly show their anti-Portuguese bias throughout. They refer continually to "brutal repression" without mentioning the provocation of African terrorism and cite alleged traits of Portuguese national character:
The recently reinforced police, in conjunction with the large military garrisons, can and have suppressed nascent subversive movements with characteristic Portuguese thoroughness and ruthlessness.
They speak of Portuguese reforms with tongue in cheek and point again and again to the disparity between principle and fact in the Angolan society. Disparities are evident, but unless the field reports are all wrong reforms are really being undertaken.
It is interesting to see the great discrepancy between the reports of the consul in Luanda and the INR publications. The consul is not all-out pro-Portuguese; he is quite critical of many aspects of the policy in Angola. But he also brings out things that show the Portuguese in a favorable light, for example the steps toward economic and educational reform, the good race relations which obtained in Angola until 1961. He stresses his conviction that statements about Portuguese brutality and the extent of rebellion have been greatly exaggerated, a conviction substantiated by reports from the British and American attaches. But these points do not appear in the Department's intelligence publications. They are not explicitly discounted or denied; they are simply ignored.
As a trainee, I have been led to believe that intelligence should present and analyze the facts in any situation in as completely objective a way as possible, and further that it should present all of the relevant facts regardless of whether or not they support a given government policy. In varying degrees the publications on the Angolan situation I examined did not live up to this ideal but manifested an anti-Portuguese bias and disregarded information favorable to the Portuguese viewpoint reported from the field. On the basis of the material that was available to me I would therefore conclude that the intelligence community's coverage of the Angolan situation has not been completely objective and has not presented all the relevant facts. If this is true, it raises a serious question in my mind: If policy makers do not receive complete reports and objective estimates from the intelligence community, to whom do they turn for them?