The International Politics of Intelligence Sharing

James Igoe Walsh

• Publisher: Columbia University Press
• Year: 2009
• eBook ISBN: 9780231520881
• Topic: Politics & International Relations
• Subtopic: Diplomacy & Treaties
• Index: Politics & International Relations

The cross-border sharing of intelligence is fundamental to the establishment and preservation of security and stability. The American invasion of Iraq in 2003 was based in part on flawed intelligence, and current efforts to defeat al Qaeda would not be possible without an exchange of information among Britain, Pakistan, Indonesia, and the United States. While critical to national security and political campaigns, intelligence sharing can also be a minefield of manipulation and maneuvering, especially when secrecy makes independent verification of sources impossible.

In The International Politics of Intelligence Sharing, James Igoe Walsh advances novel strategies for securing more reliable intelligence. His approach puts states that seek information in control of other states' intelligence efforts. According to this hierarchical framework, states regularly draw agreements in which one power directly monitors and acts on another power's information-gathering activities-a more streamlined approach that prevents the dissemination of false "secrets." In developing this strategy, Walsh draws on recent theories of international cooperation and evaluates both historical and contemporary case studies of intelligence sharing. Readers with an interest in intelligence matters cannot ignore this urgent, timely, and evidence-based book.

1 UNDERSTANDING INTELLIGENCE SHARING

CURVEBALL AND KSM

The charge that Iraq was actively developing chemical and biological weapons was an important rationale for the United States’ invasion of the country in 2003. To make this charge, the United States needed some evidence about Iraqi weapons programs, but reliable evidence was difficult to obtain. Relations between the two countries had been very hostile since the Gulf War of 1991, so the United States did not have diplomats stationed in the country who could search for information about Iraqi weapons. The president of Iraq, Saddam Hussein, maintained a brutally tight grip over his society, and any scientist, government official, or military officer even suspected of sharing weapons information or otherwise collaborating with foreigners was very likely to be tortured and killed. The United States did control satellites and aircraft that could thoroughly reconnoiter Iraqi territory for weapons activity. But the Iraqis had learned to hide underground and out of sight those activities that they wanted to keep secret. During the 1990s, the United Nations closely monitored Iraq’s military activities to ensure its compliance with provisions of the cease-fire ending the 1991 Gulf War that prohibited Iraq from developing weapons of mass destruction. These United Nations inspectors had been a valuable source of information about Iraqi activities, but Hussein expelled them in 1998.

This meant that the United States had very little capability itself to collect reliable information on Iraqi weapons development, so it turned to other countries to provide relevant intelligence. The German intelligence service, the Bundesnachrichtendienst (BND), shared information it had obtained from Rafid Ahmed Alwan. Alwan defected from Iraq to Germany in 1999 and was later interrogated by the BND, which, in an act of unintentional irony, gave him the code name Curveball. Curveball claimed that he had worked at a facility in Iraq that manufactured mobile biological weapons laboratories. Although the BND passed this information along to an American intelligence agency, it would not permit its American counterparts to question Alwan directly. Curveball’s claims formed an important part of the American argument for war: Secretary of State Colin Powell drew on them for his presentation to the UN Security Council seeking international support for the invasion, and President George W. Bush mentioned mobile biological weapons labs in his State of the Union address just two months before the invasion.

Curveball had lied. He had lied about his educational credentials (claiming to have graduated first in his class at university when in fact he had graduated last), about the reasons why he left Iraq in 1999 (he was being investigated for embezzlement), and about his work history (he drove a taxi and worked for a television production company in Iraq). All his claims to have helped manufacture biological weapons labs—key building blocks in the American case for war—were fabrications. His lies were revealed after the invasion when no mobile labs or other evidence of a weapons of mass destruction program were discovered. Curveball had duped his German controllers and, more important, the Americans who based their decision for war partly on the intelligence he provided. How could such an error have occurred? Perhaps, as many commentators have argued, American leaders wanted war with Iraq and simply used Curveball’s claims as a convenient justification.¹ A major part of the problem, however, was that Curveball was telling his falsehoods to the Germans, who in turn were sharing them with the Americans. This sharing made it difficult for the United States to verify to its own satisfaction that Curveball’s story was reliable, even though many American intelligence personnel had doubts about his credibility. Because of the barriers that the United States faced in collecting intelligence on Iraq, it was very difficult to ignore a source of information provided by another country.

Curveball illustrates many of the promises and problems that states encounter when they seek to share intelligence. Receiving intelligence from another country may be the most efficient or, in some cases, the only source of information. This makes shared intelligence very valuable, as it provides at least some information that decision makers can use to reduce the uncertainty they inevitably face when crafting foreign policy. But it also means that the recipient is at the mercy of the state providing the intelligence. The sending state may manipulate the shared intelligence to serve its own purposes, or it may fail to properly confirm important pieces of the intelligence that it passes along. This vulnerability is compounded by the fact that shared intelligence is shrouded in great secrecy to keep its source hidden from outsiders, thus making it even more difficult for the receiving state to independently verify its accuracy. How do governments overcome these barriers and mutually benefit from sharing intelligence? That is the main question addressed in this book.

Another story illustrates the answer that this book provides. Khalid Sheikh Mohammed was a leader in the al Qaeda terrorist organization and the architect of the multiple attacks on the United States on September 11, 2001. Mohammed, often referred to in official American accounts as KSM, was captured in Pakistan in 2003. His capture was the result of years of careful investigation and the close cooperation of intelligence agencies in the United States, Pakistan, and other countries. Quite by accident, Philippine authorities discovered that KSM had been planning attacks from their country. Additional intelligence came from suspected terrorists captured after the September 11 attacks and interrogated by American authorities in Afghanistan and by other governments across the Middle East. Acting on information provided by a detainee, American intelligence agencies were then able to monitor telephone communications among members of KSM’s organization in Pakistan. Someone with specific knowledge of his whereabouts gave this information to Pakistani officials. Although Pakistani forces were responsible for apprehending KSM, American officials were on the scene monitoring the situation. Pakistan then transferred KSM to American control, under which he was interrogated about his role in the September 11 attacks and other plots.²

This description of KSM’s capture illustrates that intelligence sharing is not simply an additional source of useful information but can often be a requirement for successful action. Even though the United States had the technical means to track some of the communications of KSM or his associates, this was not sufficient to engineer his capture. In addition, the United States needed intelligence provided by detainees and agents controlled by other countries to piece together a sufficiently detailed map of KSM’s activities. It needed the active cooperation of Pakistan in tracking KSM to his hideout in Rawalpindi. Intelligence sharing, therefore, can be a necessary part of successful foreign policy choices.

KSM’s case also demonstrates how successful intelligence sharing can be tied to other forms of cooperation. To secure the cooperation of the Pakistani authorities, for example, the United States after September 11, 2001, gave the country enormous amounts of intelligence of its own as well as economic and military aid. Thus the United States helped secure Pakistan’s willingness to supply the intelligence it needed by in turn providing Pakistan with its own intelligence as well as money and arms. Intelligence is a valuable commodity, and states bargain with one another to obtain the best possible return before agreeing to share it.

Finally, the United States worried that some countries might renege on their promises to share intelligence. Indeed, many of the countries that participated in the hunt for KSM may have had good reasons not to share their findings with the United States. For reasons discussed in chapter 5, Pakistan was especially suspect on this score. One step that the United States took to prevent such defection was closely supervising the pursuit of KSM. American officials collected some intelligence themselves, participated in organizing and carrying out the raid on the house where KSM was hiding, and insisted on taking custody of him. A key argument of my book is that such direct, hierarchical control of another state’s counterterrorism and intelligence efforts is an important tool for overcoming the problem of defection from promises to share intelligence. Even though states may want to receive intelligence from others, the secrecy surrounding it makes it difficult for them to determine whether it has been shaded or manipulated in some way before it is shared. Accordingly, hierarchy—by which I mean some direct control over another state’s intelligence activities—is a useful mechanism for overcoming such concerns about defection that has not been analyzed in existing works on sharing intelligence.

This book describes the benefits that states seek when they share intelligence, the barriers that they encounter when doing so, and the conditions under which forging a hierarchical relationship allows them to surmount these barriers. Intelligence sharing is a form of international cooperation, and we can learn much about the practice from explanations of cooperation developed for other domains. Intelligence sharing delivers to at least one participating state the benefits of more or better intelligence. The most important barrier to intelligence sharing is the fear that other participants will defect, in the sense of violating their agreement to cooperate, for example, by manipulating shared intelligence to serve their own ends. Moreover, the secrecy surrounding intelligence makes it very difficult for one state to tell when its partner has defected. This secrecy means that two common explanations of cooperation—mutual trust between participants, and the development of institutions and practices designed to provide information about compliance—are unlikely to be very effective in the area of intelligence sharing. Relational contracting, an alternative explanation of cooperation, has been applied to international politics and can add important insights to our understanding of the extent and nature of international intelligence sharing. Relational contracting is a part of transaction cost economics, which was first developed to explain the origins of business firms and was later applied to international politics and many other issues. Relational contracting draws attention to how states can incorporate hierarchical control, oversight, and monitoring mechanisms in their cooperative agreements in order to minimize defection by participants. Hierarchical agreements to share intelligence create a dominant state responsible for making important decisions and overseeing implementation. Such agreements also create the subordinate states that receive benefits such as shared intelligence, foreign aid, and protection from external threats in exchange for accepting the dominant state’s control. Hierarchy allows states to overcome their concerns about defection and to engage in mutually beneficial cooperation. This is particularly useful in the area of intelligence sharing, as it otherwise is difficult for states to determine whether their partners are complying with an intelligence-sharing agreement.

INTELLIGENCE AND INTELLIGENCE SHARING

Intelligence is the collection, protection, and analysis of both publicly available and secret information, with the goal of reducing decision makers’ uncertainty about a foreign policy problem.³ Intelligence is a type of, but is not synonymous with, information. Intelligence is information, or a process of obtaining information, that someone prefers to be kept secret. The targets of intelligence collection and analysis keep information about their capabilities and intentions secret from others. Military forces are a common focus of other countries’ intelligence efforts, and the targets of such intelligence collection try to mask their true capabilities and vulnerabilities in order to maintain the advantage of surprise over potential opponents. In turn, organizations charged with collecting intelligence on such targets do not reveal information about the sources and methods they use, since doing so would allow the targets to take countermeasures to secure their secrets from outsiders. Governments use secrecy to prevent others from knowing what they know or what actions they may take. But this understandable effort to secure secrets also makes it difficult for countries that agree to share intelligence to determine whether their partners are abiding by their commitments.

Intelligence is shared when one state—the sender—gives intelligence in its possession to another state—the recipient. Why do states share intelligence? Sharing intelligence is useful because decision makers often face a great deal of uncertainty when crafting foreign policy. They may be uncertain about the true intentions of friends and foes, the capabilities of others to deliver help or harm, the full range of policy options available to them, and the outcomes of implementing these policies. Intelligence about others’ political intentions and material capabilities is especially important to the areas of defense and security policy, in which the results of policy errors—such as surprise attacks or entanglements in protracted military conflicts—can be very costly.⁴ National governments devote substantial resources to collecting and analyzing intelligence. Nonetheless, decision makers are rarely satisfied with the quality and quantity of intelligence made available to them, and they often complain about the failure of intelligence agencies to anticipate important developments. This desire for intelligence is what leads governments to share intelligence with their counterparts overseas. The most important benefit from sharing intelligence is that it can give decision makers new perspectives on the problems they face and the likely effects of the policies they select.

What senders and recipients decide to exchange varies. Most countries share little or no intelligence with one another. In some cases, one state shares its intelligence in exchange for something else, such as foreign aid, security assurances, or diplomatic support. In other arrangements, the parties share intelligence with one another. Examples include pooling resources to analyze an intelligence target of interest to both countries, coordinating their networks of agents providing human intelligence, or jointly investing in new technologies to collect intelligence or in expensive intelligence assets such as satellites or listening posts. Agreements also vary in how extensively the parties share intelligence. In some relationships, such as that between the United States and Great Britain during the Second World War, the parties routinely share intelligence on a wide range of issues of mutual concern. Other agreements are more limited, with the parties sharing intelligence on some issues but not others. For instance, many current intelligence-sharing arrangements between the United States and countries in the Middle East and South Asia, which focus on intelligence concerning terrorist groups, take this form. Another way that intelligence-sharing agreements differ is how much one state directly controls another state’s collection and analysis of intelligence. During the cold war, instead of direct control, the United States and Britain collaborated closely, but each retained the authority to determine its own intelligence activities. The United States, however, directly managed and oversaw West Germany’s intelligence activities and, more recently, has supervised the restructuring of intelligence and security agencies in Latin America corrupted by drug-trafficking organizations.

The gains from sharing intelligence are greater if the participating states specialize. Specialization allows each to develop greater expertise regarding the targets and to create more focused and higher-quality collection and analysis techniques than one country could manage alone. Examples are countries that exchange intelligence from their signals intercept stations, reconnaissance aircraft, or satellites covering different areas of the world; that employ different means of collecting intelligence on the same target and sharing the results with one another; or that coordinate their networks of agents providing human intelligence so that they do not overlap. Cooperating states that agree to specialize in particular intelligence efforts can together generate more and better intelligence than would be possible if they each tried to provide adequate coverage of the same targets. This reasoning assumes that the intelligence resources developed for one target cannot be shifted at low cost to another target, which often is the case. For instance, a network of human intelligence agents providing information about one country or issue typically has little information about other countries or issues; listening posts for collecting signals intelligence are often aimed at particular targets and cannot be easily redirected to other targets; and analysts are trained in the language and history of one target and thus would require substantial retraining to address other targets.

SECRECY AND SHARING

Why might two states not share intelligence with each other? Often one country simply is not interested in another state’s intelligence. Costa Rica, for instance, might have little to gain from sharing intelligence with Nepal. In other cases, though, two states could benefit from sharing. During the Second World War, the United States shared a great deal of intelligence with its allies Britain and Canada, but much less with another ally, the Soviet Union. But both the United States and the Soviet Union could have benefited by sharing more intelligence on German tactics, battle plans, or troop movements.⁵ Why did they fail to realize these gains? Theories of international negotiation identify two general barriers to cooperation in such situations: the bargaining problem and the enforcement problem.⁶ The bargaining problem pertains to disputes over how the participating states will share the costs and benefits of a cooperative arrangement, such as an agreement to share intelligence.⁷ Governments may refuse to share intelligence if they cannot negotiate an agreement to apportion these costs and benefits acceptably. Then, once countries do agree on how to share intelligence, they must enforce the agreement. Some parties might have incentives to renege on their promises to share intelligence, and others must either prevent this or suffer the consequences.⁸

First consider the bargaining problem. Those states interested in sharing intelligence need to identify potential partners with useful intelligence, negotiate an agreement on what intelligence each will collect, how they will share it, whether they will offer any compensation such as military assistance or diplomatic assistance, and how much each state will invest in developing or redeploying intelligence resources to service the sharing agreement. Each of these issues provides benefits to one state and imposes costs on the other.

Countries may find it difficult to negotiate an agreement that divides these costs and benefits in a way that is acceptable to every participant. The first difficulty is the cost of negotiating. Each state must decide what intelligence it will send or wants to receive, the value of this intelligence, and the ability of its prospective partner to provide it. Each country will want the agreement to focus on its own priorities for intelligence collection and analysis. The participating countries also may have different preferences regarding quality control, human rights, and security standards. Countries with a strong tradition of protecting human rights, for example, might insist that their partners refrain from abusing these rights, and those without such protections might oppose such requirements as new and unnecessary burdens on their intelligence agencies. The participating states also have to construct a sharing arrangement that can withstand both anticipated and unforeseen contingencies, such as if the targets of the intelligence collection efforts unexpectedly change their behavior. Which state will pay which costs to set up and maintain the sharing agreement may also lead to disagreements. The intelligence-sharing agreement may try to benefit from specialization by calling for each country to concentrate on particular activities. Such specialization would require both countries to pay...

Table of contents

• Cover 
• Half title
• Title
• Copyright
• Contents 
• Acknowledgments
• 1. Understanding Intelligence Sharing: Curveball and KSM
• 2. Transatlantic Intelligence Sharing During the Cold War
• 3. Intelligence Sharing for Counterinsurgency: Vietnam and Colombia
• 4. Intelligence Sharing in the European Union: Institutions Are Not Enough
• 5. Intelligence Sharing and U.S. Counterterrorism Policy
• 6. Findings and Implications
• Notes
• Index