RESDAL - Últimas producciones de los miembros

Democracy and Effectiveness

Steven C. Boraz and Thomas C. Bruneau
July 2006

Steven C. Boraz, a U.S. Naval Intelligence officer, recently completed a Federal Executive Fellowship at the RAND Corporation in Santa Monica, California. Thomas C. Bruneau teaches in the Department of National Security Affairs at the U.S. Naval Postgraduate School in Monterey, California. The views expressed in this essay are the authors’ alone, and do not necessarily reflect the views of the U.S. Navy or the U.S. government.

One of the most difficult and least explored challenges confronting new democracies is that of reforming their intelligence services. Even for long-established democracies, the need for civilian agencies dedicated to protecting national security through the gathering and analysis of intelligence (as well as occasional covert action on the basis of such intelligence) poses serious problems. Democracy requires openness in the flow of information and discussion, while intelligence work often demands secrecy. Maintaining agencies to do such work in the midst of a generally open political culture is a challenge for any democracy. Democratizing or newly democratic countries, however, must deal with the even more arduous task of transforming intelligence bureaucracies that once served undemocratic regimes.

An essential part of the work of democratic consolidation, then, is the cultivation of intelligence organizations that will respect the democratic system, even as they routinely deploy secrecy (and occasionally force) in its defense. Two of the countries discussed in the essays that follow—South Africa and Taiwan—have met the challenge of intelligence reform in varying ways; the third country discussed—Russia— has seen an intelligence establishment inherited from Soviet days promote a recent backslide toward authoritarianism ⁽¹⁾.

To date, scholarship on democratization has tended to focus on the need to remove military and police forces from authoritarian control, and has neglected the question of who controls the national intelligence community (IC). Reforming intelligence and placing it under civilian control can be more difficult than achieving the same with militaries. Resistance may come not only from within the IC, but also from those who are tasked with overseeing intelligence. Nonetheless, with some adaptations, scholarship on the topic of civil-military relations provides a useful template for understanding the question of intelligence reform in a democratic regime. Ideally, the result of well-conceived reform will be an IC that is not only more “democracy-friendly,” but also more effective at its core mission of helping to protect the nation from actual and potential enemies who will typically seek to hide the truth about their own intentions and capabilities.

What precisely is intelligence? Mark Lowenthal defines it as a process, a product, and an organization ⁽²⁾. These in turn must all aim at two goals: to inform policy, and to support operations (whether of a police, military, or covert nature) in defense of national security. To accomplish these two missions, intelligence organizations carry out—and must integrate—four primary functions: collection, analysis, counterintelligence, and covert action. Collection managers and analysts work together in order to gather information and discern its significance; counterintelligence personnel work to protect state secrets while relying on analysis to help root out spies; and covert actions must be grounded in effective collection, analysis, and counterintelligence.

Collection. Intelligence agencies typically gather information through a variety of means including human intelligence, which is data collected directly by people and includes information from ambassadors and defense officials, information obtained at conferences and social events, and information obtained clandestinely. Technical means of intelligence collection include signals and imagery intelligence derived from communications intercepts, radar, telemetry, and images captured from overhead or the ground. Technically derived data captured by means other than signals or images is called measurement and signatures intelligence. Finally, there is open-source intelligence, by which professionals use more or less publicly accessible research tools such as periodicals, the Internet, seminars, and the like.

Analysis. What turns raw information into useful intelligence is the act of analysis—at once the core skill and perennially biggest challenge in the craft of intelligence. Analysis can never be reduced to a technical exercise. It requires not only processing gigantic quantities of data from multiple sources, but deciding which conclusions to draw and why, often in the face of considerable uncertainty.

The process of putting collection and analysis together to create reliable, accurate intelligence is referred to as the intelligence cycle ⁽³⁾. This cycle begins when a policy maker—which in a democracy will be an elected official or someone duly authorized by an elected official directs the IC to collect information, analyze it, and report the resulting intelligence to decision makers.

The collection and analysis process presents three significant problems for any democracy. First, if intelligence is to be of real value to decision makers, it must be independent. It is a challenge for any intelligence agency to provide accurate and independent intelligence when there may be pressure from the policy makers to produce “correct” answers ⁽⁴⁾. Second, intelligence organizations must be legitimate and professional enough to respect the rule of law if direction from politicians is illegal ⁽⁵⁾. Third, decision makers often do not know enough to task the IC appropriately. When civilians do not know enough, it will most often be the IC itself that will fill the void and provide its own direction, the dangers being obvious ⁽⁶⁾.

Counterintelligence. Generally, U.S.-based intelligence scholars identify counterintelligence as the protection of the state and its secrets against other states or organizations. In a more global context, however, the term counterintelligence may be more aptly termed “security intelligence.” Security intelligence has been defined as “the state’s gathering of information about and attempts to counter perceived threats to its security deriving from espionage, sabotage, foreign-influenced activities, political violence and subversion.”⁽⁷⁾

Covertaction. Covert actions are those activities—ranging from propaganda to paramilitary operations—which are intended to influence another state by means that cannot be traced to the state behind the actions, or which that state can at least “plausibly deny.”⁽⁸⁾ This is where intelligence-agency activities pass beyond the gathering and interpretation of information into the realm of deeds meant not merely to understand events, but to shape them, possibly by force. It is where intelligence activities most closely resemble military operations, with the difference that covert intelligence activities are typically characterized by a greater and more systematic pursuit of secrecy.

While not every country will need, want, or be able to afford extensive capabilities in each of these areas, nearly all sovereign states carry out at least some intelligence activities and thus possess some form of intelligence organization. These four functions, therefore, define the general framework within which the problem of democracy and intelligence must be understood.

The Template of Civil-Military Relations

We believe that, with a few adaptations occasioned mostly by the secrecy involved in intelligence work, the challenge of controlling intelligence agencies can be usefully thought of as similar to the task of fashioning democracy-friendly forms of civil-military relations (CMR).

There is a respectable scholarly literature on CMR that stresses the importance of maintaining democratic civilian control as well as understanding the roles and missions that militaries do and do not perform well ⁽⁹⁾. The four functions of intelligence listed above suggest a path by means of which we may analyze intelligence-community structures.

From the point of view of a democratic regime, the three great issues in CMR are: 1) Do electorally and constitutionally accountable civilian officials control the military? 2) Do they handle it in such a way that it can be effective at achieving its roles and missions from warfighting and peacekeeping to counterterrorism or emergency-response work? 3) Is all this being accomplished with maximal efficiency?

Because of the secrecy that necessarily surrounds intelligence activities and budgets, the third question borrowed from the field of CMR is hard to answer with anything like a credible analysis of costs and benefits. The first two questions, however, may be usefully asked not only about militaries but about civilian intelligence agencies. Whether the spies answer to elected civilians is often fairly easy to say. From our personal observations we know that they do in Argentina, Brazil, El Salvador, and Romania, to mention just a few countries. The question of whether the spies are good at what they do will usually be murkier (with the lion’s share of case data coming from established democracies), but there are usually enough leaks and failures to offer reasonable insights into the matter across a range of countries. While it is regrettable that only two-thirds of the CMR framework applies to matters of intelligence, CMR remains more useful than any competing alternative with which we are familiar, and hence is still the best choice as an intellectual guide to the problem of democratic intelligence control.

This control may be said to exert itself through the processes of direction and oversight. Direction is the guidance that civilian authorities give to their nation’s intelligence community regarding its overall mission. Such guidance can be embodied in day-to-day orders and feedback as well as in a written national-security strategy. Direction is therefore forward-looking, or at least concurrent. Oversight, by contrast, is a systematic process of reviewing an intelligence community’s actions, organization, budget, internal policies, and legal constraints with the goal of improving effectiveness as well as ensuring democratic civilian control ⁽¹⁰⁾.

Control over what the IC does may come from members of the executive, legislative, and judicial branches; from within the IC itself; or from external actors such as the press or NGOs. The executive normally sets the IC’s mission and basic organization. As primary consumers of intelligence, executive-branch officials give the IC most of its ongoing direction. Legislatures normally create the key organizational, budgetary, personnel, and legal-oversight mechanisms of an IC, as well as provide balance to the executive branch. Independent courts may use their authority to safeguard citizens’ rights against government intrusion. In more mature democracies, legal-accountability mechanisms are also often built into intelligence organizations themselves in the form of in-house counsels and inspectors general (IGs). In general, internal controls include not only counsels and IGs but also the IC’s own professional ethos and institutional norms. Controllability may also be promoted by the existence of multiple intelligence organizations. External controls in democracies include a free press, independent lobbies and think tanks, and, especially in the case of new democracies, NGOs strengthened by overseas ties and support in the work of monitoring their country’s intelligence agencies. Argentina, Colombia, and Romania provide salient examples of where the press and think tanks are vigilant in monitoring and reporting intelligence activities. Few democracies, even mature ones, have organizations or insttutions that span this spectrum ⁽¹¹⁾.

Even when legislative control exists, controversies over intelligence control between the executive and legislative branches can reshape the norms under which ICs must operate. The U.S. Hughes-Ryan Amendment, passed by Congress in 1974 amid questions about the IC’s involvement in Vietnam and Watergate and its infiltration of student groups, is a case in point. This law requires the president to submit to the relevant congressional committees in “timely fashion” a written “finding” stating his view that a given covert action is important to U.S. national security. Many saw this as an intrusion into executive authority, and exceptions were made to the rule in order to ensure that Congress had not overstepped its constitutional bounds. The notification system was infamously circumvented during the Iran-contra affair in 1985–86, resulting in a 1991 rewrite of the law ⁽¹²⁾. Even within the last year, there have been public controversies over the executive’s decisions to allow U.S. military forces to conduct certain covert actions, and to authorize the National Security Agency to conduct electronic-surveillance operations that may involve the monitoring of U.S. citizens’ communications.

Impediments to Reform

We have found in our studies of intelligence reform that politicians are reluctant to seek to control intelligence agencies. There appear to be a number of reasons for this. First, particularly in the case of newer democracies, elected officials may prefer to avoid any dealings with an intelligence apparatus that carries the stigma of having been a tool of repression under the old regime. A second and related reason is that politicians may wish to be able to disavow knowledge of operations and so avoid seeming as if they have condoned illegal activity. Third, most civilian politicians do not know enough about intelligence to be able to have an informed opinion. Learning the intricacies of the secret world can be a difficult task anywhere, and even a dangerous one in a newer and none-too-secure democracy in which authoritarian holdovers remain strong, as in Russia or Moldova. A fourth reason is that for legislators there is usually little or no inducement to undertake such work: A seat on a classified intelligence-oversight committee, after all, is not normally the best perch from which to garner publicity or troll for votes. A final reason is the ill-defined national-security role that lawmakers often play. Of the few new democracies that even have written nationalsecurity strategies, fewer still give much role to the legislature in the process.

While elected or politically appointed officials may be reluctant to put a bridle on the intelligence community, so likewise may the IC feel restive at the prospect of such “outside” control. To begin with, there is a natural tension between intelligence professionals, who think of themselves as needing no help from novices, and politicians, who may not fully know or understand what the IC is doing, but who nonetheless want to make sure that laws are being obeyed and public funds are being well spent. A second and closely related reason is that intelligence organizations often believe that more freedom from budgetary, legal, and other restrictions for the IC translates into improved security for the country. As Philip B. Heymann notes: “The deepest problem of controlling domestic intelligence agencies is that they have both the capacity to hide disobedience and a justification for not taking constraints seriously ⁽¹³⁾.” A third reason is that major portions of the IC are subordinated to the military, which typically opposes outside control of intelligence capabilities for fear that the focus on protecting and aiding operational forces will be lost. This was an issue that surfaced in the debate over recent reform in the U.S. intelligence community.

The final—and probably most important—reason is that intelligence professionals tend to believe, fairly or not, that politicians fail to make national security their first priority, that they cannot or will not handle secret information properly, and that they will blame intelligence personnel if anything goes wrong. In countries such as Brazil, El Salvador, South Africa, and Uruguay, where political parties that include former guerrillas have come into government, mistrust is intensified.

Elements of civil society that wish to help promote democratic control of intelligence face several hurdles. The highest is the matter of access. Those outside government must rely on leaks or an informal entrée into the IC that is hard to come by given the adversarial climate which often subsists between intelligence officials and their civilian critics. The media, likewise, will be seen as people who make headlines out of the IC’s failures but who cannot even be told of its successes, given the need to keep IC sources and methods secret.

Beside the issue of control is that of effectiveness—a factor that democratic regimes as much as any others must take into account when pondering any changes affecting the intelligence community. An obvious and fairly simple way in which IC effectiveness can be improved is to institute standards for the recruitment and training of intelligence personnel. An additional and more complex step, as highlighted recently by the work of the 9/11 Commission in the United States, is to resolve the problems of coordination that can all too readily beset ICs which are made up of separate and potentially competing organizations. Removing barriers to interagency cooperation and informationsharing is key. In addition to all the normal obstacles to bureaucratic reform, such as tensions over “turf” and scarce resources, those who would reform ICs must overcome intelligence agencies’ worries that distributing their “product” more widely will compromise the secrecy of sources and methods.

Established democracies such as the United States, Australia, Canada, and recently the United Kingdom have tried to boost IC effectiveness by using multiple agencies to provide peer reviews of finished intelligence products. They have also created a process termed “team-B analysis” to challenge established assessments within the IC and to offer alternative interpretations, sometimes using organizations outside of the IC. While some democracies use these processes, referred to collectively as “competitive analysis,” in order to produce an IC consensus, other countries make alternative interpretations a regular part of finished intelligence products ⁽¹⁴⁾. Newer democracies typically have yet to come to terms with competitive intelligence and tend to provide the executive with finished products from multiple agencies (which is also the case in France), leaving the coordination of “team-B analysis” to the executive level—a process that can lead to overly politicized outcomes. Some newer democracies, such as Romania, have created a clearly defined role for each IC agency and established an intelligence-coordinating office that should help make competitive analysis less necessary.

Advancing Democratic Control

Even with the greater emphasis on effectiveness in the more established democracies, there is a continuing concern with democratic civilian control. Nor is this surprising given democracy’s need for transparency and the need for secrecy in intelligence matters. The basic requirements for democratic control of intelligence matters are clear enough: The executive, the legislature, and the courts must continue to play the roles sketched above; the IC must be organized and run along professional lines and without a single agency holding all the power; and external watchdogs must take their work seriously.

In order to promote democratic civilian control and IC effectiveness, countries may undertake several tasks. Our comparative research and experience in countries that have pressed these tasks successfully (Argentina, Brazil, Romania, South Africa, and Taiwan) and not so successfully (Guatemala and Russia) lead us to offer the following list of endeavors, to be pursued more or less simultaneously:
1) Raise public interest and pressure. We have found that a good first step toward a more effective and democracy-compliant IC is the launching of an informed public debate. Specific issues may include determining which institutions will have an oversight role and whether an IC is organized properly to support decision makers. The challenge is to break through the pervasive public apathy toward or fear of intelligence. In some older democracies, including France, Great Britain, and the United States, NGOs and the media have stimulated a fairly regular debate, which is periodically galvanized by intelligence failures that become public. The aftermath of 9/11, the failure to find weapons of mass destruction in Iraq, global terrorism, and transitions to democracy have brought this debate to the fore in many countries. In Argentina, for example, a small number of forward-thinking politicians regularly discuss the need for civilian control and other intelligencerelated matters. The Brazilian government’s commitment to revise the intelligence system has generated public consideration. In Romania, the media have played a crucial role in promoting democratic control of intelligence. Such debates aid democracy by 1) helping “outsiders” more realistically to assess the need for, value of, and limits on the work of the IC in a popularly governed society; 2) creating legitimate avenues for civilians who wish to become expert in intelligence matters; and 3) putting pressure on officials to act with greater transparency.

2) Increase civilian awareness and competence. It is also important to motivate civilians, especially those in defense ministries and legislatures, to learn about intelligence so that they can control it. More civilian involvement in intelligence means more transparency. Many postauthoritarian countries have histories of military domination over intelligence. Civilians must be urged to study intelligence matters and to push IC professionals for due cooperation, if not respect. This process may begin with a formal and public governmental commitment to review the work and organization of the IC with the goal of establishing new policies and possibly new agencies. This happened in Brazil and South Africa, mostly as a result of political and institutional bargains made during processes of democratic transition that saw the old-guard IC’s prerogatives diminished and finally largely eliminated.
Democracies must ensure that civilians can have viable careers in intelligence. Otherwise, as in CMR generally, civilians will not come forward. One thing that countries can do is to open up their militaryintelligence training schools to civilians who might one day become involved in the oversight process. Civilians can also begin to learn about intelligence by reading the open-source literature from several countries, and taking advantage of cooperative intelligencetraining arrangements with other nations.

3) Institutionalize processes that support transparency and effectiveness. Countries must ensure that developing national-security strategies, ensuring feedback from policy makers to the IC, and coordinating the efforts of various intelligence agencies become tasks which are pursued in a regular, institutionalized way. New presidents and prime ministers should undertake systematic reviews (and if necessary, updates) of national strategy, perhaps under a legal mandate such as is furnished in the United States by provisions of the National Security Act of 1947 and the Defense Reorganization Act of 1986. The opportunity to provide the IC with direction keyed to democratic transfers of power is an excellent one. Also, a chief executive’s cabinet or staff may include civilian intelligence or strategy coordinators who can work directly with the IC in ways that both enhance the national-security document and underline the reality of civilian democratic control.
To promote greater effectiveness, democratic governments should provide for standing committees in the appropriate branches of government that have access to intelligence and can regularly review how the IC is doing its job. Doing so can substantially improve collection, analysis, and coordination as well as deter abuses. A standing oversight committee can also review budgetary and staffing decisions, vet nomineesfor top intelligence posts, and make organizational recommendationsto the IC and government. Argentina, Brazil, and Romania currently have such committees in their legislatures.

There is a natural tension between this kind of oversight body and the executive branch, for which the intelligence services primarily work. For oversight to function properly, legislatures must be able to call intelligence personnel into an inquiry (a closed one if it is necessary to discuss classified information). In many countries, including established democracies, we have seen IC agencies or the executive branch stall, citing national-security issues, when in fact it is a question of legality or propriety. This said, countries should ensure that inquiries are not so burdensome as to hamper the overall effectiveness of the IC—admittedly a difficult balance to achieve.

Making institutional changes within the IC will also be crucial. One of the greatest impediments to creating an effective IC is bureaucratic infighting among the agencies that compose it. The IC’s leaders will need to work out and emphasize formal requirements for joint or interagency operations, determine how the competitive-intelligence process should work, and ensure that the IC has a figure or organization that can serve as the IC’s voice when speaking to the various branches of government.

While governments often discuss the idea of applying objective, quantifiable metrics to measure intelligence effectiveness, we counsel caution toward such schemes. Counting the number of images or signals gathered is one thing; putting a number on the accuracy or usefulness of a complex and dynamic analytical interpretation is something else altogether. Our supposition, rather, is that focusing on improving coordination, professionalism, transparency, and trust—all of which are crucial norms for intelligence under democracy anyway—will naturally result in a more effective IC serving under knowledgeable politicians who may not be able to quantify IC performance, but who will know a “job well or poorly done” when they see it.

4) Foster a political culture that supports and trusts intelligence in society and inside the IC. Governments need to foster national political cultures that will support robust but accountable and well-controlled intelligence communities. This will mean raising the level of trust between the IC and the organizations charged with overseeing it, as well as within the IC itself. As civilians learn about intelligence through some of the various means noted above, they will increase their credibility in the eyes of intelligence professionals. Joint educational and professional activities designed to bring together IC staffers and civilians who have official responsibilities regarding intelligence may help to forge politically healthy personal bonds.

Trust, of course, is a two-way street: The IC must accept control by democratically elected civilians, and those civilians must in turn refrain from releasing classified information for personal or political reasons. While trust cannot be legislated, laws that punish leaks of secret information could help to ease the IC’s wariness about opening up to more contacts with civilians.

Trust also needs to obtain among the several agencies that make up the IC. The presence of a lead agency or single intelligence director who can balance and channel competition among arms of the IC should help, though this will always work best if the lead agency has real budgetary authority. Also useful is an oversight system that is independent and outside the executive branch, and which can recommend resource allocations based on merit rather than political favoritism. Higher intra- IC trust will promote greater effectiveness through, among other things, the smoother and speedier sharing of information. In too many of the cases that we have studied, bureaucratic considerations have impeded one agency with knowledge from revealing it to other agencies that might have made good use of it.

5) Professionalize the intelligence services. The professionalization of the IC may be thought of as the essential groundwork of internal control. Professions typically are characterized by expertise, corporateness, and responsibility ⁽¹⁵⁾. As regards the IC, “expertise” means special skill at collection, analysis, counterintelligence, and covert action. While IC professionals perform such a diverse array of tasks that there will likely be few common schooling requirements even within the intelligence services of a given country, this does not mean that there is no room for the professional education of IC personnel. On the contrary, we suspect that one of the surer—and sadly underused—ways to boost IC effectiveness and professionalism is entry-level and continuing education programs designed to ensure a starting baseline of knowledge and to offer specialized certifications (in counterintelligence, say, or analysis) to IC staffers as they pursue their careers.

“Corporateness,” that which unifies or characterizes individuals as intelligence professionals, arises in the IC from access to classified information. While a concern with secrecy or confidentiality plays a significant but limited role in professions such as the law, medicine, and the clergy, this concern suffuses the intelligence profession. The practices of compartmentalization, security clearances, and “need-toknow” access to information create a sense among IC staffers working on classified projects that they are members of a unique group. A certain sense of impunity may also develop, since if nobody else knows, then why should those who do not know control those who do? In order to counteract this tendency, standards of respect for democratic civilian authority must be clearly stated and taught, and penalties for unprofessional behavior must be meted out with equity.

Fostering Responsibility to Democracy

Third, the “responsibility” of the intelligence professional is to defend the democratic state in ways accountable to democratic authority. Too much stress on expertise and the corporateness of “those who know the secrets” can result in a profession that largely governs itself according to its own definition of responsibility. Newer democratic regimes may find their situation particularly precarious in this regard, since the regimes that they replace typically have no tradition of accountability to the people or the rule of law, and indeed may even have featured “independent security states” in which the IC answered essentially to no one, not even its ostensible authoritarian masters.

Fostering a sense of responsibility to democracy is incredibly important: Even in stable democracies, incidents come to light which cause concern that intelligence officers are not serving the state, are serving it only in limited organizational terms and not in line with the democratically elected leadership, lack a strong-enough corporate ethic to say no when told to do something illegal, or feel that illegal activity is justified because it supports national security. The placement of legal counsel and inspectors-general within intelligence agencies is one possible fix.
Regular training in the laws and regulations governing what intelligence officials can and cannot do will also help.

Argentina, Brazil, Romania, and South Africa, for example, are making major efforts to promote a sense of responsibility to democracy in their ICs. Connected with this is the need, in many countries, actively to recruit more civilians into intelligence careers, rather than following the path of least resistance that will result in an IC top-heavy with exsoldiers (many new democracies, it should be recalled, are dealing with traditions of military-dominated ICs). And what if civilians cannot be found in large numbers? Can the ethic of responsibility still be decisively promoted? Too few countries are giving this ethic the sustained and systematic backing that it deserves ⁽¹⁶⁾. In the older democracies, the larger society supports responsibility to the democratic state, and political institutions are generally not questioned, so there is less need to promote the responsibility ethic. In newer democracies, the need to actively and explicitly promote this ethic is clear, as are the needs to promote an open discussion of intelligence and to interest civilians in learning about or even working in the field ⁽¹⁷⁾.
In sum, professionalization is fundamentally important in the intelligence corps and it will be among the most difficult things to achieve in a democratic transition. Professionalization will also have the most to do with creating a cadre of individuals who can form the core of an effective institution.

Intelligence reform will undoubtedly continue to be an extremely important issue, and sometimes a dilemma, for every democratic nation. For those countries that are (or will be) on the path toward democratic consolidation, restructuring intelligence organizations is an exceptionally difficult task, with many pitfalls and no clear roadmap. Yet there is no question that democratic consolidation can never be said to have occurred in a country until its intelligence apparatus comes under effective democratic civilian control. Recent experience in the U.S. and British ICs highlights the complexity of intelligence reform even in longstanding democracies. Moreover, it shows that intelligence reform is not a one-time event, but, like democracy itself, requires ongoing attention, oversight, and institutional engineering to be effective. Just as establishing control over the IC in a new democracy is a critical step in democratic consolidation, the ever-present threat of terrorist attack that hangs over so many countries requires the utmost attention to effectiveness. In our view, democratic control need not hamper effectiveness; indeed, if promoted properly, it should enhance it.

All nations engage in intelligence activities on some scale. If these activities are to inform policy and to support operations, they must be accountable to elected officials. If no outside entity is setting operational goals or laying out policies, then the IC’s work may be of little or no value to anyone outside its confines. Yet in all too many countries there is still virtually no public recognition of the need for democratic civilian authorities to step in assertively with goals and policies for the IC to pursue and follow. Without decisive action to support democratic controls, an authoritarian intelligence apparatus can remain a state within a state and prevent democratic consolidation, which has been the case in Russia and (to a lesser extent) in Moldova.

Much like the field of civil-military relations, the field of “civil-IC relations” demands that both sides of the relationship work continually toward achieving the best balance of effectiveness and transparency for the country. Many countries have undertaken reform of their intelligence systems and generated a public debate on the matter. Global terrorism and recent public intelligence failures by the United States and Britain have also invigorated more open discussion. Countries such as Argentina, Brazil, Colombia, Romania, South Africa, and Taiwan all have had intelligence-control issues brought to various public forums in recent years, while other countries, such as Russia and Moldova, are either stagnant or backsliding with respect to democratic control of their intelligence services. Establishing democratic control over intelligence can help to achieve democratic consolidation as well as create effective intelligence organizations—a twenty-first century imperative.

[1]. Argentina, Brazil, and Romania are among the other countries that have restructured their intelligence communities within the past decade to provide greater civilian control and accountability. For more discussion of these and other cases, see our edited volume on comparative experiences with intelligence reform, to be published by the University of Texas Press in 2007.
[2]. Mark Lowenthal, Intelligence: From Secrets to Policy, 3rd ed. (Washington, D.C.: Congressional Quarterly Press, 2005), 9. Lowenthal is, in turn, heavily influenced by Sherman Kent’s Strategic Intelligence for American World Policy (Princeton: Princeton University Press, 1951), wherein Kent describes intelligence as knowledge, organization, and activity.
[3]. The intelligence cycle is generally conceived of as having five phases: planning and direction; collection; processing (where raw data are put into some sort of useable format); analysis and production; and dissemination.
[4]. Intelligence regarding weapons of mass destruction in Iraq provides a contemporary case in which some claim that the IC of the United States may have provided politically distorted intelligence. See, for example, Michael Scheuer, Imperial Hubris: Why the West Is Losing the War on Terrorism (Washington, D.C.: Brassey’s, 2004). It should be noted, however, that the Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction (also known as the Robb-Silberman Commission) found that “[t]he analysts who worked Iraqi weapons issues universally agreed that in no instance did political pressure cause them to skew or alter any of their analytical judgments.” See the Commission’s Report to the President of the United States, 31 March 2005, 27.
[5]. A clear case where intelligence agencies carried out the bidding of a president was France during the administration of François Mitterrand (1981–95). He set up a cell of intelligence officials in the Elysée Palace to protect secrets and spy on his political enemies.
[6]. South Africa under apartheid is a case in point. The intelligence services there provided their own guidance to carry out repression, kidnapping, torture, and murder to support their goal of maintaining apartheid. Even in military regimes, such as Brazil’s from 1964 to 1985, the main intelligence agency operated largely outside the sway of the military government.
[7]. Peter Gill, Policing Politics: Security Intelligence and the Liberal Democratic State (Portland: Frank Cass, 1994), 6–7. Gill groups security-intelligence services into three categories: domestic intelligence, political police, and an independent security state. Among these three general types of security-intelligence services, the independent security state’s penetration of society is the most extensive and it wields virtually unchecked power over the regime and population.
[8]. See Mark Lowenthal, Intelligence, ch. 8.
[9]. This approach to civil-military relations is developed and illustrated in Thomas C. Bruneau and Scott D. Tollefson, eds., Who Guards the Guardians and How: Democratic Civil-Military Relations (Austin: University of Texas Press, 2006).
[10]. Some of the more important works on controlling intelligence include Glenn P. Hastedt, Controlling Intelligence (Portland: Frank Cass, 1991); Loch K. Johnson and James J. Wirtz, eds., Strategic Intelligence: Windows into a Secret World (Los Angeles: Roxbury, 2004), Part VII; and Hans Born and Ian Leigh, Making Intelligence Accountable: Legal Standards and Best Practice for Oversight of Intelligence Agencies (Oslo: Publishing House of the Parliament of Norway, 2005).
[11]. In Britain, there is an Intelligence and Security Committee of Parliament (whose members are named by the prime minister) that reviews reports on intelligence. While the committee lacks full access to information and reports only to the prime minister, over the years it has had an increasing influence on British intelligence policy. In France and Japan there is no separate system for overseeing the intelligence services. See Hans Born and Ian Leigh, Making Intelligence Accountable, as well as Peter Chalk and William Rosenau, Confronting the Enemy Within (Santa Monica, Calif.: RAND Corporation, 2004).
[12]. The U.S. Fiscal Year 1991 Intelligence Authorization Act codifies two requirements for any covert action: 1) a nonretroactive presidential finding, and 2) notification of the House and Senate intelligence committees of the action as soon as possible after issuance of the finding and before the operation starts. An exception remains for “extraordinary circumstances,” in the event of which the president need only fully inform the committees “in a timely fashion.” See Public Law No. 102-88, 105 Stat. 429.
[13]. Philip Heymann, “Controlling Intelligence Agencies,” Harvard University, Project on Justice in Times of Transition 2000 at http://ksg.harvard.edu/justiceproject/Heymann percent20WP.doc.
[14]. Great Britain, up until very recently with the acceptance of the Butler Committee report, was a good example of a consensus intelligence provider. The British system now includes a separate team to challenge assessments, and will shortly be appointing a “challenger-in-chief” to test intelligence material before it is presented to policy makers. The United States, and to a lesser extent Australia and Canada, use competitive-intelligence methods to provide alternative analyses.
[15. Although the sociological literature on professions is huge, going back at least to Max Weber, that which is most pertinent here is the literature on the military as a profession. The classic is Samuel P. Huntington, The Soldier and the State (Cambridge: Harvard University Press, 1957). The most useful additions and critiques include Bengt Abrahamsson, Military Professionalization and Political Power (Beverly Hills, Calif.: Sage, 1972); and Peter D. Feaver, “The Civil-Military Problematique: Huntington, Janowitz, and the Question of Civilian Control,” Armed Forces and Society 23 (Winter 1996): 149–77.
[16. This was long ago recognized by Hastedt: “Only by seeking to structure how intelligence professionals see their job can one hope to prevent abuses from occurring in the first place or ensure responsiveness.” See Glenn P. Hastedt, Controlling Intelligence, 14.
[17. The other side of the recruitment question is the retirement of intelligence professionals. It is important for governments to ensure that their intelligence organizations create stable career progressions based on merit, including provisions for decent retirement after service. This ensures loyalty and gives IC staffers options beyond trying desperately to hang on to IC jobs as long as possible, or even worse, turning to illegal activities since their skills are not readily transferable to other legal occupations.

Source:
Journal of Democracy, Volume 17, Number 3, July 2006
© 2006 National Endowment for Democracy and The Johns Hopkins University Press
Reproduced width permission of The Johns Hopkins University Press