FAS | Government Secrecy | e-Prints ||| Search | Join FAS

©Journal of Peace Research, vol. 32, no. 2, 1995, pp. 213--228

This article analyzes the question of information control and how it influences the conduct and the study of international relations. Three theories are advanced as possible explanations for secrecy: The first, the External Threat explanation, suggests that government secrecy is designed to protect sensitive information from external enemies. The second explanation, the Bureaucratic Politics explanation, views secrecy as a relatively unsystematic process that results from the (collectively) irrational features in any government bureaucracy. The third explanation, the Internal Threat approach, argues that government officials use secrecy to mislead the populations of their own countries. These three explanations are applied to a case study of US information policy with regard to the Congo Crisis of 1960-61. The article concludes that the Congo case tends to support the Internal Threat explanation of secrecy.

1. Introduction

Whatever its larger consequences, the end of the Cold War has proven beneficial for social scientists and historians. Access to previously secret information in the former USSR and other ex-communist countries has enriched our understanding of how these countries functioned, both in their domestic and foreign policies. In the past several years there has been a much more limited loosening of information access in the USA, and this access has provided exciting possibilities for research. Social scientists are only beginning to sift through the new, previously secret materials, but it is generally believed that they will significantly influence our understanding of the Cold War (Alperovitz & Bird, 1994).

The question of access to government information is obviously one of major importance for the researcher, especially in political science and related disciplines; however the study of information policy has been somewhat uneven. During the 1960s and 1970s, there was considerable interest in the question of secrecy, among journalists, policymakers, and scholars. Many of these, to be sure were descriptive or advocacy oriented. However, there were also serious studies that systematically compared secrecy practices among Western democracies and placed secrecy in the broader perspective of public policy analysis (see for example essays in Galnoor, 1977; & Franck & Weisband, 1974). Even such prominent figures as Carl J. Friedrich (1972, 1977) and Theodore Lowi (1977) contributed to this discussion.

However, such studies went out of fashion after 1980, and there has been a distinct decline in research on secrecy. Ironically, declining interest in secrecy among scholars coincided with a substantial increase in the practice of secrecy by governments, especially in the USA.⁽¹⁾ Despite its continued importance, there has been no systematic study that seeks to explain the causes of government secrecy. The present essay will analyze the causes of secrecy, especially in the areas of international relations and foreign policy. Specifically, we will consider the following questions: Why does a democratic government fail to release all of its information? Why are some government documents classified and retained as secret, while others are not? Finally, what are the implications of secrecy for methodologies of political research? The article considers these questions in the context of an empirical case study of US information policy during the Congo Crisis.

Before embarking on the empirical discussion, I present three separate explanations for government secrecy: The first, the External Threat explanation, suggests that government secrecy is designed to protect sensitive information from foreign powers and other external enemies. The second explanation, the Bureaucratic Politics explanation, sees secrecy as a relatively unsystematic process that results from the (collectively) irrational features in any government bureaucracy. The third explanation, the Internal Threat approach, argues that government officials use secrecy to mislead the populations of their own countries.

2. The External Threat Approach

This approach assumes that state officials are motivated by a concern with the national interest and, especially, national security. In order to maintain and enhance this security, officials must withhold information from potential foreign rivals. Certainly a government cannot reveal national defense plans to its enemies, at least not without grave risks. Also, intelligence agencies must protect the their sources and methods. Moreover, a state must withhold secrets, such as the details of trade negotiation strategies, even from relatively friendly countries with which it is, from time to time, in competition. So it is the governments of foreign countries, not members of the public, that are the objects of government secrecy. This External Threat explanation is widely promoted by the government itself, but some academics also subscribe to it. In his study of the Central Intelligence Agency, Rodrick Jeffrys-Jones (1989, p. 3) writes:

[A]llegations of excessive secrecy have in fact whipped up legislative fury since the early days of the Republic. Here there is real scope for misunderstanding. In keeping certain information hidden from foreign powers, the director of the CIA must necessarily refrain from instructing (and may even actually deceive) the American people.

Deception of the public is an unfortunate yet inevitable side-effect of this process. This is, by far, the most widely accepted explanation for government information policy, and it is familiar to students of strategic decision-making processes.

The methodological implications of this External Threat approach should be somewhat encouraging to the researcher. The approach implies that the government will make no effort to withhold information if the information is already known by the enemy powers. Moreover, the government will not usually withhold documents for very long periods, and as soon as certain practices or strategies are fully obsolete, the government will declassify the relevant documents. Above all, it is assumed that governments will not withhold information simply because the information might embarrass state officials or generate public opposition. Thus, the External Threat approach implies that secrecy will necessarily be limited in scope and circumscribed with regard to the types of information that are useful to the social scientist. Excepting researchers who focus on military or intelligence history (especially those who focus on the details of specific strategies), the social scientist should not be unduly impeded by government secrecy.

3. Bureaucratic Politics

Most theories assume that foreign policy, whatever its basis, is the result of a coherent strategy that is designed to achieve some assigned objectives. The Bureaucratic Politics approach challenges this assumption and argues that policy-making in all areas -- including information policy -- is rarely quite so simple. One of the most basic aspects of any bureaucratic entity is adherence to standard operating procedures, and this adherence can sometimes result in suboptimal or irrational outcomes (Allison, 1971; see also the critique by Bendor and Hammond, 1992). Bureaucracies are accordingly viewed as lumbering, rule-ridden, mechanisms incapable of decisive, coherent action (Joseph, 1981, p. 25), and they tend to act in particular ways because they have always acted in such ways. Another aspect of bureaucratic functioning is competition among various agencies, subagencies, and personalities. In its most extreme form, the Bureaucratic perspective implies that there is no guiding objective in government activity; policy formulation results from inertia and from incoherent maneuvering among diverse agencies.

Let us apply this perspective to secrecy. In certain foreign policy agencies, secrecy is standard operating procedure. Intelligence agencies, for example, routinely classify a wide range of documents simply because agency rules call for such classification, even though much of this classified material must surely cover mundane subjects of no interest to academics, journalists, members of the public, foreign enemies, or anyone else. Researchers who have used government archives can readily attest that many previously classified materials contain little or no important information.

Moreover, the rigors of interagency competition can influence document classification procedures. Government officials (or agencies) will seek to deceive each other, and bureaucratic conflict may be a significant factor in the process of document classification. Sometimes, officials will withhold documents in order to gain advantage in inter-bureaucratic competition or, in other cases, because of personal feuds among officials. Paul Anderson (1981, pp. 745--47) applies the Bureaucratic Politics model to the question of government information practices. He notes that, even in private meetings, officials must tailor their arguments in order appear more persuasive and to be consistent with acceptable standards of policy-making etiquette.

For example, Anderson notes that government officials may be motivated by parochial bureaucratic factors, interest group pressures, or other political considerations; however, the accepted standard of bureaucratic discussion does not permit officials to frame their arguments in terms of these things and, instead, demands that the national interest be invoked to justify all foreign policy. Officials will, in this model, always face a tension between the reality of political pressures, on the one hand, and the need to suppress mention of such pressures, on the other. Anderson does not explicitly discuss secrecy, though it is easy to see that secrecy would play a role in the process he describes: Officials will seek to conceal information that suggests political rather than national interest motivations for the positions that they advocate. Thus, the target of secrecy is neither the public nor foreign enemies, but competing government officials or agencies.

Overall, the Bureaucratic explanation suggests that much of the material that is inaccessible to researchers are classified only because of arcane feuds or operating procedures. And the degree of secrecy will be circumscribed. All states (except, perhaps, those on the verge of disintegration) must have a significant degree of interagency coordination, and the need for such coordination will constrain the practice of bureaucratic rivalry, as well as the associated secrecy. The irrationalities of standard operating procedures result in a more or less random classification of documents;⁽²⁾ in the process, some analytically important materials may, unfortunately be withheld, but there is little systematic withholding of historically relevant materials. The methodological implications of the bureaucratic perspective are therefore encouraging for the researcher.

4. The Internal Threat Approach, Variant I

This explanation of secrecy argues that government officials use secrecy (at least partly) as a device to mislead the public and to ensure elite control over foreign policy. This perspective is often implied in much of the realist literature on international relations. To be sure, there is no detailed, deductively derived realist theory of government secrecy and, in fact, no writer in the realist tradition has ever even addressed the issue of government secrecy in any explicit way; however, a realist perspective on secrecy is, nevertheless, implied in much of the existing literature. For example, Hans Morgenthau (1967, pp. 142--43) notes that a conflict between the requirements of good foreign policy and the preferences of public opinion is in the nature of things and continues:

. . .the government must realize that it is not the slave to public opinion; that public opinion is not a static thing to be discovered... it is a dynamic, ever changing entity to be continuously created and recreated by informed and responsible leadership; that it is the historic mission of the government to assert that leadership.

Morgenthau suggests the novel weapon of propaganda, which can be used to manipulate public opinion. Secrecy is not directly mentioned here but it is certainly implied; the use of secrecy and the selective dissemination of one-sided information will help the statesman to assert informed and responsible leadership, which realists like Morgenthau advocate.

Selective information dissemination, advocated by Morgenthau, may sometimes involve deliberate misrepresentation.⁽³⁾ In other instances, misrepresentations will be eschewed and only partial or distorted -- rather than outright false -- information will be disseminated, while any unfavorable information will be suppressed. The easiest way to suppress information is to mark it secret, to classify it.⁽⁴⁾

What, according to the realist, motivates officials to act secretly? The answer to this question must be tentative, given the lack of explicit discussion on this topic. However, the realist literature implies that officials seek to advance the national interest and security (see Krasner, 1978);⁽⁵⁾ it is this commitment to the national interest that motivates them to mislead the public and to conceal information. Some writers note that secrecy and deception can produce positive results. Thomas Bailey, for example, states with approval that Franklin Delano Roosevelt systematically acted to deceive the US public about his efforts to lead the country into World War II (cited in Novick, 1988, p. 308). Friedrich (1972, p. 179), more explicitly, states that secrecy is eminently functional in many government operations, in both the domestic sphere but especially in matters of foreign and military policy.⁽⁶⁾ In short, realists insist on the need for elite control of foreign policy, even in a democratic form of government; secrecy is sometimes a necessary means for achieving such elite control.

Critics of realism, such as Noam Chomsky (1989, pp. 16--17), challenge these views as elitist and undemocratic. Woodrow Wilson was also critical of government secrecy (at least, this was his view before he entered government service). In Congressional Government (quoted in Friedrich, 1972, p. 181), he asserted that there is not any legitimate privacy [i.e. secrecy] about matters of government. Government must... be absolutely public in everything that effects it -- and Wilson made no exception for foreign policy. Realists, however, have long contended that such idealistic views, however well intended, are ill-suited to the rigors of international conflict.⁽⁷⁾ In short, the realists view secrecy and public deception as vital tools for an effective foreign policy.

The Internal Threat Approach, Variant II

Among the critics of the realist approach, some propose rudimentary explanations for secrecy. David Sadofsky (1990, pp. 84--85) implies that the national interest is often a subterfuge, a shallow rationalization; government secrecy according to Sadofsky, functions not to protect national security but, rather, to protect the interest of government policy-makers. Sadofsky thus suggests a distinct and apparently competing interpretation of secrecy.

It is interesting to note that Sadofsky's hunch -- that governmental self interest motivates secrecy -- is quite consistent with the extensive literature on rational action in government (see Bartlett, 1973; Buchanan, 1975; Downs, 1967; and Niskanen, 1975). There is a wide consensus among major studies of the subject, that government officials are motivated at least partly (though not, for most analysts, exclusively) by the desire to advance their own interests and their respective agencies. From the standpoint of rational action, the significance of secrecy can be easily established: Rational officials will classify information if such classification advances the officials' interests. However, bureaucrats will not necessarily pursue their interests in a consciously cynical fashion; it is well established that individuals and groups tend to allow their conceptions of the public good to be influenced by their own interests. Karl Mannheim (1936) has convincingly shown that human beings or even whole social groups rationalize behavior and assume that their own interests are really universal interests, while George Orwell (1954) has argued that language can obfuscate this process of rationalization. The present discussion may appropriately be considered in ideological terms, and bureaucrats may be expected to rationalize self interested behavior in all areas, including the use of secrecy and deception.

According to the rational action perspective, secrecy is intertwined with considerations of self interest. Rational policymakers will selectively release information that reflects favorably upon themselves and their bureaux, and will withhold damaging information. If an official acts incompetently or illegally, or if the official acts in ways that are inconsistent with accepted standards of conduct, such information will probably be withheld. As Anthony Downs (1967, p. 77) notes: . . .all types of officials tend to exaggerate data that reflect favorably on themselves and to minimize those that reveal their own shortcomings.

It seems reasonable to assume that self interested bureaucratic behavior must at least sometimes conflict with the interests of the public (and this would hold true for any definition of public interest). Government officials have a major advantage in such conflicts: they control much or nearly all the relevant information (Niskanen, 1971, pp. 29--30, 39--40). Government officials can pursue questionable actions that would be difficult to justify in public and can then classify the relevant documentation. Officials can plausibly assert that classification advances the public interest, and since members of the public cannot see the classified documents they cannot contradict the officials' assertions. The classification system is, by its very nature, well suited to official self interest.

6. The Internal Threat Approach: A Synthesis

The rational action perspective on secrecy is, in many obvious ways, contrary to the realist approach. The two perspectives have very different views of the motivations of policy-makers. The realists assume that policy-makers generally act to further the national interest, while the rational action perspective assumes, in contrast, that policy-makers often act to further their own interests. The difference between the rational action and realist approaches to secrecy may be illustrated with a recently declassified document, which makes reference to radiation experiments conducted by the US government during the early period of the Cold War (see US Atomic Energy Commission, 1947). The document in question is a memorandum written in 1947 and declassified in 1994, and it directed government employees as follows: It is desired that no document be released which refers to experiments with humans and might have adverse effects on public opinion or result in legal suits. Documents covering such work field should be classified "secret."' The document also authorized the release of any information pertaining to radiation research that could be beneficial to human disorders and diseases (emphasis added).

The rational action approach would view classification in this case as an effort by government officials to protect themselves and their agencies against public embarrassment, budget cuts, or even criminal prosecutions. The realist approach, in contrast, would interpret these events somewhat differently. Realists would argue that such classification was intended to serve the national interest, since the radiation experiments being described were vital to national security and -- in the context of the Cold War -- it was sometimes necessary to undertake actions that conflicted with ordinary conceptions of morality. As David Charters notes (1985, p. 334): The dark underside of international affairs is inherently untidy and unpleasant. It leaves little room for comfortable moral, political, or operational positions.

Despite all of these differences, the realist and rational action perspectives have a crucial feature in common: They both agree that government officials withhold information in order to deceive the public. The two perspectives differ with regard to why secrecy takes place and, concomitantly, they differ in the way they analyze the motivations of policy-makers; but they agree that the public is the target of secrecy policies.⁽⁸⁾

For the remainder of this article I will combine Variants I and II and term this hybrid simply the Internal Threat approach. The defining feature of the Internal Threat approach is that the general public is considered the principal target of government secrecy. The practical implications of this perspective are disturbing for the researcher. First, the perspective implies that the range of classified information would be broad; even information that is known to enemy powers may remain classified -- since it is the public, not enemy powers, that is the principal target according to this explanation. Secondly, we would expect classification periods to be of long duration. Documents pertaining to events that had long ceased to be active foreign policy concerns may remain classified, since exposures can discredit government activities in general. Finally, the Internal Threat perspective assumes that governments will often withhold information that may be of great interest to researchers. By its very nature, the Internal Threat perspective assumes that officials make special efforts to conceal information on controversial actions -- such as the radiation tests cited above -- and such controversial material is often historically important. Thus governments do not randomly suppress information; on the contrary, they systematically suppress some of the most historically relevant information. If the Internal Threat approach is correct, then much of our research is and must be conducted without important or even vital information.

Comparison of Theories

At this point, several qualifications are needed. It should be apparent that there is significant overlap among these various theories and that the above three categories are not mutually exclusive. Many realists would no doubt subscribe to aspects of both the External Threat and the Internal Threat explanations, while advocates of rational action might concur simultaneously with the Internal Threat approach and the Bureaucratic Politics approach. However, this discussion will treat the three perspectives separately, in order to simplify our analysis. This article will consider each theory as a mutually exclusive ideal type, which has been deliberately exaggerated in order to facilitate effective comparison.⁽⁹⁾

Overall, these three theories present a range of different analyses about the causes and the effects of secrecy. For convenience, the main points of each perspective are presented in outline form (see Table 1). The most significant differences among the theories may be summarized as follows: The Internal Threat approach argues that the general public is the principal target of government secrecy. This approach, if substantiated, would pose serious methodological questions for researchers, since it implies that the scope of secrecy would be broad and that governments have a systematic tendency to classify information that is relevant to researchers. The External Threat and Bureaucratic Politics approaches, in contrast, suggest that the scope of secrecy would be considerably more circumscribed and that much of the secret information is somewhat less relevant to researchers. Now that we have outlined the three perspectives, I turn to the case study of the Congo Crisis, where the theories will be applied.

8. The Congo Case

The Congo Crisis of 1960--1 is well suited for this analysis of secrecy for two reasons. First, it was one of the major conflicts of the Cold War. In the USA, it received regular attention from two presidents (Eisenhower and Kennedy) and, if one looks at the index from the New York Times, it is apparent that the Congo received more attention during the period 1960--1 than any other Third World crisis area (including even Vietnam). Second, there is now substantial information pertaining to covert operations during the Congo Crisis by the USA and other countries that has been culled by academic researchers and US Senate investigators (see especially Gibbs, 1991; Kalb, 1982; Mahoney, 1983; US Senate, 1975; and Weissman, 1974. These studies are based, in part, on a large volume of US and UN documents that have been declassified over time. Although many documents pertaining to the Congo remain classified, there does exist a significant quantity of declassified materials. With these new materials it is possible to analyze the type of information that was classified and the type of information that was publicly disseminated. By examining this data, we can evaluate, at least tentatively, the three approaches to government secrecy described above.

The crisis began when the Congo (now called Zaire) gained independence from Belgium in June 1960. The Belgians had made few efforts to prepare the country for self-government and, at the time of independence, there were no more than thirty Congolese university graduates. The new country almost immediately was beset by random violence and civil war. The province of Katanga, a major producer of copper, cobalt, and other basic minerals, seceded from the Congo within days and formed a separate state. Later, the diamond-producing region of South Kasai also seceded. The Congolese National Army was in complete disarray throughout this period, and the central government proved powerless to contain the disorder. The Congo was also the object of intervention by several foreign countries. The USA, the USSR, Belgium, France, and the United Kingdom all intervened extensively during the confusion of the Congo Crisis. A UN peace-keeping force -- one of the largest in history --patrolled the country and played a large role in its politics of the Congo during the period 1960-4.

A major issue throughout the early period of the Congo Crisis was the role of Patrice Lumumba, the head of the largest political party and also the country's first prime minister. Western officials were generally quite hostile toward Lumumba, and they regarded him as excessively nationalistic and unreliable in the East--west conflict. Lumumba's decision, in August 1960, to accept Soviet military aid, was especially contentious. Early scholarly accounts of Lumumba were highly negative, presenting him as authoritarian, emotionally unstable, and procommunist (see, for example, Lefever, 1965, pp. 38--42). More recent analyses, however, have been considerably more favorable and have noted that Lumumba was the only democratically elected leader in the history of the Congo, and that many of his actions seem much more reasonable in light of what is now known about the extent and nature of foreign intervention in the crisis (see especially Minter, 1984, pp. 28-35).

Whatever the specifics, US officials detested Lumumba and were determined to overthrow him, and this became the principal objective of US policy during the first six months of the Congo Crisis. The efforts to remove Lumumba triggered what became one of the largest and most important operations in the history of the Central Intelligence Agency. Many years later, in 1984, former CIA Director William Colby commented: Of the covert operations undertaken [since the 1950s]... I would say some have been very successful and some have been disasters... The Bay of Pigs was certainly a disaster. But consider our program in the Congo (Colby, 1984, p. 36).

The covert operation plans were outlined by CIA Director Allen Dulles during August 1960, and they emphasized building opposition against Premier Lumumba among Congolese parliamentarians, with the objective of bringing about a no-confidence vote against the prime minister. President Eisenhower reportedly reviewed the plans and expressed extremely strong feelings on the necessity for very straightforward action in this situation, and he wondered whether the plans as outlined were sufficient to accomplish this.⁽¹⁰⁾ The president's statement appears to have encouraged the CIA to expand the scope of its operational plans. Dulles then cabled the CIA station in the Congo and authorized actions aimed at replacing Lumumba; also, Dulles vaguely suggested even more aggressive action if it can remain covert (US Senate, 1975, p. 16).

Although the USA considered sending American troops to the Congo, this option was rejected by President Eisenhower as excessively risky. Instead, the Eisenhower administration sought to project US power through proxy forces recruited by the CIA. The CIA made a special effort to gain influence in the Congolese National Army, through a young and politically-connected officer, Joseph Desire Mobutu.⁽¹¹⁾ Recent studies of the Congo Crisis generally agree that CIA funding supported Mobutu's activities. Madeleine Kalb (1982, p. 96) concluded that It was money provided by the CIA and the other western embassies that kept him [Colonel Mobutu] in business during the Congo Crisis, while Stephen Weissman (1974, p. 95) cites evidence that Mobutu was paid by the CIA. Even former CIA Director Colby (1984, p. 36) now agrees that the Agency helped Joseph Mobutu during the Congo Crisis.

Amply supplied by the USA (and possibly other Western powers), Mobutu distributed large amounts of money to the officers and men under his command; through this arrangement he was able to establish bonds of loyalty among his soldiers. Mobutu's military unit probably numbered only a few hundred soldiers, but his troops were paid exceptionally well (by Congolese standards), and this unit was virtually the only really functioning element of the Congolese National Army. Also, Mobutu and his men were strategically located in Leopoldville, the capital city, and were thus well placed to influence events. This force was a major conduit of influence for American foreign policy. Another source of US government influence was the UN peace-keeping force. The USA provided a disproportionate share of the funding for the UN force (about 40% of total funding for the duration of the operation) and several of the UN officials in charge of the Congo operation, notably Secretary-General Dag Hammarskjold and Special Representative Andrew Cordier, secretly collaborated with US officials (see Gibbs, 1993; and Collins, 1992). The USA thus had formidable capabilities to inject its power into the Congo.

The first (documented) effort to overthrow Lumumba came in September 1960 during a constitutional dispute. Technically, the prime minister of the Congo was appointed by the president who, at the time was Joseph Kasavubu. On September 5, 1960, President Kasavubu announced that he was dismissing Lumumba and appointing a new prime minister. There is significant evidence that both US and UN officials influenced Kasavubu's action. The British ambassador to the Congo later wrote that '"at one time" Kasavubu sent for Cordier to "ask how a coup d'etat should be organized"' (quoted from Weissman, 1974, p. 90, n.84). Both Cordier and Hammarskjold coordinated their activities with the State Department. On the day of Lumumba's dismissal, Cordier sent an

'urgent recommendation' to the US government via Hammarskjold suggesting that it send an official to [the capital city] Leopoldville -- 'not too high-level, not too junior -- who could observe the local scene, see what the Russians were up to, and make a firsthand report to Washington (Kalb, 1982, pp. 74-75).

Cordier also consulted with the US ambassador to the Congo and the next day, September 6, he arranged for UN troops to close the airport -- to preclude any airlift of loyal troops to the capital by Lumumba.⁽¹²⁾ Cordier then ordered UN forces to close the radio station as well, which prevented Lumumba from broadcasting an appeal for support. In short the UN, in cooperation with the USA, encouraged Kasavubu to act against Lumumba and then attempted to bloc Lumumba's return to power. While the UN support for President Kasavubu was publicly known at the time -- Hammarskjold openly defended the president's action -- the collaboration with US officials was completely secret.

The close association between the USA and the UN deteriorated to some extent after this incident. Cordier was recalled as the director of the Congo operation, while Hammarskjold appears to have become concerned that the UN was leaning too far in the direction of the USA and that such close identification with a single power was weakening the credibility of the U.N. For the next several months, the policies of the USA and the UN diverged to some extent (for details, see Gibbs, 1991, pp. 96--98). Moreover, the effort to dismiss Lumumba was unsuccessful. Lumumba argued his case before the Congolese legislature and received an overwhelming vote of confidence, while Kasavubu's dismissal of the prime minister was countermanded.

The CIA, apparently out-maneuvered, remained determined to remove Lumumba. The USA now advocated a military coup d'etat by Colonel Mobutu and his men. On 14 September, Mobutu deposed Lumumba, closed the parliament, and established himself as dictator. He organized a council of Congolese university students to advise him, called the College of Commissioners, and established a rudimentary de facto government. There is clear evidence that the CIA played a role in this coup. Weissman (1974, p. 95) writes:

The main point -- that the CIA was heavily involved in the. . . emergence of Col. Joseph Mobutu and his College of Commissioners -- has been confirmed by a former official of the American Embassy who observed the events in question. Further substantiation was provided by two former government officials who spoke directly to the CIA station Chief in [the Congolese capital] Leopoldville, Lawrence Devlin.

Former CIA officer Victor Marchetti (1988) also confirms that the Agency was involved in Mobutu's coup.

Let us pause here and focus on the significance of these events for the topic of this article, i.e. information policy. During these events, the US government released no information pertaining to its intervention in the Congo; at the same time, information was selectively released which suggested that the USA was not intervening at all and was, on the contrary, seeking to protect the Congo from potential intervention by other powers. The basic position was publicly stated by the US representative to the United Nations, James Wadsworth (1960, p. 530): . . .the Security Council should make it unmistakably clear once more that no one -- and I repeat no one -- should be permitted to intervene unilaterally in [the Congo]. The information on US intervention in the Crisis, such as the subversion of the Lumumba government or the support for Colonel Mobutu's coup, was suppressed.

In any case, US officials continued to fear Lumumba, even after the coup removed him from power. Lumumba sought refuge in his residence in the capital, but it was widely believed that he would attempt a comeback. US officials urged Mobutu to arrest Lumumba; however, UN forces, now showing some independence from US foreign policy, guarded Lumumba's house and prevented the arrest. A stalemate reigned for several months, until Lumumba left his home in November and fled north where, en route, he was arrested by Mobutu's troops. Lumumba was then imprisoned at a military installation not far from the capital city.

Lumumba was subsequently removed from prison and, on 17 January, 1961, he was assassinated. A full discussion of the assassination would go beyond the scope of this article, but we will focus on the possible role of the United States. Throughout the events described above, the CIA was considering a variety of plans to kill Lumumba, especially with poisons, and these assassination plots have been exhaustively documented in an investigation by the US Senate (1975, ch. 3). The assassination plots were authorized by top level officials, including the CIA Director, Allen Dulles, the Agency Deputy Director for Operations, Richard Bissell, and the Africa Division Chief, Bronson Tweedy.

The obvious question is: Did the CIA assassinate Lumumba? It should be noted that the Senate investigation only acknowledges that the CIA had plotted to kill Lumumba, but emphasizes that these plots were all unsuccessful. Crucially, the report concludes: there is no evidence that the United States was in any way involved in Lumumba's death at the hands of his Congolese enemies (US Senate, 1975, p. 256).⁽¹³⁾ Other investigations, however, draw different conclusions. Richard Mahoney (1983, p. 71) states, there can be little doubt that the CIA -- though not the actual assassin -- was a moving force behind the murder, while Madeleine Kalb (1982, p. 189) notes little doubt that US officials encouraged Lumumba's Congolese opponents to eliminate him. According to John Stockwell (1978, p. 237), a former CIA officer with long experience in Africa, Lumumba was beaten to death by henchmen of Congolese politicians who had close relationships with the CIA. Whatever the CIA's role in the eventual killing, there is general agreement among all sources that the Agency certainly plotted to kill Lumumba.

9. Analysis

It should be apparent that a significant amount of information pertaining to US intervention in the Congo was suppressed; at the same time the US government disseminated information that misleadingly suggested that the USA was not intervening. Now, let us consider the significance of these facts for the three theories of government secrecy. First, the External Threat approach suggests that the USA withheld information in order to deceive competing foreign powers, especially the Soviet Union, about US activity in the Congo. Such secrecy was necessary and, without it, the Soviet Union might have sabotaged US operations. It is easy to think of certain types of information that might have been withheld for reasons that seem consistent with the External Threat approach. Certainly, the US could not have released the operational details of its covert action in the Congo, without jeopardizing the outcome.

However, there emerges a major problem with this approach: Much of the information on US covert action was already known to the Soviet Union, at least in its principal features. The collaboration between the USA and the UN command, for example, was openly discussed in Soviet press coverage (see Izvestia, 1960). US support for Mobutu also appears to have been common knowledge. An article in Pravda (1960a) noted that 'American imperialists supply Mobutu and his bandits with enormous sums of money for the struggle against Lumumba,' while another article (Pravda, 1960b) characterized Mobutu as a 'puppet colonel' and a creation of the NATO powers. These are fairly accurate assessments of the situation, at least insofar as Mobutu was indeed dependent on foreign support. (Other countries were aware of these facts as well: In 1961, a Yugoslav diplomat told his US counterpart, We know you have influence with Mobutu [Kalb, 1982, p. 278].) Even the US efforts to assassinate Lumumba were not altogether secret. At one point, Pravda (quoted in Kalb, 1982, p. 189) reported that the Americans and the Belgians were "preparing to do away with" Lumumba. If the Soviet press was aware of these facts then, surely, they were well known throughout much of the Soviet foreign policy bureaucracy; the US failed to release information pertaining to events that the USSR already knew about. The External Threat approach may explain some details of US information policy but as a general explanation it is seriously inadequate.

The Bureaucratic Politics perspective helps to explain other features of the information process in the Congo case. It is certainly true that much of the secret information must have been classified due to standard operating procedures or for idiosyncratic reasons. The present author can attest that the presidential libraries contain (among other things) significant amounts of information on the Congo Crisis that would be of no interest to anyone. Some of the classified information seems so trivial that it is difficult (for the layperson) to understand why officials bothered to classify it.

On further consideration, however, Bureaucratic Politics fails to provide an adequate account of information policy. First, random classification, due to standard operating procedures, fails to account for much of the secrecy in this case. On the contrary, there seems to have been a systematic effort to suppress virtually all information pertaining to US intervention in the internal politics of the Congo -- none of this information was released until long after the end of the Congo Crisis. (Indeed much of the information discussed above was only released under pressure from Congress during special hearings, in 1975, on covert operations.) And there is little evidence that interbureaucratic differences influenced information policy to any great extent. Although the CIA did withhold some information from the State Department,⁽¹⁴⁾ the two agencies seemed to be in general agreement with regard to intervention in the Congo.⁽¹⁵⁾ State Department officials were usually well informed with regard to CIA activities in the Congo and, according to one account, there was a policy consensus in Washington about the need for the removal of Lumumba (US Senate, 1975, p. 16).⁽¹⁶⁾

The Internal Threat approach -- which emphasizes government efforts to hide information from the public -- is strongly supported by the Congo case. According to this approach, governments seek to conceal potentially controversial activities or ones that could generate public opposition. In the Congo case, secrecy successfully concealed government activities (such as the efforts to assassinate Lumumba) that were potentially very controversial. Other covert operations, such as the coaching of President Kasavubu, the efforts to undermine Lumumba's position in the military, the support for Mobutu's coup would all have been very difficult to justify in public. Moreover, the Internal Threat approach correctly predicts that governments often obscure information that is historically relevant. Decisive events of the Congo Crisis -- e.g. the dismissal of Lumumba or the coup by Mobutu -- were influenced by foreign intervention, yet this intervention was withheld from public knowledge.

10. Conclusion

I hope you recognize how obvious is the inference that it is just the shocking nature of these passages which has led to their suppression. -- Sigmund Freud (1975, p. 146), The Dream Censorship

The most significant finding in the Congo case was that government information policy systematically distorted the public record and created the misleading impression that the USA was not, in fact, intervening. Indeed, early scholarly accounts of the Congo Crisis (e.g., Young, 1965; also Lefever, 1965) relied overwhelmingly on public information and, as a result, they missed the significance of US intervention. It is also interesting to note that the Congo case is inconsistent with some basic assumptions about international relations. Hans Morgenthau (1967, p. 226), for example, wrote of political assassination as follows:

According to its official records, the Republic of Venice, from 1415 to 1525, planned or attempted about two hundred assassinations for purposes of its foreign policy... The documents record virtually no offer of assassination to have been rejected by the Venetian government... In the same period, the cardinals brought their own butlers and wine to papal coronation dinners for fear they might otherwise be poisoned; this custom is reported to have been general in Rome, without the host's taking offense at it. . . . Obviously, such methods to attain political ends are no longer practiced today.

Our discussion of the Congo suggests that Morgenthau was mistaken; contemporary international relations does indeed entail assassination plots. Morgenthau (writing in 1967) found no evidence of assassination in the public records and, mistakenly, assumed that it no longer occurred. In short, government secrecy successfully misled many researchers.

Although further research is needed on this topic, our (tentative) findings in this case raise significant methodological issues. Much of the literature in international relations is based on public -- and possibly unreliable -- sources of information. Studies on the causes of war (Maoz & Abolali, 1989; and Weede, 1984), for example, focus almost exclusively on publicly available information. Such studies systematically miss the significance of covert war (see critique by Forsythe, 1992), and how such wars can influence theories of war in general. Recent discussions of such diverse issues as the Gulf War, the Bosnia--Hercegovina conflict, and international democratization usually fail to mention that much of the information on these subjects remains unavailable. Classified documents pertaining to these events will probably remain closed to researchers for several decades (if, indeed, they are released at all). Such methodological complications, which are rarely addressed in the literature, require augmented consideration.

Our findings on the Congo case (if confirmed by further studies) would also have implications for democratic theory. Recent analyses often assume that democratic political systems -- with regular elections and freedom of expression -- ensure public control of the government (see Diamond, et al., 1988). Our discussion raises serious questions about this assumption. The public could not have effectively controlled foreign policy in the Congo case, since the specifics of that policy were not known to the public. Governmental information policy can be seen as a means of social control. It is not a perfect one -- occasionally documents will be leaked -- but it can be effective in maintaining bureaucratic power and manipulating public opinion in ways that seem incompatible with existing theories of democratic government. Our findings in the Congo case imply that discussion of international issues is sometimes based on perceptions that are artificially constructed by government information policy (on the constructed nature of political discussion in general, see Edelman, 1988; and Schneider & Ingram, 1993). Theorists of democracy might consider the significance of information policy and how it can affect democratic control of governments, especially in the area of foreign policy.

Before closing, I consider one final point: It may be objected that this discussion is anachronistic and that, after the Vietnam War and the Watergate scandal, secrecy was effectively curtailed in the USA. Indeed, there is a widespread view that the Freedom of Information Act and other reforms created a far more open foreign policy in the USA and, as a result, the public scrutinizes policies in great detail. Thus, Robert Keohane (1984, pp. 94--95) writes American officials often lament that the US government leaks information "like a sieve. . . ,"' while Alden and Schurmann (1990, p. 66) claim that US foreign policy is formulated in a floodlit society, where secrecy is impractical.

Such views are mistaken; in fact, US foreign policy became far more secretive during the 1980s, as legislative and administrative actions considerably narrowed the range of information supplied by the federal government.⁽¹⁷⁾ Document sets declassified by the government during this period were considered unreliable. For example, documents published by the State Department in the series Foreign Relations of the United States (FRUS), generated controversy in the history profession, with Roger Dingman noting unmistakable evidence of dramatic and devastating changes in the editorial policies and processes which govern the publication of documents on American foreign policy (paraphrased in Washington Post, 1990). A diplomatic historian, Bruce R. Kuniholm, reviewed a FRUS volume on US policy toward Iran and commented that certain sections constitute a gross misrepresentation of the historical record, sufficient to deserve the label of fraud (quoted in Washington Post, 1990), while Warren I. Cohen (1990) stated that the Iran volume was a fraud, a gross distortion of American activity there. Criticism in the history profession culminated in a resolution by the Organization of American Historians, which criticized an appalling increase in the amount of incomplete and deleted documentary material (quoted in Washington Post, 1990). Other professional organizations also noted increased restrictiveness. The American Library Association (1988) issued a pamphlet entitled Less Access to Less Information by and about the US Government. Military doctrine increasingly emphasized the need for propaganda and public deception, while specific military actions in Grenada, Panama, and the Persian Gulf entailed exceptionally high levels of secrecy (Sharkey, 1991).

With the demise of the Cold War and, in 1993, the inauguration of a new presidential administration, there is a general expectation that government information policies will become somewhat less restrictive in the US. However, only modest changes in information policy have been undertaken by the Clinton administration, and there is every indication that government secrecy will continue long after the end of the Cold War (see Secrecy and Government Bulletin 1993 and 1994; and Twentieth Century Fund, 1992, p. 15). The notion that US foreign policy is conducted under floodlit conditions is inaccurate.

Classification of documents contributes to an atmosphere of deception in politics, and such deception will be present, to some extent, in all political systems. It is also worth noting that US information policy is relatively unrestrictive -- at least when compared with other countries such as the (now former) USSR or even Western European democracies (see for example, Gleditsch & Hogetveit, 1984). Our conclusions here would apply with even greater force to other countries, where secrecy is more widely practiced. Information policy thus has significant implications for the practice and the study of democratic politics. It is a subject that deserves increased attention from researchers.

1. Some useful descriptions of recent information policy can be found in Bennet (1988), Feinberg (1989), and Guida (1989).

2. However, these standard operating procedures could be interpreted in different ways. It may be objected that intelligence agencies do systematically withhold information, since entire categories of documents are routinely classified. However, there is no systematic tendency to withhold information that is particularly important (to any audience); whether the classified information is important or mundane is the result of random processes.

3. With regard to misrepresentation, see analysis by Sissela Bok (1978). It should be noted that the existence of misrepresentation has been disputed by Arthur S. Hulnick (1992, p. 92) who writes: "The CIA does not lie to the press or to Congress."

4. Of course, classification is not the only way to withhold information. Other methods include destroying information or simply not mentioning sensitive issues in writing. Telecommunications obviously contribute to the latter form of suppressing information. A former CIA officer notes, in his memoirs, that the CIA does not record all of its activities with regard to highly sensitive operations, such as assassination plots (Stockwell 1979, p. 160n).

5. It should be noted that Rosenau (1968) provocatively argues that the term "national interest" is so vague that it is not very useful as a social science concept. However, for purposes of argument this essay will accept the contention of Krasner (1978), who argues that the national interest remains a valid concept.

6. However, Friedrich (1977, p. xiv) acknowledges that, in many cases, secrecy can also be dysfunctional and can exceed legitimate boundaries.

7. Also, it may be argued that the public does not even want to know government secrets, since many of these might be disturbing. I thank Gordon Tullock for suggesting this possibility.

8. It is interesting to note that a recent report on covert action by the Twentieth Century Fund (1992, p. 14) acknowledges that the public can be a target of secrecy: "Sometimes, covert action has been hidden only from Americans, not from our adversaries." The report does not elaborate on this observation and implies that such occurrences are unusual. Overall, the report endorses continued covert action, even after the end of the Cold War.

9. Max Weber (1949) notes that ideal types are, by definition, one-sided and exaggerated models. He argues that such exaggeration is a valid and even vital component of social scientific inquiry, provided that the exaggeration is recognized by the researcher.

10. This was reported by National Security advisor Gordon Gray, who paraphrased the president's views (US Senate, 1975, p. 15).

11. Note that Mobutu (now called Mobutu Sese Seko) has been the president of Zaire since his coup d'etat in 1965. At the time of this writing, he is still in power.

12. Stephen Weissman (1974, p. 91) writes that "According to a member of the American Embassy, Cordier consulted Ambassador Timberlake before taking this step [closing the airport]." Note that Gendebien (1967, p. 76) provides information on possible direct CIA involvement in this incident.

13. John Ranelagh (1987, p. 344) defends the Senate report conclusions. Though he acknowledges that the CIA undoubtedly plotted an assassination, Ranelagh adds:

Lumumba was not assassinated through the efforts of the CIA or by the CIA's killers... He broke away from UN custody on November 27, determined to lead his supporters against Mobutu and the UN force. He was captured by Mobutu's troops three days later. On January 17, 1961, he was flown to Elisabethville, the capital of the secessionist Katanga province, and he was murdered (probably by Mobutu supporters) there within hours of his arrival.

14. At one point CIA headquarters sent a cable to the field, noting that the Agency was considering sending arms and supplies to anti-Lumumba factions; the cable specifically stated the State Department representatives should not be informed of these plans (US Senate, 1975, p. 18, n.1).

15. CIA Director Dulles sent a cable to the field authorizing action "to replace Lumumba with a pro-Western group" and noted that this policy had been '"seen and approved at competent level," by the State Department' (US Senate, 1975, p. 16). Weissman (1974, p. 89) also presents evidence that the US ambassador was involved in the covert operations.

16. The quote is from CIA officer Bronson Tweedy. It should be noted that the Senate investigation found some evidence that President Eisenhower ordered or at least knew about the assassination efforts, but this evidence is conflicting and inconclusive (US Senate, 1975, chap. 3).

17. It should be noted that the CIA was never especially forthcoming, even during the high point of government openness during the 1970s. Madeleine Kalb (1982, p. xv) writes of her research on the Congo Crisis: "The CIA... was highly obstructive. Appeals for additional information... produced interminable delays -- and in the end all the cables were withheld. The few documents that were released contained no information that could not have been found in The New York Times; and there were an astonishing number of completely blank pages." It should be noted that these remarks pertained to research Kalb conducted during the mid 1970s. Similarly Peter Wyden wrote of his experiences researching the Bay of Pigs incident: "The CIA bureaucracy... stonewalled and confirmed suspicion about itself by refusing to admit the time off the clock on the wall." This research was also conducted during the 1970s -- when government openness was at a historic high point (Wyden, 1979, p. 331).
Also note that the CIA has long maintained unofficial files for especially sensitive operations, that do not officially exist and are thus permanently exempted from disclosure. According to former CIA officer John Stockwell (1978, p. 228n), "Since the Freedom of Information Act, the agency increasingly uses a system of 'soft,' 'unofficial,' or 'convenience' files for sensitive subjects... Such files are not registered in the agency's officials records system and hence can never be disclosed under FOIA [Freedom of Information Act]."

Alden, Edward & Franz Schurmann. 1990. Why We Need Ideologies in US Foreign Policy. Berkeley: Institute of International Studies.

Allison, Graham. 1971. Essence of Decision: Explaining the Cuban Missile Crisis. Boston, MA: Little, Brown.

Alperovitz, Gar & Kai Bird. 1994. Understanding the Cold War, Foreign Policy, no. 94.

American Library Association. 1988. Less Access to Less Information by and about the US Government: A 1981--1987 Chronology. Washington, DC.

Anderson, Paul A. 1981. Justifications and Precedents as Constraints in Foreign Policy Decision Making, American Journal of Political Science, vol. 25, no. 4.

Bartlett, Randall. 1973. Economic Foundations of Political Power. New York: Free Press.

Bendor, Jonathan & H. Hammond. 1992. Rethinking Allison's Models, American Political Science Review, vol. 86, no. 2.

Bennet, James R. 1988. Censorship by the Reagan Administration, Index on Censorship, August.

Bok, Sissela. 1978. Lying: Moral Choice in Public and Private Life. New York: Vintage.

Buchanan, James M. 1975. The Limits of Liberty. Chicago, IL: University of Chicago Press.

Charters, David. 1985. The Role of Intelligence Services in the Direction of Covert Paramilitary Operations, in Alfred C. Maurer, Marion D. Tunstall, & James M. Keagle, eds., Intelligence Policy and Process. Boulder, CO., Westview.

Chomsky, Noam. 1989. Necessary Illusions. Boston, MA: South End Press.

Cohen, Warren. 1990. At the State Dept., Historygate, New York Times, 8 May.

Colby, William et al. 1984. Should the US Fight Secret Wars? Harper's, September.

Collins, Carole. 1992. Fatally Flawed Diplomacy: Cordier and the Congo Crisis of 1960, Africa Today, vol. 39, no. 3.

Diamond, Larry, Juan J. Linz, & Seymour Martin Lipset. 1988. Preface in Larry Diamond, Juan J. Linz, Seymour Martin Lipset. eds., Democracy in Developing Countries: Africa. Boulder, CO: Lynne Rienner Publishers.

Downs, Anthony. 1967. Inside Bureaucracy. Boston, MA: Little, Brown.

Edelman, Murray. 1988. Constructing the Political Spectacle. Chicago, IL: University of Chicago Press.

Feinberg, Lotte E. 1989. Managing the Freedom of Information Act and Federal Information Policy: The Reagan Years, Government Information Quarterly, vol. 6, no. 4.

Forsythe, David P. 1992. Democracy, Covert Action, and War, Journal of Peace Research, vol. 29, no. 4.

Franck, Thomas M. & Edward Weisband, eds. 1974. Secrecy and Foreign Policy. New York: Oxford University Press, 1974.

Freud, Sigmund. 1975. A General Introduction to Psychoanalysis. New York: Pocket Books.

Friedrich, Carl J. 1972. The Pathology of Politics. New York: Harper and Row.

Friedrich, Carl J. 1977. Introduction, in Itzhak Galnoor, ed., Government Secrecy in Democracies. New York: New York University Press.

Galnoor, Itzhak, ed. 1977. Government Secrecy in Democracies. New York: New York University Press.

Gendebien, P. H. 1967. L'Intervention des Nations Unies au Congo, 1960--1964. Paris: Mouton.

Gibbs, David N. 1991. The Political Economy of Third World Intervention: Mines, Money, and US Policy in the Congo Crisis. Chicago, IL: University of Chicago Press.

Gibbs, David N. 1993. Comment fut scelle le destin d'un pays aux richesses convoitees, Le Monde Diplomatique, September.

Gleditsch, Nils Petter & Einar Hogetveit. 1984. 'Freedom of Information in National Security Affairs: A Comparative Study of Norway and the United States,' Journal of Peace Research, vol. 21, no. 1.

Guida, Richard A. 1989. The Costs of Free Information, The Public Interest, no. 97.

Hulnick, Arthur S. 1992. Learning about US Intelligence: Difficult but Not Impossible, International Journal of Intelligence and Counterintelligence, vol. 5, no. 1.

Izvestia. 1960. Colonialists Gamble on a Quisling, reprinted in Current Digest of the Soviet Press, October 5, 1960.

Jeffrys-Jones, Rodrick. 1989. The CIA and American Democracy. New Haven, CT: Yale University Press.

Joseph, Paul. 1981. Cracks in the Empire: State Politics in the Vietnam War. Boston, MA: South End Press.

Kaiser, Frederick M. 1989. The Amount of Classified Information: Causes, Consequences, and Correctives of a Growing Concern, Government Information Quarterly, vol. 6, no. 3.

Kalb, Madeleine. 1982. The Congo Cables: The Cold War in Africa from Eisenhower to Kennedy. New York: MacMillan.

Keohane, Robert. 1984. After Hegemony. Princeton: Princeton University Press.

Krasner, Stephen. 1978. Defending the National Interest. Princeton: Princeton University Press.

Lacqueur, Walter. 1985. A World of Secrets. New York: Basic Books.

Lefever, Ernest W. 1965. Crisis in the Congo. Washington, DC: Brookings.

Lowi, Theodore. 1977. The Information Revolution, Politics, and the Prospects for an Open Society, in Itzhak Galnoor, ed., Government Secrecy in Democracies. New York: New York University Press.

Mahoney, Richard. 1983. JFK: Ordeal in Africa. New York: Oxford University Press.

Mannheim, Karl. 1936. Ideology and Utopia. New York: Harcourt, Brace, and World.

Maoz, A., & M. Abolali. 1989. Regime Type and International Conflict, 1816-1976, Journal of Conflict Resolution, vol. 33, no. 1.

Marchetti, Victor. 1988. Telephone interview, 18 June.

Maurer, Alfred C., Marion D. Tunstall, & James M. Keagle. 1985. General Introduction: The Conceptual Framework in Alfred C. Maurer, Marion D. Tunstall, & James M. Keagle, eds., Intelligence Policy and Process. Boulder, CO: Westview.

Minter, William. 1984. The Limits of Liberal Africa Policy: Lessons from the Congo Crisis, Transafrica Forum, vol. 2, no. 3.

Morgenthau, Hans. 1967. Politics among Nations. New York: Knopf.

Niskanen, William A., Jr. 1971. Bureaucracy and Representative Government. Chicago: Aldine and Atherton.

Novick, Peter. 1988. That Noble Dream: The Objectivity Question and the American Historical Profession. Cambridge, MA: Cambridge University Press.

Orwell, George. 1954. Politics and the English Language in George Orwell: A Collection of Essays. Garden City, NY: Doubleday--Anchor Books.

Pravda. 1960a. Lumumba's Arrest was Sanctioned by the State Department, reprinted in Current Digest of the Soviet Press, 11 January, 1961.

Pravda. 1960b. Another Plot, Another Failure, reprinted in Current Digest of the Soviet Press, 23 October, 1960.

Ranelagh, John. 1987. The Agency: The Rise and Decline of the CIA. New York: Simon & Schuster.

Relyea, Harold C. 1987. National Security and Information, Government Information Quarterly, vol. 4, no. 1.

Rosenau, James. 1968. National Interest, International Encyclopedia of the Social Sciences.

Sadofsky, David. 1990. Knowledge as Power: Political and Legal Control of Information. New York: Praeger.

Schneider, Ann & Helen Ingram. 1993. Social Construction of Target Populations: Implications for Politics and Policy, American Political Science Review, vol. 87, no. 2.

Secrecy and Government Bulletin. 1993. New Draft Order on Secrecy: An Ambiguous Beginning, October.

Secrecy and Government Bulletin. 1994. Draft Executive Order, January/February.

Sharkey, Jacqueline. 1991. Under Fire: US Military Restrictions on the Media from Grenada to the Gulf War. Washington, DC: Center for Public Integrity.

Stockwell, John. 1978. In Search of Enemies. New York: Norton.

Twentieth Century Fund. 1992. The Need to Know: The Report of the Twentieth Century Fund Task Force on Covert Action and American Democracy. New York.

US Atomic Energy Commission. 1947. Memorandum from Colonel O. G. Haywood, Jr., April 17. (Note that this document was declassified in 1994 and provided to the author by the Federation of American Scientists.)

US Senate. 1975. Alleged Assassination Plots Involving Foreign Leaders. Washington, DC: US Government Printing Office.

Wadsworth, James A. 1960. Statement of September 14, Department of State Bulletin, October 3.

Washington Post. 1990. Historians Say Secrecy Distorts Foreign Policy Chronicle, April 16.

Weber, Max. 1949. Objectivity in Social Science and Social Policy, in The Methodology of the Social Sciences. New York: The Free Press.

Weede, Erich. 1984. Democracy and War Involvement, Journal of Conflict Resolution, vol. 28, no. 4.

Weissman, Stephen. 1974. American Foreign Policy toward the Congo, 1960--1964. Ithaca, NY: Cornell University Press.

Wyden, Peter. 1979. Bay of Pigs: The Untold Story. New York: Simon & Schuster.

Young, Crawford. 1965. Politics in the Congo. Princeton: Princeton University Press.

___________________________
DAVID N. GIBBS, b. 1958, is Assistant Professor of Political Science at the University of Arizona and the author of The Political Economy of Third World Intervention: Mines, Money, and US Policy in the Congo Crisis (University of Chicago Press, 1991). He is currently working on a revisionist account of UN peace-keeping operations.

*The author thanks the following persons for their comments: Thomas Ferguson, David Wilkins, Michael Sullivan, Diana Rix, Gordon Tullock, Michael Schaller, Thomas Chrisiano, Randolph Siverson, Edward Muller, Allen Whiting, William Dixon, Tim McKeown, Cary Nederman, and Kathleen Schwartzman.

FAS | Government Secrecy | e-Prints || Search | Join FAS