by James Igoe Walsh, (New York: Columbia University Press, 2009), 208 pp.
Reviewed by J.B. Webb
Usually I think about the risks of intelligence sharing as a problem of how reliable Washington’s partners are. But the massive Wikileak.com disclosures of classified US military documents on Afghanistan and Iraq and cables from the US Department of State naturally draws the questions “Might our partners think we are untrustworthy and will some of them scale back sharing?” Fortunately, James Walsh’s The International Politics of Intelligence Sharing offers a timely way to rethink what drives intelligence-sharing relationships and perhaps some reassurance. Walsh, a political scientist at North Carolina who humorously portrays himself as researching and teaching about “bad things,” offers an argument on intelligence sharing that stresses state interests rather than trust alone, suggesting the fallout may not be so bad.
Walsh sees intelligence as a commodity. States share out of mutual interest or to extract things like foreign aid and security assurances. (7) He argues that the secret nature of intelligence gives rise to two key problems. The “sellers” of intelligence can’t be sure that “buyers” will adequately protect what they receive, and “buyers” cannot be sure of the veracity of the intelligence they get from “sellers.” (13) To solve this classic cooperation problem, Walsh dips into social economic theory for answers. He applies relational contracting, a branch of transactional economics, to address the sticky bargaining and enforcement problems that come from the intelligence sharing dilemma of never knowing if your partner is going to double-cross you. (15-25)
Walsh crisply takes his readers through four hypotheses and finds mixed support for them. His hypotheses are:
Large gains are a necessary condition for intelligence sharing.
Intelligence is shared through anarchic institutions.
If one state is concerned that its partner will defect, it will seek to construct a “hierarchical” relationship.
Power imbalances are a necessary but not sufficient condition for creating such a hierarchy.
To test his hypotheses, Walsh draws on a range of historical US intelligence relationships, such as its transatlantic intelligence relationships with Britain, Germany, and France against the Soviet Union during the Cold War; US intelligence cooperation with South Vietnam against the Vietcong in the 1970s; Washington’s engagement of Bogotá in counternarcotics in the 1980s; and finally Washington’s current counterterrorism policies. The only non-US-related case examined is the development of intelligence sharing within the European Union in the 1990s.
In cases in which mutual interests are strong and the value of sharing intelligence is high, Walsh finds that states have little need for a highly structured relationship. The US-UK relationship during the Cold War is the shining example here. (38-43) Walsh uses Washington’s concern about the reliability of Paris during the Cold War to illustrate that when one partner deems intelligence sharing to be of little value, shared interests are weak, and where there are concerns that a partner might be unreliable, states forgo the risk of forming a working intelligence relationship. (47-48)
His most thought provoking analysis is in cases where state interest in sharing intelligence and worries about unreliable partners are both high, much like the US-Pakistan relationship as seen by some today. He emphasizes that “hierarchical control”—the case in which one state directly controls aspects of another state’s intelligence activities—is the key to hedging against suspect but needed partners.
It is the combination of interests and signaling of sustained commitment, I think, that are the most significant aspects of his argument. In the examples he uses of Germany during the Cold War and South Vietnam during the Vietnam War, Walsh shows that the United States was able to make these intelligence relationships work because it gained control over parts of its partners’ services. This was possible because the United States was a dominant power, had strong institutions, more resources for attacking the problems, and, most importantly, because Washington communicated a keen sense of the importance of fighting the Russians and the Vietcong, respectively, buttressing shared interests along the way (51–55; 72–78). In contrast, Walsh shows that US intelligence sharing with Colombia was significantly more limited because US shared interests with Bogotá were only moderate at best and because Washington provided fewer resources to fighting drug cartels as interest in combating the drug problem fluctuated. (79, 84–87)
Walsh’s argument that, by providing money, training, and other intangibles, states can lead other states to share intelligence against their competing interests is persuasive. And while I also agree with intelligence field officers who might argue that trust is far more important to sharing in the field, it is the degree of shared interests that truly matters in intelligence relationships between nations. That is why I am skeptical of Walsh’s assertion that states may be convinced to “specialize” in certain areas of intelligence (9, 19, 123–4) for the purpose of sharing. After all, intelligence services exist to help safeguard the state against a range of threats, and it is hard to imagine that policymakers would ever narrow their services’ focus to single national security interests.
Some readers undoubtedly will take issue with the theoretically broad approach Walsh takes. But where we should really focus debate is on his concept of “hierarchy” in intelligence relationships. Though the United States, for example, may be able to provide pledges of support that other states want in exchange for information, the fact remains that in some cases Washington’s has an extremely great need for certain information or certain skills to get information. When that is combined with the limited number of partners capable of “selling” that which Washington needs, the United States ends up not being the dominant player in the intelligence relationship, despite its resources.
The focus on interests and how to maximize shared ones, rather than mutual trust, is a key idea to keep in mind as the United States engages its intelligence partners in the wake of the Wikileaks debacle. I suspect our partners will want to know what we new things we will do to better safeguard secrets in the future. Underscoring and recommitting to the shared interests that drive our intelligence relationships, however, is probably more important, and something Washington could consider communicating to its partners as it cleans up from this mess.