Intelligence alliances are often highly secretive and consequently poorly understood. Some argue that no secret services are friends with one another and point to the large amount of recent economic espionage by supposed allies. Others claim that clandestine agencies cooperate too closely and cannot be trusted by the states that employ them. The attacks of September 11 blurred the boundaries between friends and enemies even more, as Western agencies have now teamed up with unlikely partners in order to pursue a vigorous war on terrorism. The campaign in Afghanistan, however, reminds us of the importance of intelligence alliances and shows that the real priority is finding a robust doctrine that allows some regulation of the diverse friendships required by the war. Far from constituting a world of unyielding competition, intelligence alliances can teach would-be practitioners of liberal institutionalism about structures that radiate trust, order, and devotion to the public good.

Intelligence alliances are among the most closely guarded secrets of clandestine agencies. Inside large alliances, these organizations behave remarkably like states, with their own treaties, embassies, and emissaries. Such intelligence groupings are more common than most realize, and most Western agencies enjoy treaty relations with dozens of foreign equivalents, setting out rights, permissions, and, most importantly, their place in the hierarchy of intelligence powers. The United States is at the top of this hierarchy, not only by virtue of its vast intelligence spending—over US$30 billion per year—but also because of its position as a global intelligence alliance manager.

Perspectives on the inner workings of intelligence alliances differ sharply. Some argue that clandestine agencies pursue national interests ruthlessly against friends and enemies alike. This secret statecraft school of thought is typified by the much-quoted adage: “There are no friendly secret services, only the secret services of friendly states.” Conversely, others argue that agencies of different states operate together like an international brotherhood, and that this clandestine kinship means they often owe more allegiance to each other than to the states they purport to serve. Whatever our perspective, the events of the last decade have created turbulence in the realm of intelligence alliances. The end of the Cold War left established allies with seemingly less need to cooperate after their common enemies had disappeared. Globalization and the quickening pace of technological change gave developed states more incentive to spy on each other in search of high-tech gains.

Public discussions of intelligence alliances have tended to focus upon the low politics of economic espionage between established allies, something that comes close to commercial theft. But the importance of this aspect has been exaggerated. In the last decade, real cooperation over issues such as nuclear proliferation has become increasingly important. Today, it is the war on terrorism and the exponential growth of US intelligence and military activities that pose urgent questions for the West. If the 20th century was characterized primarily by war, then the 21st century might well be one marked by terror. Terrorist organizations and the clandestine agencies of Western states operate in similar ways. In this new landscape of silent warfare, clandestine agencies may cease to be a supporting arm of defense and diplomacy, instead becoming themselves the cutting edge of foreign policy. As these agencies increasingly pursue global issues, intelligence alliances will become yet more important. What kinds of intelligence alliances will be needed in this new landscape and how will they be managed?

Hidden Partnerships

Established allies fight over policy-related intelligence, corroding trust and undermining possibilities for high-level policy convergence. One would hope that a sharing of intelligence among friends would promote similar outlooks. But in reality, even the closest intelligence allies are reluctant to accept finished analysis from one another, turning instead to raw data for fear that they will become victims of analytical spin. Whether allies ultimately receive raw data or mere analysis is a function of their place in the intelligence hierarchy. The ebb and flow of raw data can also be used as a weapon even against familiar partners, as when the United States famously cut the flow of intelligence to New Zealand in response to the latter’s withdrawal of facilities for US nuclear ships in the 1980s. Although this caused some surprise among observers, such arm-twisting among intelligence allies is not unusual. During the Yom Kippur War in 1973, British Prime Minister Edward Heath’s decision not to allow US intelligence gathering from British bases in Cyprus resulted in a temporary halt in the US signals intelligence tap. In these incidents, intelligence flow was used to underline sharp differences over policy.

More recently, Pakistan’s foreign intelligence agency, Inter-Services Intelligence (ISI), provides a fascinating example of policy divergence. What happens when a friendly regime, such as Pakistan, professes support for Western powers against the activities of Al Qaeda but its agencies do not? One of the complexities for intelligence officials is that a friendly secret service is rarely like one’s own. In Pakistan, the foreign ministry is a relatively ceremonial apparatus, while the ISI has secured control of many vital areas of Pakistani foreign policy, including Afghanistan and Kashmir. The ISI is now heavily factionalized, with some elements having assisted the remnants of Al Qaeda that fought in southern Afghanistan in February 2002. Inevitably, the number of US operations in Pakistan that are undeclared to their Pakistani hosts has risen. This situation is not unprecedented and results from the ISI’s continued close clandestine kinship with those that are no longer favored by the state.

This conflict has become more widespread in the months following the attack on the World Trade Center, as Western agencies enjoy the cooperation of a surprising range of allies. These include the Libyan intelligence services that the West has condemned for their association with terrorism in the 1990s and the Syrian intelligence services that played a major part in sponsoring terrorist groups in the 1980s. Both Libya and Syria were made pariahs, their agencies reviled in the West as no better than the terrorist groups with whom they have worked. But both countries have shown signs of diplomatic rehabilitation, in part facilitated by clandestine kinship, because Western contact with their agencies has always continued at some level. Other active US allies during the ongoing war against terrorism have included the Russians and the Chinese.

Clandestine kinship means that many of these seemingly new partners are, in reality, old friends. Despite periodic disputes over mutual penetration since the end of the Cold War, Russian agencies have worked well with both British and US agencies across substantial areas of mutual interest. Yet in 1990, respected US Central Intelligence Agency (CIA) veterans held divergent views as to whether this cooperation would even be possible. Former CIA Deputy Director for Intelligence Ray Cline warned that all that awaited would-be fraternizers with the East was entrapment or deception. By contrast, former CIA Director William Colby was upbeat about the possibilities and personally traveled to meet many of his former opponents, seeking to build bridges on matters of common interest. The dividends of cooperation were not long in coming. For example, during the Gulf War, the Russians provided Britain’s Secret Intelligence Service (SIS) with vital information on Iraqi President Saddam Hussein’s armory of Scud missiles and the nature of modifications that had been undertaken to enhance their performance.

Meanwhile, the United States has maintained a fruitful intelligence alliance with Beijing based on a mutual interest in Russian matters. In intelligence terms, China is also an old friend, with joint intelligence activities against Russia dating back to US President Richard Nixon’s normalization of relations in 1972. Some commentators have attributed the reluctance of successive US administrations to criticize human rights abuses in China to the continuing presence of two vital US-engineered listening stations inside China that are focused on Russia. The Sino-US intelligence relationship has grown recently owing to Beijing’s concerns with Islamic militancy in its western provinces. Against this background, the recent incident involving the Chinese detention of the US EP-3 spy plane on Hainan Island might be seen as a brief interruption of an awkward but nonetheless time-honored intelligence alliance.

Calculated Risks

Established intelligence alliances have never been severed because of economic espionage or intellectual property theft. But in 1995 some intelligence relationships were broken off when the CIA had to back away from unsavory sources and dubious services with records of murder and torture. That year, US lawyer Jennifer Harbury, campaigning for information on the fate of her dead husband, a captured Nicaraguan guerrilla, was able to show that his fate had been determined by a senior Nicaraguan military official who had twice been a student at the US Army School of the Americas. The official also turned out to be a CIA source who had received payments from the agency. After this case, new draconian guidelines were developed requiring scrutiny of all intelligence assets. These restrictions became known as the scrub order.

The CIA was required to cease contact with individuals who had a record of violence in their past. James Woolsey, whose term as director of central intelligence had ended shortly before the scrub order was introduced, worried that this would prevent intelligence gathering in those areas where it was needed most and protested that “political correctness and fighting terrorism often don’t work well together.” In 2000, the US Congress commissioned a report on counter-terrorism from diplomat L. Paul Bremer. His report, which presciently featured the World Trade Center on the cover, warned explicitly about the dangers of catastrophic terrorism striking the United States. It urged that the scrub order be abandoned because it effectively outlawed the recruitment of many people with inside information on terrorism. But Congress refused and the administration’s general message seemed to be that such risk-taking was not acceptable and perhaps no longer necessary.

The scrub order has now effectively been scrapped and these short-lived attempts to regulate the nature of clandestine kinship all but abandoned. Numerous other restrictions on operational activity have been relaxed or ignored. Relationships with agencies that were considered unsuitable in the mid-1990s no longer present a problem. In the words of US President George Bush on September 17, 2001, the war waged by terrorists is intrinsically unpleasant and the result is a contest in which “there’s no rules.”

Assertions that it is now time to throw away the rulebook and adopt the dirty tactics of one’s adversary have a deep resonance for historians of the CIA. In 1954, Major General James Doolittle, famous leader of the 1942 air raid on Tokyo, prepared a National Security Council study for US President Dwight Eisenhower arguing that the United States needed to abandon restraints in covert warfare in response to Soviet actions. By the 1960s, this road had led to 13 failed attempts to assassinate Fidel Castro using a range of unsavory allies, including the US mafia. In the 1970s the pendulum swung the other way, as the hearings of the US Senate’s Church Committee began to expose these activities. Agencies became cautious and restraints were imposed. Hugh Tovar, the CIA’s one-time covert action chief, lamented that his area was a “dying art form.” But the pendulum continued to oscillate, and by 1980 Ronald Reagan became US president after campaigning that it was time to “unleash” the CIA. Iran-Contra was just around the corner.

Some oscillation is inevitable, but violent oscillation is damaging and demoralizing for the agencies who rightly see themselves as victims of political fashion. A standard that recognizes the importance of the information at stake is needed to cover recruitment of dangerous sources and relations with dubious agencies. It must be straightforward enough to offer practical guidance to intelligence officers recruiting sources on the ground. It also needs to be sophisticated enough to allow agencies and governments to make reasoned judgements about just what kinds of threats require what sort of risky relationship. This is not too much to ask, since calculated risk-taking is the core business of agencies; the underlying logic will now have to become more explicit. The much derided scrub order of 1995 was intended to have some of this proportionality but proved too cumbersome. On the ground, the distinctions were lost and the general effect was to chill the recruitment of all risky and violent sources.

The real radical shift required here lies in the area of congressional or parliamentary oversight rather than regulation. Traditionally, intelligence alliances were a black box into which oversight bodies have rarely been allowed to look. Resistance to such oversight has a long history. In 1946 British signals intelligence personnel were forbidden to travel to the United States to give evidence to a US congressional inquiry into Pearl Harbor. In 1975 Britain’s SIS and the British Government Communications Headquarters were appalled by the Church Committee hearings and sought assurances that this inquiry would not become multinational in scope. But in a global era, when clandestine agencies rarely work alone on large issues, the near invisibility of liaison arrangements to oversight by elected representatives is problematic. Oversight mechanisms have not kept pace with global issues. This shortcoming has recently been underlined by a Dutch government inquiry into intelligence and the dismal events at Srebrenica in 1995, for while national aspects of that inquiry have gone well, the alliance dimension has been difficult. Will we eventually need multinational oversight teams to look at multinational operations? Doubtless, the agencies will shake their heads, but the problem will still have to be faced.

US Dominance

The United States is the world hub of intelligence alliances. Accordingly, alliances cannot be separated from the business of reforming or reshaping the US intelligence community. Cynics have observed that when it comes to US intelligence, only a policy of enlargement does not encounter immediate resistance. This observation rings true because the US intelligence community has some unusual arrangements and duplications. The CIA has its own special forces while the US military has its own mini-CIA buried within the Department of Defense, conducting clandestine human intelligence in competition with the CIA.

Similar to the British and the French, many characteristics of the US intelligence community are perennial. Historically, it has been strong in core technical areas such as signals intelligence and imagery. It has an impressive record in covert action and regime support. Yet it has been weaker in the area of classical human intelligence gathering. There are many reasons for this weakness, some having to do with the growing dominance of the Department of Defense as a player in Washington, demanding a high-tech intelligence capability that is designed to help deliver full-spectrum dominance on the battlefield. It is partly explained by the growth of a secretive presidential foreign policy, which has required substantial covert action that can be hidden not only from foreign enemies but also from the public. It also reflects the CIA’s justifiable pride in its analytical prowess.

As a result, despite its size, the United States finds itself dependent to some degree on intelligence alliances with its middle power allies. It produces more intelligence but it also has a vast appetite. It is a massive exporter of technical intelligence while it is also surprisingly dependent on friends for certain kinds of espionage. The work of the Israeli Mossad and the French Directorate-General of External Security in Africa, for example, has been essential in providing Washington with a window on affairs within that continent. Some argue that allies have been “overcharging” the United States for this human intelligence, but the overall result has been a mutual dependence that is healthy and that ensures a greater reservoir of unique skills in the service of Western policy.

Recent operations in Afghanistan have offered a good illustration of the benefits of intelligence alliances. This is a country in which a large number of British intelligence officers have served in the last two decades, either during the war against Soviet occupation or subsequently in counter-narcotics operations. In late 2001, as the military campaign began, Britain was able to field substantial numbers of SIS officers with detailed knowledge of the region, its leaders, and its languages. On the ground, the contrasting British and US approaches to secret work proved to be complementary. While CIA officers emphasized locating and eliminating Al Qaeda units, SIS focused on political operations that would build regime solidarity and cohesion in an environment where local politics could be rather fissiparous. Using traditional and time-honored methods, allied forces persuaded Afghan warlords to put their shoulders to the wheel, helping turn the corner in a difficult campaign.

In the next decade, the United States may decide to increase spending on human espionage, but the need for what some have called “professional bargains” in this area will persist. The picture might be different in the areas of technical and military intelligence. In the recent campaign in Afghanistan a range of new capabilities, many of them information-led, helped to deliver success in a conflict that some confidently asserted would be almost unwinnable. This echoes the Gulf War, a historic victory with remarkably few casualties in which the effectiveness of information in warfare became apparent. What does this mean for intelligence alliances on the battlefield? As Paul Kennedy has noted, the United States will soon account for more than 40 percent of world defense spending. The problem here is not quantitative but qualitative. By 2020, the information-led armies of the United States may no longer be inter-operable with the rather antiquarian forces of even its closest partners. This may not matter much in military terms, but where world politics requires coalition warfare, this may prove to be a significant problem. One major challenge for intelligence alliances is to maintain Western cohesion in the area of high-tech battlefields. This will require NATO to make a special effort to stay allied on cutting edge military projects.

Understanding Alliances

Do clandestine agencies follow national interest in a ruthless manner, or do they constitute a global network operating together in a way that is at best semi-detached from the states they purport to serve? The diverse menagerie of intelligence communities contains some odd creatures and inevitably both perspectives contain some truth. But in another sense, both are misleading insofar as they emphasize the “agency” of the organizations and portray them as the blameworthy controllers of events. For all their talk of getting productive, there is something to be said for viewing clandestine agencies and their secret friendships as the product of structural change, most obviously of globalization. One might even argue that the modern intelligence alliance, originated a century ago in state efforts to pursue agitators, anarchists, and terrorists moving internationally, was one of the first results of globalization.

Perhaps clandestine agencies and their intelligence alliances should be viewed less as exponents of realism and more as the smooth and experienced exemplars of liberal institutionalism. Not only were they among the first to accept that values, ideas, and knowledge can sway events, they have also been required to mediate national interests using cooperation and trust inculcated through a vast institutionalized network of information exchange. This complex web of unseen agreements and networks arguably raises expectations about cooperation and regulates some rather awkward practices by radiating established norms and conventions. By definition, liberal institutions operate indirectly, and therefore somewhat imperfectly. Perhaps the world of intelligence alliances constitutes a place where ideas and knowledge have real power and where cooperative exchange has always been viewed as a public good.