INTELLIGENCE AND DEMOCRATIZATION: THE CHALLENGE OF CONTROL IN NEW DEMOCRACIES
Occasional Paper #5
Thomas C. Bruneau
CCMR
March 2000
Draft - Not for citation or attribution without the author's permission
Introduction
The main political challenge facing leaders of new democracies that have merged worldwide during the last fifteen years is the consolidation of the democratic process. Important components of democratic consolidation include economic reform and a change in the relationship between military and political leaders that results in democratic civilian control of the armed forces. One of the most contentious issues in reform on the armed forces is control of the intelligence apparatus. This is due not only to the legacies of the prior, non-democratic, regimes in which the intelligence or security apparatus was a key element of control and linked to human rights abuses, but also to the inherent tension that exists between the intelligence community and a democratic government. Admiral Stansfield Turner, a former Director of Central Intelligence, succinctly defines this tension: "Secret agencies within democratic governments are anachronisms, because popular controls break down when citizens cannot know everything their government is doing."(1)
This paper will discuss the process of bringing intelligence agencies under democratic control, and the challenge this process poses within the context of democratic consolidation. Control of intelligence organizations is reflected in both the constitutional and legal arrangements regarding the internal and external roles of the armed forces, and the structures and processes involved in the function of intelligence agencies. However, discussion of control and intelligence are difficult. First, the terms and concepts associated with intelligence and undefined and ambiguous, open to multiple interpretations. Second, the intelligence community retains an institutional interest in maintaining the secrecy and power it is afforded. As a result, it minimizes the knowledge that outsiders have of intelligence organizations and activities. Third, there is little written about intelligence and democratization. Theoretical literature on intelligence and democracy exists, but it mainly pertains to the established democracies such as Great Britain, France, and the United States where the goal is to reiterate the need to control the intelligence apparatus lest it undermine democracy.(2) Consequently, this report has a modest purpose: To help demythologize intelligence in new democracies by providing an introduction to some of the key issues involved in intelligence organization structure and function. This report can serve as a stepping stool for those in new democracies that want to learn about civilian control of intelligence affairs.
The Counterintelligence State
In most authoritarian regimes (including the former Soviet bloc), the intelligence apparatus was a key element for maintaining power. These regimes were based on something other than democratic legitimacy exercised via free elections. They had to rely on such organizations to identify domestic opponents, neutralize their opposition to the government, and generate at least apathy, if not outright support, of government activities using such measures such as media control. In most cases these organizations were intelligence services. Intelligence agencies, given such a crucial role in the maintenance of government control, became increasingly useful, and consequently grew in size and power. As a result, intelligence organizations became largely autonomous, even within authoritarian regimes where intelligence was primarily counterintelligence and focused on protecting the state's secrets from outsiders - which meant anyone outside the central core of power. (3) Oftentimes anything could be defined as a state secret, so the scope of that which had to be controlled was immense. While in many cases the intelligence services rhetorically linked internal opposition to putative foreign enemies, the overwhelming focus of the intelligence services, in most countries was domestic opposition, and not other states.(4)
Among the many negative legacies of the intelligence services in the new democracies was their involvement in human rights abuses. The information they gathered on their own people was at times obtained with coercive methods, and used in arbitrary and violent efforts to eliminate domestic opposition. They are, in short, integrally associated with the human rights abuses that characterize most authoritarian regimes. In addition to the overall popular legacy, there is little awareness of intelligence functions and organizations. Most civilian politicians, let alone the public at large, do not know enough about intelligence to be able to have an informed opinion about it. In some countries there is real concern that the intelligence apparatus has accumulated, and is still collecting, information that could be used against average civilians and politicians. Not only is there a lack of information about intelligence communities, but fear, associated with past intelligence activities, exacerbates the challenge of actively seeking out this information.
The Challenge of Democratic Consolidation
Despite efforts by students of comparative politics to develop models of democratic transitions, these transitions are largely sui generis and defy generalization. Studies have shown that the authoritarian regimes collapsed due to their successes as well as their failures, or the actions (or inaction) of domestic elites or foreigners. In any case, power finally passed on to more or less popularly elected civilians.(5) Transitions allowed new, democratic, regimes to emerge, but they did not necessarily result in stable democratic regimes. Today, democratic consolidation is one of the primary focuses in the field of Comparative Politics. Consolidation is a useful concept because it suggests that a new regime's structures and processes are becoming stable. That is, a democratic regime is consolidated when the elites and the masses accept it as "the only game in town."(6) This acceptance is no easy task, especially if one considers the basic requirements for a regime to be termed democratic. A standard definition of contemporary democracy is as follows:
Modern political democracy is a system of governance in which rulers are held accountable for their actions in the public realm by citizens, acting indirectly through the competition and cooperation of their elected representatives.(7)
For accountability to function, several procedural conditions are necessary. They include the often-noted seven fundamental guarantees that ensure free and fair elections, freedom of speech and association, the right to run for and hold office, among others, and constitute a corpus of guarantees dependent upon a supportive culture. As more countries began to consolidate their new democracies, scholars identified an additional defining characteristic-the requirement that no un-elected body has authority over popularly elected officials.
A political situation in which these guarantees function is obviously quite different from the structures and culture of the prior authoritarian regime. Major challenges resulted from both the lack of recent experience with democracy and the difficulty of attaching values to these new structures and processes by a population unfamiliar with them. In addition, in most cases these countries also face economic problems, often accompanied by social disruption. In general, democracy is a very demanding political system for both elites and average citizens. Both need to be involved for it to function well, yet new democracies are very tentative. The development of trust and transparency in the context of the legacies of the authoritarian regime is a major challenge. It is quite possible that the intelligence apparatus is not under government control, but instead has power over the civilian officials, as seems to be the case in Russia today.(8) If the elected government does not control intelligence, it is, by definition, not a consolidated democracy.
The Meaning of Intelligence
Due to the scope and diversity of intelligence, there is disagreement on its meaning.(9) Intelligence is mainly defined by process. That is, the process of gathering and using information for some purpose. Since processes are varied, as are the sources of information and their ends, much is, by necessity, left vague. Most discussions within the intelligence community center on tradecraft; the "how to" of sources, methods, and analysis rather than the "what is"? Further definitions of the intelligence community, either by design or habit, are characterized by vagueness and ambiguity. This attitude, or approach, is probably intentional, and used to minimize information flows. Once one becomes aware of intelligence, and its limits, there is an even greater awareness that not everything is knowable, let alone known. Also, intelligence officers are trained to collect information, and not to disseminate it, except to very few of their superiors with a need to know. This tendency pervades the whole field of intelligence. They are professionals in intelligence; information is their vocation. It makes no sense to give it away, unless indeed it is disinformation.
Given our purposes here, in the context of the new democracies, we must use a broad definition of intelligence in order to convey the scope of what it can include, which is extremely broad.(10) Glenn P. Hastedt in Controlling Intelligence states succinctly: "The four elements of intelligence are clandestine collection, analysis and estimates, covert action, and counter-intelligence."(11) Loch Johnson elaborates this synthesis:
Intelligence commonly encompasses two broad meanings. First, the secret agencies acquire and interpret information about threats and opportunities that confront the nation, in an imperfect attempt to reduce the gaps and ambiguities that plague open sources of knowledge about the world. A nation especially seeks secret information to help it prevail in times of war, with as few casualties as possible. Second, based on information derived from denied and open sources, policymakers call upon their intelligence agencies to shield the nation against harm (counterintelligence) while advancing its interests through the secret manipulation of foreign events and personalities (covert action). Intelligence thus involves both information and response.(12)
For our purpose here, intelligence is understood as these four functions: collection, analysis, counterintelligence, and covert action. Intelligence also refers to the organization(s) that collect the information, as well as the information collected. As all individuals and organizations collect and process information, this information in itself is not the defining characteristic. The key characteristics include the premise that these functions are centered in and intended for the state, and that they are secret. This knowledge thus has a dual nature; it is information, but it is secret information, used by the state in potential or real conflicts. What follows is a very brief review of these four functions.(13)
Collection
Intelligence organizations collect information. Major issues in collection address the types of information collected and the means used to do so. At a minimum, intelligence organizations use "open sources" - public information that includes periodicals and materials found on "the web," and at seminars and conferences. There is an ongoing debate regarding the usefulness of open versus classified because a great deal of information on a wide variety of topics is readily available.(14) Other types of collection methods include human intelligence, or HUMINT, and scientific and technical intelligence, such as SIGINT (from intercepts in communications, radar, and telemetry), IMINT (including both overhead and ground imagery), and MASINT (which is technically derived intelligence data other than imagery and SIGINT). HUMINT is information collected directly by people and includes information provided by ambassadors or defense attaches as part of their normal reporting routines, information obtained at public and social events, and information obtained clandestinely through spies, reading others' mail, and documents. HUMINT is the traditional "espionage," or spying, mainly through the use of agents in another country who provide secret information to their managers who forward it to their home agencies.
Wealthier countries have large investments and capabilities in scientific and technical intelligence. In the United States, the bulk of the approximately $27 billion annual intelligence budget goes to these technical forms of collection.(15) They include the interception and processing of communications by phone, radio, and computers. The processing may consist of decoding as well as translation. Another source of scientific and technical intelligence is photo or image reconnaissance. Originally, it was simply an attaché taking picture of ships, planes, facilities, and the like. It evolved with aerial reconnaissance to include highflying airplanes that were undetectable and theoretically unreachable by a potential enemy. More recently, scientific and technical intelligence consists of satellite photoreconnaissance or imagery. This technology is becoming increasingly available. Today, anyone with the money can purchase from private firms photos that were unavailable or highly classified a few years ago.(16)
Analysis
Raw intelligence is not very useful until it is analyzed. Analysis, or the anticipation of analysis, also shapes collection requirements. Analysis, or what to conclude from raw information, has always been the big challenge in intelligence. In retrospect, the United States should have known about Japanese intentions at Pearl Harbor, the Argentines should have known about American and British reactions to the invasion of the Malvinas, and Saddam Hussein should have known that the United States would react forcefully to the invasion of Kuwait. The processing of gigantic quantities of data as well as the choice of policy recommendations to derive from information makes analysis quite a complex process.
The intelligence professional must convince policy makers of the accuracy and relevance of the intelligence. Production is only the first step; the intelligence must then be marketed. Analysis, in short, is not a simple technical issue but rather includes methods, perceptions, and political preferences. Much of the analytical literature on intelligence in the US and USSR focuses precisely on whether, and to what extent, leaders use the information provided to them by intelligence organizations.(17)
Counterintelligence
The most basic purpose of counterintelligence is to protect the state, and its secrets, against other states or organizations. This seemingly clear and straightforward function can in fact become, in the words of the long-time and controversial head of counterintelligence at the CIA, James Angleton "'the wilderness of mirrors,' where defectors are false, lies are truth, truth lies, and the reflections leave you dazzled and confused."(18) Abram N. Shulsky defines the scope of issues involved:
In its most general terms, counterintelligence refers to information collected and analyzed, and activities undertaken, to protect a nation (including its own intelligence-related activities) against the actions of hostile intelligence services. Under this definition, the scope of counterintelligence is as broad as the scope of intelligence itself, since all manners of hostile intelligence activities must be defended against.(19)
Memoir accounts by retired intelligence professionals, as well as books by students of intelligence, indicate that counterintelligence has the greatest negative implications for democracy, due to its reliance on the surveillance of the citizenry.(20) The implications for democracy are much more severe in new democracies where counterintelligence was the principal function of intelligence services. The intelligence service sought to root out real and imagined enemies of the state, often resulting in yet more opposition leading to a spiral of violence. If even in established democracies a certain amount of paranoia is inherent in counterintelligence - "there is an enemy at work here and we must root him out," in less institutionalized and non-democratic Third World countries this attitude routinely resulted in extreme violation of human rights and impunity for the intelligence agents.(21)
Covert Actions
Covert actions, or as the British term them "special political actions" and the Soviets "active measures," are actions intended to influence another state by means that are not identified with the state behind the actions. While covert action is not always included in government documents as part of intelligence, it is the topic most often featured in the books on controversial actions, including intelligence fiascoes. The Watergate cover-up, activities related to the coup in Chile and death of Allende in 1973, and various assassination attempts of nations' leaders led to the creation of the Church and Pike committees in the U.S. Congress in the mid-1970s, and resulted in the assertion of greater control of the CIA by the legislative branch.
There are three main types of covert action. The first is propaganda, which can include the utilization of the media in another country to convey a certain message. The second is political action, such as the funding or other support to government leaders, political parties, unions, religious groups, the armed forces, and the like to follow a certain course of action in another country. The third type of covert action is paramilitary activity, which involves the use of force. It can consist of smaller actions, such as assassination or the arming and training a small contingent of dissident tribal groups, or it can be large such as the Bay of Pigs invasion. The problem is that any large action cannot remain covert for long, and even with smaller actions it does not take much imagination to determine the country behind the action. While there is a considerable literature on U.S. and Soviet covert actions, and little on other countries, this does not mean that other countries do not also engage in covert actions. Indeed, countries seek to use their foreign intelligence service, including military attaches, to not only gather information but also to influence policies in another country. Mr. Richard Bissell, who through experience, knew much about American covert action, has elaborated a straightforward rationale for covert action:
It becomes overwhelmingly obvious that we are deeply concerned with the internal affairs of other nations and that, insofar as we make any effort to encourage the evolution of the world community in accord with our values, we will be endeavoring purposefully to influence these affairs. The argument then turns out to be not about whether to influence the internal affairs of others, but about how.... Open diplomacy, however, has its limitations as a policy tool. There are times when a great power can best attain its objectives by acting in total secrecy. On certain occasions, however, a great power may seek to influence the internal affairs of another nation without its knowledge or without the knowledge of the international community. These circumstances require covert action.(22)
This justification is not limited to a great power. Obviously, not every country has robust capabilities in all four intelligence functions, but the fact that they exist, that any nation has these capabilities, means that this is the global framework within which intelligence must be understood. Intelligence is created to defend the state. It must defend it within the context of potential enemies, and use available instruments. All countries have some degree of awareness of the intelligence capabilities of other countries, and of the fact that they will be involved in, or even the target of, collection and covert action.
Intelligence and Democracy
All countries have an intelligence apparatus of some scope and capability. The question for new democracies is: what kinds of intelligence organizations do they need and how can they be controlled? While the challenge is especially severe in the new democracies, democratic control of intelligence is a challenge everywhere for at least four reasons. First, as Pat Holt states "Secrecy is the enemy of democracy."(23) Why? The answer is secrecy encourages abuse. If there is secrecy how can there be accountability, the operative mechanism of democracy? Since intelligence organizations are secret they themselves can avoid the checks and balances on which democracy is based. Second, intelligence agencies are not only secret, but they also collect and analyze information, and information means power. Intelligence organizations take on agendas and purposes of their own. Secrecy limits public scrutiny. Peter Gill uses the analogy of the "Gore-Tex" state to illustrate the degree of domestic penetration by the security intelligence services. Information flows in one direction and not two directions; to the intelligence services and not from them to state and society.(24) Intelligence may be autonomous from state control and, through the use of information that others do not have, determine policy. There are two further perceptual, or behavioral, elements beyond secrecy and the unique control of information that hinder democratic control of intelligence organizations. Third, intelligence agents and organizations routinely break laws abroad. Indeed, in most cases they do not admit to who they are or for whom they work. Furthermore, spying is illegal everywhere. Intelligence managers provide undeclared funds to foreign nationals as agents so that they may author articles to tap phones, steal documents, and the like, all of which are illegal activities. There may be a problem in making the distinction between breaking laws abroad and not breaking them at home. Fourth is the self-justification that intelligence is critical to defense of the nation. In the words of Peter Wright, "It [intelligence] is a constant war, and you face a constantly shifting target."(25) It is up to the intelligence organizations to root out spies, domestic and foreign, who are threats to the nation. They may easily perceive that they, more than anyone else, really know what is going on; how dangerous the threat really is. The intelligence officers' task is to identify threats to the nation, and there are always threats; the only question is, how serious are the threats. They know things, and others do not, and this may lead to a certain condescending attitude regarding others who are not in the know, who are not initiated into the club.
In light of the general difficulty of controlling intelligence, and, considering the background in most new democracies, what are the choices to be made and what are the implications of different options for democratic control? Initially, and this is really a requirement that is the same regarding the armed forces in general, democracies must establish a clear and comprehensive legal framework. Intelligence is "slippery," and if the legal framework is not clear and explicit, intelligence agencies can never be brought under control. The legal framework must emerge from the democratic structures and processes, and must seek to ensure the continuation of the democratic values that they seek to promote in the area of intelligence. In South Africa, for example, soon after the transition to majority rule in 1994, the government initiated through the legislative process, reform of the intelligence apparatus. This involved three major bills in parliament that clearly defined and restructured the intelligence system. (26) In Brazil it took a considerably longer time from the transition to civilian government in 1985 for reform to occur; indeed, it was only in 1999 that the ABIN (Brazilian Intelligence Agency) was created to replace the authoritarian regime's National Intelligence Service (SNI). The Brazilian Congress played a central role in the creation of ABIN, and the legislation does provide a clear legal basis for civilian control of intelligence. (27)
There are three general decisions to be made regarding intelligence, which should be stipulated, clearly and explicitly within the legal framework. The level of implementation of the four intelligence functions, and the amount of resources devoted to this implementation, needs to be determined. This decision rests on an assessment of the global and regional situation, alliances, recent history, and available resources. However, this is also a political choice. How much is intelligence worth? Obviously it is worth a great deal if it provides the nation with the means to maintain its independence in the face of a hostile neighbor. Intelligence also can be valuable in lieu of larger forces. It can allow a country to focus its forces on the most serious threats thereby minimizing redundancy and higher operational costs. An assessment of what intelligence is really worth requires a political decision. Does the mere fact of having a certain level of intelligence capability avoid hostile intentions and actions? This decision also depends on its relationship with other, more powerful, countries that may share intelligence capabilities with it. Neither of these decisions can be made in a vacuum, and they should be integrated into an overall framework for decision - making in defense. The main point is, however, that there must be an analysis of what the nation requires and how much it is willing to pay for it. This is, of course, a general issue in civil - military relations.
Evidence consistently suggests that the top level of the executive branch of government should control these decisions and their coordination. In the United States the National Security Council holds this responsibility. In Brazil, after the most recent reforms, it is the Secretariat of Institutional Security directly under the president. And, in South Africa after the reforms in the mid - 1990s, it is the National Intelligence Coordinating Committee that reports directly to the president and the cabinet.
The second choice concerns the balance in intelligence between civilian and military organizations, both in terms of production (collection and analysis) and consumption. In most countries, intelligence has been a military monopoly in production and consumption. During democratic consolidation there are decisions to be made as to whether military intelligence should be replaced in whole or in part by new civilian organizations. Should the military have responsibilities only in military intelligence and civilians assume responsibility in strategic intelligence and counterintelligence? Equally important as collection is consumption. To whom is the intelligence product distributed? Only the president of the country, his director for intelligence, members of the cabinet such as Interior, or only the military, the congress, and who else? Obviously access to the information, and the form in which it is made available, has great implications for the potential power of those who receive it. In the 1996 Guatemala peace agreement, the "Accord on Strengthening of Civil Power and the Function of the Army in a Democratic Society" stipulates that there will be created "a civil department of intelligence and analysis of information'. By April 2000 they had defined the new structure, which included one military and two civilian intelligence organizations. Unfortunately, implementation was delayed due to the failure of a referendum on constitutional revisions later in the year.
A sub-theme of this balance between civilian and military institutions is the issue of internal and external intelligence. Does the same organization have responsibility for domestic intelligence as well as foreign intelligence? The former is of course mainly counterintelligence. Are these functions fused? If so, what are the controls so that it is not used for personal political purposes? In most democracies the functions are separate. In the United States, the Federal Bureau of Investigation handles counterintelligence within the United States, and the Central Intelligence Agency has performed both functions outside the country. In most European democracies the functions are divided between counterintelligence and foreign intelligence, the organizations doing their tasks wherever necessary, at home or abroad. This has not been much of an issue in most of the new democracies since they focused mainly internally. It should be noted that domestic intelligence is cheap in comparison to external intelligence, and most countries cannot afford to do the latter professionally.
The third choice concerns the relationship between intelligence and policy. This also logically involves the issue of coordination among the intelligence organizations. Is all intelligence formally coordinated by a director of central intelligence as in the United States, but separate from policy (the DCI is not in the cabinet)? Or, is intelligence separate as with MI 5 and MI 6 in Great Britain, but as both are located within the Foreign Office, linked with policy? The main issue here concerns an ongoing debate about the implications for objective intelligence analysis when it is closely linked to policy as opposed to the supposed loss of efficiency by having intelligence that is not linked. There are great variations in how different democracies handle this issue.(28) The answer depends on the political traditions and structures of the country, but the underlying issue of policy-relevant but not policy-driven intelligence is what must be assessed. One of the critiques in the United States of covert action is that these actions fuse both intelligence and policy within the CIA. Rather than providing intelligence objectively, the agency develops the policy, conducts it, and largely evaluates its success. Hastedt, who has published one of the few books on controlling intelligence, makes his position explicit on this issue: "The purpose of intelligence is to inform and warn policymakers. The choice of what to do lies with the policymaker. If intelligence is brought too closely to policy making it runs the risk of being corrupted."(29) In the new democracies it is too early to determine how they are dealing with this issue since they are still in the process of defining and implementing new structures and processes in the (often newly - created) ministries of defense and intelligence organizations. Argentina is probably the furthest developed in the implementing of new structures in both areas, but even here the structures are still changing substantially and much remains personality - driven. (30)
All three of these decisions hold implications for democratic control over intelligence. The first choice, about intelligence functions, has obvious implications, especially regarding whether counterintelligence should be included or excluded. The second, civilian vs. military location of the intelligence function has implications in terms of civilian control over the armed forces, and then civilian control over intelligence. Third, a very close link with policy can make intelligence less a function of information gathering and analysis, and more a tool used by political leaders to retain power. In the cases with which I am most familiar - Argentina, Brazil, Guatemala, and South Africa - the discussions suggest that they are dealing well with decisions one and two, but that three remains unresolved.
Explicit Mechanisms of Control over Intelligence
A common mechanism to control intelligence is through its separation into different agencies. Policymakers should prevent any single agency from having a monopoly on intelligence. This is the model in the United States. A possible arrangement could separate intelligence organizations into branches for each of the armed forces services and the police, as well as separate organizations for focus on domestic and foreign intelligence. This proliferation of organizations may or may not be efficient, due to bureaucratic squabbling, but it eliminates the chances of monopoly by any single organization or individual, and creates opportunities for more democratic control. Most countries that are seeking to reform their intelligence structures have moved in this direction. Argentina, Brazil, Guatemala, and South Africa, as well as other countries in Europe and Central America, have created civilian intelligence organizations to complement (compete with?) the military intelligence organizations. In Brazil and Guatemala, for example, there are two civilian organizations.
A second mechanism for democratic control is an oversight mechanism or mechanisms. Does anyone have oversight over intelligence or does the apparatus, and it alone, have responsibility for monitoring its own performance? The latter option is extremely dangerous. In the United States, oversight has expanded to the current situation where not only do the intelligence agencies have inspector generals, but the executive has oversight bodies and the two houses of congress also have oversight committees.(31) In Great Britain, oversight remains very limited but the democratic institutions are hallowed. It seems necessary today in countries that are seeking to consolidate their democracies that if intelligence is to be under democratic civilian control then there must be oversight. How far it extends, and under what terms it operates, will vary tremendously. Oversight has immediate implications for control but also has implications for popular support for intelligence.
Since knowledge equals power, it is important to specify who has access to the intelligence and in what form. Is it limited only to the military or do civilians in the executive also have access? What about the legislature? Do any or all of them have access even before operations such as covert actions? This issue concerns not just immediate distribution of intelligence (which here extends to covert actions as well) but the general availability of information after a certain period of time. The possibility of wider distribution also holds implications for control. If the agencies know that in the future the files will be open for public scrutiny, they must be careful of their behavior.
There is a dilemma inherent in the issue of control, and that is the trade-off between democratic control over intelligence and the effectiveness of the intelligence apparatus in defending the nation. This dilemma can be reduced to the tension between accountability, which requires transparency, and the intelligence function, which requires secrecy. For example, does legislative oversight result in agents being uncovered? Democracies wrestle with this dilemma constantly and there is no easy or sure solution. Rather, it requires constant attention and adjustment. In discussing legislative oversight in other countries the issue of the reliability or sense of responsibility of legislators always emerges. It is very difficult to make appropriate judgments on this issue, but it should be noted that since legislative oversight was imposed in the U.S. in the 1970s there have been far fewer cases of members of Congress or their staffs releasing classified information than leaks from the Executive branch. The issue is not just oversight or not, but how it is implemented and by whom.
The possibility exists that democratically elected civilians may not in fact be interested in controlling the intelligence apparatus in new democracies. In virtually all of these countries, the use of elections to determine access to power is a new and relatively fragile means of determining who wields power. Even in old and stable democracies leaders often prefer "plausible deniability," rather than access to the information required to control a potentially controversial or dangerous organization or operation.(32) Logically, this would be even more the case in newer democracies. First, the politicians may be afraid of antagonizing the intelligence apparatus through efforts to control it because the intelligence organization might have embarrassing information concerning them. Second, they may be afraid because the intelligence organization in the past engaged in arbitrary and violent actions, and the politicians are not sure that these practices have ended. Third, there are probably no votes to be won in attempting to control an organization that most people either don't know about or want to ignore.
Frequently the issue of democratic control of intelligence can be profitably discussed only in those countries that have already grappled with the more general issues of civilian control of the military, and which have begun to institutionalize the structures and processes for this control. In the others, the environment remains too opaque or tense for open discussion of intelligence organizations and oversight. Intelligence is far from the first issue the new civilian leadership wants to confront.
Towards Democratic Control of Intelligence
Countries that want to begin to exert democratic civilian control over the intelligence apparatus must undertake several tasks. These tasks are similar to those necessary for asserting civilian control over the military in general, but are more acute due to secrecy and the penetration of state and society in line with the counterintelligence function.(33) The tasks that follow are not prioritized, and in fact should be pursued simultaneously. They concern civilian competence, public interest and then pressure, and the profession of intelligence.
The first task is to motivate civilians to learn about intelligence so they can control it. In most authoritarian regimes intelligence was monopolized by the military, and civilians had no role whatsoever. These countries will be unable to control intelligence unless they prepare civilians to learn enough to understand intelligence requirements, and be aware of intelligence functions. This information, in turn, will help civilians to achieve some degree of cooperation, if not respect, from the intelligence professionals. Although this is a difficult task, one has to start somewhere. One should begin with the formal and public commitment by the government to reviewing intelligence to establish a new policy. This has been done in such countries as Argentina, Brazil, Guatemala, and South Africa. Intelligence reform took place in these countries mainly due to the bargain resulting from the democratic transitions. The commitment must also open the possibility for civilian positions in intelligence. Otherwise, as in civil-military relations in general, no civilians will come forward if they do not anticipate viable careers. Then, civilians can begin to learn about intelligence by reading the unclassified literature from several countries, and taking advantage of cooperative training arrangements in intelligence with other nations. For example, CCMR offered a week - long seminar specifically on this topic in Argentina in September 1998, and are in correspondence with civilian intelligence specialists in the four countries noted above. CCMR also offers a semester long course on the topic of "Intelligence and Democracy" in a masters' degree program at the Naval Postgraduate School. In addition, the country needs to establish regional programs for civilians to share their insights and further develop their common fund of knowledge.
The second broader task is to encourage a political culture, which supports the legitimate role of intelligence in a democracy but does not allow it to run rampant. James A. Schlesinger made this point: "to preserve secrecy, especially in a democracy, security must be part of an accepted pattern of behavior outside of government and inside."(34) The responsibility must be held by democratically elected civilians to control intelligence, but to not release classified information for personal or political reasons. As in the general case of democratic civil-military relations, this culture can be encouraged by generating a public debate. The challenge is to break through the current apathy or fear by the population of intelligence by initiating the debate. In some older democracies, such as Canada, France, Great Britain, and the United States there is a fairly regular debate stimulated by non-governmental organizations (NGOs) and the media, which is periodically dramatized by intelligence fiascoes that become public. The role of the media is crucial, and their awareness of intelligence can be encouraged in the same manner as the public. Again, the debate can be stimulated by the politicians' commitment to establish a policy on intelligence. Such a debate has been initiated in a few of the newer democracies.
The Peace Accords in Guatemala between the government and the guerrillas signed in December 1996 stipulate in several sections that intelligence will be transformed and put under civilian oversight. These commitments have led to public seminars on intelligence, publications by NGOs, and articles in the newspapers.(35) In Argentina there also is a debate initiated by a small number of civilians realizing that democratic consolidation requires civilian control over intelligence.(36) And, more recently, in Brazil, the government's commitment to revise the intelligence system has generated a public debate. This debate serves a number of important functions. It demythologizes intelligence thereby allowing outsiders to more realistically assess the necessity of intelligence and its value for a country; it creates legitimate space for civilians who want to become intelligence specialists; and it puts pressure on the government to be more transparent. It should be noted that international NGOs are very willing to assist other countries in generating this debate. (37)
The third task is not about civilians or the public in general, but concerns the selection, training, and overall preparation of intelligence professionals; those who specialize as intelligence agents working for the state. The focus on intelligence as a profession is particularly apt since these professionals, more than any other single profession, are controlled even in a democracy more by professional norms than by outside controls (such as oversight).(38) In contrast, in addition to their self-policing, or ethic, doctors are regulated by the legal system and licensing boards, lawyers by the legal system and bar associations, politicians by the legal system and elections, and the armed forces by budgets, promotions, and a myriad of civilian control mechanisms. The intelligence professionals, however, are controlled only in the last analysis, if that, by the external structures and processes noted above. They are granted impunity to break laws abroad and have tremendous leeway within their own country and organization. As illustrated in virtually all the books and articles dealing with intelligence agents, secrecy allows them to operate with a tremendous amount of autonomy. There are few checks because they operate secretly, they are ensconced in a bureaucracy with other like-minded agents and develop a closed-club mentality, and they are very suspicious of outsiders, including at times their superiors.
Intelligence as a Profession
A profession can be defined in terms of the three criteria of expertise, corporateness, and responsibility.(39) In the case of the intelligence professional the criteria are as follows:
First, their expertise is defined in line with the four intelligence functions of collection, analysis, counterintelligence, and covert action described above. The range of what intelligence professionals do is extremely diverse. What unifies them, or defines them as intelligence professionals, is secrecy. Unlike other professions, but for certain limited aspects of patient or client privacy or privilege, the intelligence professional is defined by secrecy. (The military profession also has elements of secrecy but mainly these pertain to intelligence.) In reference to covert actions one of the foremost American intelligence professionals, Richard M. Bissell Jr. states:
The professional competence of a clandestine service consists of, and is measured by, its ability to carry out operations secretly (or deniably), much as lawyers' competence consists in their ability to win cases, and doctors' in their ability to prevent or treat illness. The clandestine service may number among its members brilliant journalists, able warriors, and superior political analysts, but the professional skill for which, presumably, they are hired is the ability to organize and conduct operations covertly. This is a rather specialized skill not widely found outside of intelligence and internal security services.(40)
And, in reference to counterintelligence one of the foremost British intelligence professionals states:
The profession of intelligence is a solitary one. There is camaraderie, of course, but in the end you are alone with your secrets. You live and work at a feverish pitch of excitement, dependent always on the help of your colleagues. But you always move on, whether to a new branch or department, or to a new operation. And when you move on, you inherit new secrets, which subtly divorce you from those you have worked with before. Contacts, especially with the outside world, are casual, since the largest part of yourself cannot be shared.(41)
Their expertise is thus diverse, as is intelligence itself, and the defining characteristic of the profession is secrecy.
Second, their corporateness is defined by their access to secret systems, documents, information, sources, and operations. As doctors enter the profession through boards, internships, and residencies, professors by comprehensive examinations and Ph.D. dissertations, and lawyers by the bar exams, intelligence professionals enter via security clearances. Clearances are the control mechanism for entry into and continuing in the profession. There are few educational requirements in common for intelligence professionals, even between different intelligence organizations within one country, and there is little else that defines their corporate identity besides their access to classified information.(42) In intelligence everything is compartmentalized; different levels of clearances plus the need to know determine access. Even agents with similarly high clearances do not, and are not supposed to, discuss information unless they have the need to know in terms of their current projects and responsibilities. The security clearances and the shared work in secret and on secret information and projects, create identification as a member of a unique club. It may also breed a certain arrogance, or a sense of impunity since they are overseeing privileged information.
Third, the mission of the intelligence professional is to serve in defense of the state. But if we consider the first two criteria of expertise in secret matters and access via security clearances, we are led inexorably to a profession, which largely governs itself according to its own definition of responsibility. In new democracies this is doubly serious, as the state was not accountable to the general population and the intelligence agents may not have even been responsible to the small group controlling the state. Who can know and who is to control? The sense of responsibility is incredibly important, and even in stable democracies enough incidents come to light to cause concern that the agents are not serving the state. Or, better, they are serving it in their limited organizational terms and not those of the democratically elected leaders. This sense is captured in a quote from James Angleton while testifying before Congress on why the CIA had not destroyed stocks of a toxic poison: "It is inconceivable that a secret intelligence arm of the government has to comply with all the overt orders of the government."(43) It is difficult to accommodate this kind of attitude with the procedures and culture of democracy.
To Change a Profession
This review of the defining characteristics of the intelligence profession suggests that major efforts must be made in the new democracies to promote and inculcate a sense of professional responsibility by making the agents and agencies responsible to the state via the democratically elected leaders. How to do this? Only by committing great attention and resources to recruitment, training, and obligating that the professionals remain involved in the larger polity and society. The specifics of this prescription have to be defined separately for each nation. One of the biggest difficulties is that the government will most easily recruit retired military into civilian intelligence positions. They may have taken off the uniform, but their attitudes remain the same as those of everyone around them. If new personnel cannot be found, then can their ethic of responsibility be changed? In most countries, including the older democracies, there is little explicit attention to promoting this ethic.(44) In the older democracies the larger society supports responsibility to the democratic state and the institutions are not under question, so there is less need to promote the ethic. In the newer democracies there is clearly a need to promote it along with an open debate on intelligence and interesting civilians in the field.(45) To the best of my knowledge, none of the new democracies have yet focused on changing the profession. However, as stated more than once in this report, much dealing with intelligence is secret and outsiders would be unlikely to know about reforms in training. Or, better stated, those who knew would not be likely to divulge the information.
Conclusion
All nations engage in intelligence activities at one scale or another. They must do as other countries do. No nation can afford to not know what is going on outside and inside their country. Similarly, nations must be ready to counter other countries' efforts to influence developments in their country. In most of the world, intelligence services of authoritarian regimes were central to the survival of those regimes and acted in the most negative manner imaginable. Today, in the midst of challenges to democratic consolidation, seeking to ensure democratic control over intelligence is both necessary and extremely difficult. In many countries there is virtually no public recognition of this fact. Without decisive action, however, the intelligence apparatus will remain a state within a state and prevent democratic consolidation. Like all else in civil-military relations, the challenges are many and it requires continual efforts on the part of civilians and officers to achieve the most appropriate balance of efficiency and transparency for the country. As illustrated at times in the body of this report, a small but significant group of countries have undertaken to reform their intelligence systems and have generated a public debate. This is an area in which international assistance is available, and on which there is a remarkable and rapidly increasing body of useful literature.
A Note on Sources and Expertise
The literature on intelligence is routinely broken down into four categories: memoirs of retired intelligence professionals; exposes by disgruntled former professionals, journalists, and activists; government reports, studies, and documents; and academic studies. Of these four categories only the last is largely objective. The other three are motivated by personal, partisan, or national goals, and thus contain some kind of bias or "agenda." Further, the literature in any one category is not so abundant that the interested student can dispense with material in any one of these categories. This is not the place to assess the literature in general, but to highlight that there is much material on the United States and now Russia, less on European democracies and South Africa, and very little on the new democracies in book or journal articles. Now, with the Internet there are available sources of information on aspects of intelligence throughout the world. There is not, however, to the best of my knowledge, any literature to provide the background and discussion of issues in which to locate this current Internet information on the new democracies. In sum, the material is sketchy and an overall conceptual framework is yet to be written.
To write this article I drew on the available literature less for inspiration and analysis, and more for examples of the points I wanted to make. My background and current activities provided the framework for the study. I attended graduate school in Political Science at the University of California at Berkeley during the 1960s. After completing my studies in 1969 I taught at McGill University, Montreal until 1987. During that time I researched "hot" topics; first politics and religion in Brazil (during the authoritarian regime) and later the Portuguese Revolution and its path to democratic consolidation. During that period of two decades, I had occasion to meet intelligence agents abroad who would ask me lots of questions but never told me anything. Lacking reciprocity I avoided contact with them. In 1987 I joined the Naval Postgraduate School and in 1989 became chairman of the Department of National Security Affairs. In that position I qualified to receive a high-level security clearance because I had to attend meetings and read documents requiring it. Having the clearance allowed me to learn quite a lot about the United States intelligence community with seminars and meetings at the CIA, Defense Intelligence Agency, National Security Agency, Office of Naval Intelligence, Office of the Joints Chiefs of Staff and Office of the Secretary of Defense at the Pentagon, regional headquarters of the armed forces, and American embassies abroad. Because our department offers one of the two masters degree programs in intelligence (the other is the Joint Military Intelligence College), we have alumnae in intelligence positions throughout the community. Through these contacts, and involvement with our courses, faculty, and students, I became interested in intelligence as a field of study. I could then appreciate the "one-way street" of information in my earlier experience since I had not been "cleared" at that time. Unfortunately for the field of study, as noted in the text of this article, most of intelligence is "slippery" in that there is much information on systems, tradecraft, and wiring diagrams, but little analysis of intelligence as an organization and a system. The effort is put into the analysis of the information and not the organization.
When the Center for Civil-Military Relations was founded in 1994 I became involved as director of the programs in Latin America. It was clear to me, with my background in Brazil and Portugal, that intelligence is a core topic in civil-military relations. Consequently we include a block of study on it in most of our programs in the region. We also developed a full week program on the topic of intelligence and democracy held in Buenos Aires in August 1998. The experience throughout Latin America brought me into contact with officers who are intelligence professionals and a small number of civilians who are interested in intelligence. The examples used to illustrate the text above are drawn from these contacts. Hopefully the democratic consolidation of that continent and others will continue apace, and the elected civilian leadership in all of the countries will feel secure enough to assert control over intelligence.