America needs Pakistan more than Pakistan needs America. Pakistan is the pivot of the world, as we are placed, the frontier on which the future position of the world revolves.
— Mohammad Ali Jinnah, first governor general of Pakistan, 19471
How does a state judge how much leverage it holds in an alliance? Alliance theory literature tells us that bargaining power is a function of dependence and having alternatives, rather than of raw power. The state that needs what its prospective ally has to offer the most and has the fewest attractive alternatives is assumed to have less leverage within the relationship.2 States can strategize to expand that leverage somewhat, for example by downplaying their need for the alliance or exaggerating other options available to them. Small states can exaggerate their vulnerabilities, insisting they need a great power’s support if they are to survive.3 Skill in bargaining matters too. Successful bargaining, after all, involves not only knowing what one wants but understanding the other party’s need for and alternatives to what you have to offer. This suggests that a state’s ability to draw maximum benefit from an alliance rests not only on “alternatives, commitment, and control,” but also on access to intelligence about the other party’s internal discussions.4 This includes the costs and benefits that a leader expects to receive from entering into a pact, not only in terms of national security, but also for his or her own domestic position, as well as how domestic politics constrain available options.
Alliance literature has largely focused on formal, publicly acknowledged agreements. Many works, such as Mira Rapp-Hooper’s excellent Shields of the Republic,5 limit their dataset further, examining only alliances that are based on mutual defense treaties. And yet, there is another kind of alliance that also shapes U.S. foreign policy: namely the secret national security agreements that underpin clandestine intelligence operations abroad. Such agreements can involve substantial U.S. commitments but may have few or no written underpinnings. For example, after the December 1979 Soviet invasion of Afghanistan, President Jimmy Carter committed the United States to channel its support to the Afghan resistance through Pakistan’s military Inter Services Intelligence Directorate during a telephone call with President Mohammad Zia-ul-Haq. Former CIA Afghan program chief Jack Devine confirms that even into the late 1980s, the program still ran without any formal, written agreements.6 Nonetheless, the United States spent some $20 billion on the clandestine program itself while Pakistan went from being a sanctioned state receiving minimal U.S. aid to receiving some $5 billion during the 1980s, making it the third-largest recipient of U.S. assistance.7 While the scale of the money and influence involved in the program is unusual, the informal nature of the agreement is not. The CIA ran its support for Angolan opposition forces from Kabila Airbase in what was then Zaire for more than six years based on a nod from Mobutu Sesu Seko.8 Sobukwe Odinga, who reviewed 13 U.S. strategic security partnerships in Africa created between the start of the Cold War and today, found that these partnerships tended to be “more informal and sustained by commitments that are less clear-cut.”9 Senior retired CIA officers confirm that, in their experience, handshake accords with local intelligence services were the norm throughout their decades-long careers.10
If these secret pacts, ranging from service-to-service intelligence sharing to base concessions and joint counter-terrorism operations, involve agreements between nations and potentially substantial funds, should they not also be considered a form of alliance? And if they are, how does their context shape negotiations and calculations of leverage? I argue that the secret and informal nature of these agreements lends itself to a more volatile relationship overall. Small states may gain disproportionate rewards for participating in these types of accords, since U.S. leaders are less likely to have to defend publicly any concessions they have to make. At the same time, the agreements are more vulnerable to changes in political circumstances, leaders’ whims, or miscalculations. Leaders are more prone to disagree over the terms of the accord as the alliance enters the bargaining phase, given the absence of any anchoring or mutually accepted documentation. Meanwhile, the smaller circle of advisers and the absence of public discussion involved in secret deals increases the likelihood that leaders will make their decisions based on limited and imperfect information.
In order to better understand how bargaining theory applies to clandestine security relationships, I analyze the dynamics of a set of secret U.S.-Pakistani national security agreements from their creation in the 1950s through their collapse a decade later. The relationship began with great fanfare and high hopes on both sides but ended with each accusing the other of betrayal. Between the mid-1950s and 1965, the United States and Pakistan each benefitted from the secret collaboration. Washington operated several intelligence collection sites in Pakistan that gathered imagery, telemetry, air samples, and other intelligence on the Soviet and Chinese nuclear weapons programs as well as Soviet missile sites and Sputnik launches. Pakistan, meanwhile, received roughly $1 billion in U.S. military aid during this period, along with an additional $500,000 in economic support. This mutually profitable relationship proved unsustainable, however, as each side came to believe that the other was taking unreasonable advantage of the arrangement. This case suggests that a recognition of clandestine equities and the power imbalances inherent in many secret deals may prove useful in deepening our overall understanding of alliances among nations.
I chose this case because the events took place far enough in the past that a substantial body of declassified records is available and accessible. These include discussions between U.S. presidents and their advisers about America’s secret operations in Pakistan and how those operations played into U.S. Cold War strategy. There are also records of the debates within the White House, the National Security Council, and the intelligence community over Washington’s negotiating strategy with Pakistan. These allow me to track over time how U.S. leaders viewed the importance of the intelligence collection sites on Pakistani soil, as well as what they were willing to do to maintain access to these sites. To understand the Pakistani side of the discussion, I draw upon a number of Pakistani officials’ memoirs and diaries, while recognizing that these are invariably shaped by the authors’ hindsight and drive to burnish their legacies. When possible, I compare memoirists’ accounts of events with contemporary press accounts and U.S. diplomats’ real-time descriptions of exchanges to offset this.
This article’s account of U.S.-Pakistani secret bargaining proceeds as follows. First, I outline how the secret nature of clandestine pacts may enable small states to exercise greater leverage than that usually seen in international alliance bargaining. Second, I detail how Pakistan wooed the United States and, in the late 1950s, eagerly signed both a secret lease agreement for a major U.S. intelligence collection facility and multiple mutual aid pacts. Third, I describe how U.S. and Pakistani leaders maneuvered for greater leverage within their clandestine relationship and how these efforts fueled mutual mistrust. Finally, I explore the lessons that the collapse of the intelligence base agreement offers for our understanding of how national leaders determine their bargaining leverage within clandestine relationships.
Secret Intelligence Pacts and Bilateral Relations
Before looking at the specific case of the United States and Pakistan, it is important to first better understand intelligence-sharing pacts — and specifically secret intelligence pacts. This includes how these clandestine deals can allow foreign leaders to influence U.S. policymakers, how they often give U.S. intelligence partners significant and disproportionate bargaining leverage, and the dangers of making such agreements in secret. Secret alliances largely follow the same patterns explored in works such as Glenn Snyder’s Alliance Politics.11 States calculate the costs and benefits of entering an alliance and bargain to ensure that they draw maximum benefit from the resulting pact. Negotiations continue after alliances are concluded as nations pursue their interests under constantly changing conditions. A key factor in this continued maneuvering is the balance of dependence. All things being equal, the state that is less dependent upon the other tends to have the upper hand since it can always threaten to abandon the relationship altogether.
These same patterns hold true in clandestine relationships, but with some differences. Intelligence sharing gives foreign states new ways to influence U.S. policymakers. While often limited to sharing information, intelligence-liaison relationships between intelligence services can also include joint collection operations, training, clandestine operations against a third party, and secret facility leases. Liaison programs can offer foreign leaders the chance to bypass or reinforce traditional diplomatic channels by giving them direct connections to the White House. This expanded influence can go both ways. For example, CIA chiefs of station operating large liaison programs can sometimes build closer relationships with the local head of state than the serving U.S. ambassador can.12 They then use these relationships to explain U.S. policy and press for U.S. interests. In general, though, liaison programs increase Washington’s stake in its clandestine ally, creating an incentive to give more to preserve the relationship. Groups within the Washington policymaking bureaucracy often develop into strong internal lobbyists for delivering what the ally wants in order to preserve the liaison relationship.13
All things being equal, the state that is less dependent upon the other tends to have the upper hand since it can always threaten to abandon the relationship altogether.
For example, CIA Director Richard Helms repeatedly lobbied President Richard Nixon in support of the Shah of Iran’s weapons requests, arguing that U.S. listening posts in Iran were “absolutely essential” and irreplaceable.14 Helms conceded that the Shah’s requests might be excessive but highlighted that U.S. intelligence installations in Iran were “entirely dependent upon the continuing willingness of the Shah to permit them to operate.” Nixon approved the Shah’s request despite the opposition of the Departments of Defense and State. Nixon’s decision fits a larger pattern. When U.S. presidents take a personal interest in a clandestine program, this access is magnified. For example, President Ronald Reagan’s personal engagement in the CIA’s program to oust the Soviets from Afghanistan allowed Zia to make end runs around the U.S. ambassador and Department of State when they denied his requests for aid concessions or new weapons systems.15 Zia often wrote to Reagan directly, pressing for him to expand U.S. aid in return for Islamabad’s clandestine support in Afghanistan.16
A U.S. intelligence partner also can gain disproportionate bargaining leverage when Washington is dependent on the partnership, but the partner is not. When both liaison partners are equally interested in the goal of the partnership, collaboration stands to strengthen the overall alliance relationship. The United States and United Kingdom, for example, worked together to collect information on the Soviet Politburo without needing to convince each other of the merits of the program’s overall goal. When intelligence operations reflect the interests of only one party, however, it creates an institutionalized dependency that gives the less-engaged partner significant leverage, as we will see in the U.S.-Pakistani case study below.17 The state most able to walk away from an alliance holds the upper hand in bargaining, whether that alliance is acknowledged or secret.
Jennifer Sims terms these unbalanced secret relationships “asymmetric liaisons” and argues that they foster an adversarial dynamic. Unlike classic alliance theory, which focuses on asymmetry in national power, what matters in secret relationships is the balance of interests. The liaison partner least dependent upon the clandestine program can hold that program hostage to side payments.18 Such partners can demand attention or demonstrate their displeasure with U.S. policies by slowing intelligence sharing, denying access to key sites or assets, leaking sensitive information, or otherwise hindering intelligence operations. They also can walk away from the liaison agreement, usually without any public awareness that the relationship ever existed. Odinga found U.S. clandestine liaisons with African intelligence partners to be characterized by constant struggles for control, with “horse-trading” and quid pro quos “endemic” to the collaboration.19 Eli Berman and David Lake’s review of principals and proxies engaged in clandestine operations similarly finds that the more divergent the interests between the principal and the proxy, the greater the rewards needed to retain the proxy’s engagement.20
Military and economic aid, pledges of mutual defense, and other non-intelligence concessions are thus built into asymmetric liaison agreements from the start. Just as overt bilateral alliances allow for continued maneuvering over the life of the alliance, clandestine bargains also face continual renegotiation. Indeed, Odinga makes a persuasive case that clandestine alliances may be more subject to strong-arming.21 Liaison agreements stemming from rushed ad hoc responses to immediate crises, such as the Sept. 11, 2001 terrorist attacks, may produce the most leverage for small states. The comparative lack of documentation and frequent adjustments to changing conditions in the field make it easy for states to demand new concessions just when the more dependent country’s need is the greatest.
Meanwhile, these secret pacts also receive less scrutiny, which can foster a lack of perspective. Regional and topical experts may not have the clearances and access to weigh in, while the public is excluded altogether. Indeed, the lack of scrutiny is one of the attractions of entering into an informal agreement. For example, the CIA chose not to formalize its agreement with Mobutu to use Kamina Airbase in order to avoid congressional hearings and possible leaks.22 One danger is that excluding knowledge of a secret agreement from all but a small inner circle would, presumably, increase the risk of negotiators misjudging both their own and the other party’s alternatives. If that is the case, all else being equal, we should expect to see secret agreements prove less stable than those that are openly acknowledged. Such agreements would likely involve more frequent confrontation as participant states struggle to establish leverage.
Given individual leaders’ personal association with specific clandestine pacts and programs, there is also the risk that a new leader may demand new terms or even end cooperation as governments change hands. This dynamic can foster uncertainty among U.S. policymakers, as well as Washington’s partners. Indeed, this is exactly what was at play in the intelligence-sharing relationship between the United States and Pakistan in the 1950s and 60s. President Mohammad Ayub Khan governed the country throughout America’s decade-long lease of Badaber but dealt with three different U.S. presidents. His conviction that the presidents who came after President Dwight D. Eisenhower did not see maintaining ties with Pakistan as important played a significant role in the downward spiral of the two states’ clandestine security ties.
The Development of Relations Between the United States and Pakistan
The trajectory of U.S.-Pakistani relations illustrates how shifting geopolitical conditions can reshape alliance possibilities as well as the balance of leverage between states. When Pakistan first gained independence on Aug. 14, 1947, it was Pakistan that was the suitor. Pakistan’s founding leader, Mohammad Ali Jinnah, expected Washington to be eager to support his fledgling state. Jinnah presumed that Washington needed Pakistan. With the United Kingdom no longer securing the subcontinent, the United States would need Pakistan to stand as a bulwark against communism in the region. “Pakistan is the pivot of the world, as we are placed … . If Russia walks in here, the whole world is menaced,” Jinnah told a journalist shortly after Pakistani independence. If the United States felt the need to arm nations such as Greece and Turkey, surely it would also want to “pour weapons into Pakistan,” given its proximity to the Soviet Union’s southern borders.23 At the same time, Pakistan desperately needed outside support. The new state had virtually no industry — its historic markets lay in India, now a hostile neighbor. Pakistan had inherited only 30 percent of the Raj-era Indian Army and 15 percent of the revenue streams, along with unanticipated millions of refugees.24 Meanwhile, convinced that New Delhi meant to reconquer its lost territory, Pakistani leaders spent roughly 75 percent of their budget on national defense.25
Contrary to Jinnah’s hopes, his urgent requests for arms only convinced the U.S. Department of Defense that Pakistan hoped to hand off responsibility for its defense to Washington. President Harry Truman had no interest in taking on that responsibility, particularly when State Department experts warned that it would alienate the much larger and more important India. Truman’s administration also rejected a separate request from Pakistan for a $2 billion loan, recommending it instead approach the Import-Export Bank.26 Despite these rejections, Pakistani leaders continued to press for support. “The Pakistanis were always asking us for arms,” Secretary of State Dean Acheson remembered, “and I was always holding them off.”27
U.S. leaders dismissed Jinnah’s warnings that Pakistan was a communist target. In a May 1950 briefing book in advance of Prime Minister Liaquat Ali Khan’s visit to Washington, Truman’s advisers summed up the administration’s view: “The entire South Asian region is of relatively secondary importance to the United States from a military view.”28 If anything, the Truman administration’s limited interest in the region lay in resolving Indo-Pakistani tensions over Kashmir to stabilize the region, not in fueling an arms race.
Contrary to Jinnah’s hopes, his urgent requests for arms only convinced the U.S. Department of Defense that Pakistan hoped to hand off responsibility for its defense to Washington.
Meanwhile, popular opinion in Pakistan remained skeptical of America’s intentions. Even as Jinnah and Liaquat pressed for U.S. support, other prominent Pakistani politicians cited earlier American opposition to the partition of India and Pakistan as evidence that Washington was hostile to their interests. Roosevelt had, after all, compared the idea of carving out a separate Muslim state to the U.S. Civil War, while Truman’s envoys pressed the Muslim League to cooperate with India’s leading pro-independence group, Jawaharlal Nehru’s Congress party. Although the United States recognized Pakistani independence once it was a fait accompli and was even the first nation to post an ambassador to Karachi, it continued to face popular suspicion. The overtures toward the United States represented a break from the ruling Muslim League’s pre-independence lean toward non-alignment. Sir Morrice James, a British diplomat posted to Karachi at the time, recalled that the majority of his contacts continued to favor non-alignment, viewing the East-West struggle as irrelevant to their country’s needs.29 Truman’s support for the creation of Israel redoubled accusations that the United States was imperialist and anti-Muslim. Jinnah wrote Truman a public letter denouncing the president’s “morally untenable” support for the new Jewish state.30 When Israel declared independence on May 14, 1948, thousands of Pakistanis blockaded the U.S. embassy in Karachi in protest.
The Eisenhower administration reassessed the importance of Pakistan to U.S. interests after war broke out in Korea. The National Security Council decided, in a January 1951 meeting, that the United States needed urgently to develop access to military facilities in the region, while denying South Asian resources to Moscow.31 This justified providing economic and military aid to both Pakistan and India. While the National Security Council envisioned a policy that tilted somewhat toward India, Secretary of State John Foster Dulles had other ideas. In June 1951, Dulles briefed the Senate Committee on Foreign Relations, telling its members that the administration saw Turkey and Pakistan as potential “strong points” in its “northern-tier strategy” to contain the Soviet Union. “The trouble with Pakistan at the moment,” he noted, “is that we do not have any program of military aid for Pakistan, because we don’t dare to do it because of the repercussions on India.”32 The next year, Defense Department officials began exploring a “relatively small-scale” military aid program. Maj. Gen. George Olmsted, then director of the Office of Military Assistance, argued the investment would “intensify” Pakistan’s determination to “fight the Soviet.”33
Dulles grew yet more certain that Washington’s interests lay with Pakistan during a May 1953 visit to South Asia. He cabled Washington to praise his Pakistani hosts, writing that he had the “feeling Pakistan is one country that has [the] moral courage to do its part in resisting communism.”34 The CIA (wrongly) predicted the next month that Pakistan would eventually acquiesce to Indian control of Kashmir, removing a key destabilizing factor in regional politics. U.S. intelligence analysts assessed that Pakistan would likely agree to host strategic U.S. airbases, but only in return for “substantial military and economic aid and Western Security guarantees.”35 Before Indian and Pakistani independence in 1947, the United Kingdom had used its Indian troops to fight throughout the Middle East. Dulles and Defense Department officials envisioned using Pakistani troops to similarly maintain Middle East peace, put down communist insurgencies, and slow a Soviet invasion southward. The National Security Council concurred, agreeing in 1954 that it was U.S. policy to “[s]eek to insure that in the event of general war Pakistan will make available manpower, resources and strategic facilities for mutual defense efforts with the West.”36
The United States and Pakistan signed their first Mutual Defense Assistance Agreement on May 19, 1954. The State Department’s announcement of the agreement noted Pakistan’s “desire to play a part in the collective defense of the free world.”37 Other agreements soon followed. In September 1954, Pakistan joined the Southeast Asia Treaty Organization. “Pakistan today enters what promises to be a glorious chapter in its history,” Pakistani Prime Minister Mohammad Ali Bogra announced. “United States military aid will enable Pakistan to achieve adequate defensive strength without the country having to assume an otherwise increasing burden on its economy.”38 In 1955, Pakistan joined the Baghdad Pact, later renamed the Central Treaty Organization. Dulles had envisioned this pact as the vehicle for accessing Pakistani troops to defend against communist incursions in the Middle East. Ayub, who participated in the negotiations as Pakistan’s chief of the army staff, later admitted in his memoirs that “the objectives that the Western powers wanted the Baghdad Pact to serve were quite different from the objectives we had in mind.” Ayub’s claims that Pakistan had “never made any secret of [its] intentions or [its] interests” contradict his colleagues’ claims that he had coached them to avoid mentioning India to official U.S. visitors, instead emphasizing their determination to stand against communism.39 Ayub was unconcerned about the long-term risks of the deception, writing to Bogra that he felt “confident” that “there is no danger of the Americans interfering with our affairs or … forcing their opinions on us.”40
As the only member of both pacts, Pakistan was crucial in connecting the two organizations. In the early years of the agreements, Pakistan also offered an example of a pro-America ally in the Third World that was receiving extensive development support — a symbolic challenge to the growing Non-Aligned Movement. Ayub publicly declared Pakistan “the most allied of allies.”41 During the Eisenhower presidency, Pakistan provided a reliable vote in the United Nations for American initiatives. Pakistan’s neighbors, India and Afghanistan, meanwhile, traded high-level visits and signed economic deals with Moscow — increasing the perceived value of the Pakistani relationship. Pakistani leaders, too, gained not only an important source of aid but also what they hoped would be a superpower protector. They also hoped that their alliance with the United States would, at the very least, discourage Washington from building ties with New Delhi and, ideally, cause them to actively support Pakistan with regard to India.
Nonetheless, neither the United States nor Pakistan were satisfied with the results of the alliance. While U.S. planners considered their military aid to be substantial, it fell far short of Pakistan’s expectations. Ayub professed himself “dejected” and “broken hearted” when an initial U.S. commitment in 1954 turned out to be $30 million in military equipment and training, rather than the $200–$300 million in unallocated funds that he had envisioned.42 U.S. aid to Pakistan increased to $171 million by 1955, but Ayub again denounced the amount as inadequate to U.S. diplomats, journalists, and visiting congressmen. The Eisenhower administration once again reluctantly conceded, offering to support five and a half Pakistani divisions “regardless of cost.”43 Eisenhower was less willing, however, to be used as a lever against India. While he and Dulles abhorred Nehru’s attempts to build a movement of non-aligned states — Dulles thought non-alignment “immoral” — both men considered India an important bulwark against the spread of communism in Asia.44 Eisenhower had, in fact, originally hoped to extend his 1959 agreement with Pakistan to include India, aiming to create a joint India-Pakistan defense against communist aggression.45
Pakistan’s value to Washington lay not just in its public role as a U.S. ally, but as a clandestine collection and operational platform against the Soviet Union and China.
Meanwhile, Pakistani leaders sought to use the country’s membership in the Southeast Asia Treaty Organization and the Central Treaty Organization to reinforce its position vis-à-vis India. Barely a month after signing the September 1954 Manila Pact that created the Southeast Asia Treaty Organization, Bogra pressed Dulles to include aggression from India — a noncommunist state — under the alliance’s mutual protection clause.46 Dulles reiterated that the mutual defense pact was limited to attacks by communist states, but Bogra and his successors persisted. In 1958, Prime Minister Feroz Khan Noon threatened to withdraw from both of the treaty organizations and “embrace Communism” unless the United States met Pakistan’s “hopes re Kashmir and more foreign aid.”47 The next year, Noon announced to cheers from the National Assembly that Pakistan “will break all pacts” and “shake hands with those whom we have made our enemies for the sake of others,” if threatened by India.48 “The West,” he added, “was very much mistaken in assuming that the Muslims cannot reconcile themselves to Communism.”
Ultimately, Eisenhower acceded to most of Ayub’s demands for money and weaponry. Pakistan’s value to Washington lay not just in its public role as a U.S. ally, but as a clandestine collection and operational platform against the Soviet Union and China. Indeed, the diplomatic and clandestine negotiations appear to have been closely intertwined. For example, during Ayub’s visit to Washington in May 1953 to press for U.S. military aid, he wrote in his diary that, after meeting with the Joint Chiefs of Staff over lunch, he “met [Director of Central Intelligence] Allen Dulles and thanked him for putting his brother, who was the U.S. Foreign Secretary, in the picture regarding our problems.”49
Ayub’s visit came as Washington was wrestling with a critical intelligence gap. Most available imagery of Soviet military facilities south of the Urals dated to World War II, long before contemporary nuclear and missile development programs. U.S. leaders needed better intelligence to assess and respond to the Soviet threat. Allen Dulles later described the situation to the Senate Committee on Foreign Relations: “Great areas of the USSR are curtained off to the outside world … . Their military hardware, ballistic missiles, bombers, nuclear weapons, and submarine forces, as far as physically possible, are concealed from us.”50 Eisenhower appointed a committee to find ways to minimize the risk of a surprise Soviet nuclear attack. The members wrote in their report that intelligence was the key. “We must find ways,” their report warned, “to increase the number of hard facts upon which our intelligence estimates are based, to provide better strategic warning … and to reduce the danger of gross overestimation or gross under-estimation of the threat.”51
Pakistan offered a unique location from which to collect precisely the type of intelligence that Washington most needed. The United States was already subsidizing a British seismic collection station north of Peshawar, code-named “Stowage.” This was the West’s closest collection station to the Soviet Semipalatinsk nuclear test site and was considered the most important post in the United Kingdom’s global collection network.52 The Soviet Union had built many of its key nuclear and missile test sites in Central Asia — far from NATO countries or bases in the Far East, but within reach of Pakistani airbases. The CIA officers working with Lockheed to develop the U-2 aircraft believed that the Soviet Union had yet to extend its air defense systems to the Pakistani-Afghan air corridors. A survey team identified Badaber airfield, outside Peshawar, as offering access to Soviet nuclear and missile test sites south of the Urals. In addition, Pakistan also provided a platform for clandestine operations on the Tibetan Plateau and access to Beijing’s first nuclear test site, Lop Nur, then under construction in Xinjiang Province.
The United States and Pakistan signed their 1954 bilateral mutual defense agreements even as intelligence officials were negotiating with Pakistan’s Inter Services Intelligence Directorate for permission to collect intelligence on Soviet and Chinese targets from Pakistani soil. In October 1954, the Department of State passed Pakistani representatives a secret aide memoire, committing to spend $171 million over the next three and a half years as well as offering more than $105 million in economic aid.53 A separate private agreement also allowed the United States to install seismic monitors for nuclear tests, early warning radar systems, and signals intelligence collection facilities. The next year, a U.S. Air Force survey team chose Peshawar as an ideal location for a radio intercept site. During a state visit to Washington in July 1957, Pakistani Prime Minister Shaheed Suhrawardy agreed, in principle, to Eisenhower’s request that Pakistan allow the United States to expand collection operations to include U-2 flights and take charge of Badaber.54
In August 1957, the United States launched its first U-2 flights from Pakistan. Over a 23-day period in August 1957, the CIA launched seven missions over Soviet and Chinese territory. These missions proved Pakistan’s value as a base of operations. U-2 pilots pinpointed the previously unknown locations of the Soviet Tyuratam intercontinental ballistic missile and Saryshagan anti-ballistic missile test sites. On another flight, the pilot photographed the Semipalatinsk proving grounds for four hours before a half-megaton nuclear device was detonated. Two other flights collected information on Lop Nur.55 Altogether, the seven flights collected what an internal CIA report described as “a bonanza of information that kept scores of photo interpreters busy for more than a year.”56 Among other things, this information helped the CIA to prove that there was no missile gap and that Soviet Premier Nikita Khrushchev had greatly exaggerated Soviet advances in intercontinental ballistic missile technology.57
Later in 1957, Allen Dulles gained permission — apparently also unwritten — from Pakistan’s intelligence service to use an abandoned British airfield outside Dhaka to expand America’s support to the Tibetan resistance.58 According to Kenneth Conboy and James Morrison, the chief of Karachi Station simply asked then-President Iskandar Mizra if, “hypothetically speaking,” the CIA could exfiltrate Tibetan fighters into East Pakistan for transport onwards for training.59 Pakistani officers met groups of Tibetan volunteers at the northern border and escorted them to the site, known as Kurmitula.60 The CIA then whisked the Tibetans off to training sites in Colorado, Okinawa, and elsewhere, before eventually air-dropping them back into Tibet.61 These CIA-trained fighters captured valuable intelligence on Chinese military operations while increasing the cost to Beijing of consolidating its hold on Tibet.62 The CIA also used Kurmitula from 1957 to 1959 as a base for U-2 flights over China, especially over Tibetan territory.63
The U.S. Air Force, CIA, and other intelligence agencies moved onto Badaber Airbase in December 1958, two months after Ayub had seized power and seven months before the United States and Pakistan signed a formal, 10-year renewable lease for the facility. The contract granted the United States 126 acres for its facilities, with the right — subject to Pakistani approval — to expand to 504 acres as needed. U.S. intelligence-collection programs in Pakistan continued to expand throughout John F. Kennedy’s presidency. In 1961, the Kennedy administration decided to enlarge the Badaber facility to the full 504 acres to accommodate new operational buildings and additional antennas to the intelligence collection program.64 “The Peshawar station,” Deputy Secretary of Defense Roswell Gilpatric wrote to Acheson, “is one of a very limited number of important overseas intelligence gathering stations at which essential information, including warning intelligence, can be collected.” Gilpatric described expanded collection there as “a matter of first importance.65” A December 1963 background paper for the chairman of the Joint Chiefs of Staff, Gen. Maxwell Taylor, concurred, describing America’s facilities within Pakistan as “essential to our national interest.”66
It would have been in the U.S. interest to play down Badaber’s importance when dealing with Pakistani leaders. However, there is little evidence that this was done during the initial years of the program.
Even as intelligence and defense officials awaited Ayub’s approval for the expansion, they continued to boost investment in Pakistan-based intelligence collection facilities. The U.S. Air Force, U.S. Navy, CIA, National Security Agency, and National Reconnaissance Office were among the many agencies operating from Badaber.67 Later accounts describe “rows of ranch houses,” a bowling alley, a movie theater, a Defense Department school, and a golf course.68 At its height, some 1,200 U.S. personnel worked at the site.69 Technicians stationed at Badaber later wrote of their work’s close link to U.S. national security: “We placed more emphasis on [targeting] certain [Soviet] bases to … ensure that our missiles could completely obliterate them should the need ever arise.”70
By the time President Lyndon B. Johnson assumed office in November 1963, the U.S. intelligence and defense communities also had developed a network of at least a half-dozen additional sites around Pakistan, all targeting Soviet and Chinese nuclear and missile programs.71 Two of the facilities — atomic energy and missile launch detection facilities in Lahore and Karachi — were part of the expanded Atomic Energy Detection System, which was key to policing the Nuclear Test Ban Treaty.72 All six of the facilities were ad hoc arrangements set up without formal written permission from the government of Pakistan.73 However, U.S. defense and intelligence agencies could not have occupied the sites — which were located on Pakistani military bases in Lahore, Rawalpindi, and Karachi — without the knowledge and approval of Pakistani leaders. Indeed, U.S. ambassador Walter P. McConaughy reported that Pakistani officials had inspected the units and received partial briefings on their operations. The intelligence community sought repeatedly to regularize the status of these additional facilities, unsuccessfully pressing Ayub’s government to enter formal negotiations on the issue.74
It would have been in the U.S. interest to play down Badaber’s importance when dealing with Pakistani leaders. However, there is little evidence that this was done during the initial years of the program. Instead, there is every sign that Pakistani leaders were aware of how important these bases were to Washington’s overall effort to collect intelligence on Soviet and Chinese nuclear assets. There was no hiding Washington’s urgent need to understand what was taking place at the classified sites scattered across the Soviet Urals and at China’s Lop Nur facility.
Balancing Interests and Leverage in a Clandestine Relationship
Sims’ model of an asymmetric liaison relationship is one in which the host country derives little value from the relationship while its partner derives high value, creating a dependency. In such cases, she predicts that the survival of the arrangement will depend upon the dependent partner providing constant side payments — or blackmail.75 The U.S. collection programs in Pakistan certainly fit this pattern.
Ayub clearly understood that Badaber gave him bargaining leverage over Washington and he maintained personal control over all matters concerning the base.76 “Decisions governing Peshawar are made by Ayub,” the U.S. embassy later reported, “because he regards it rightly as a policy instrument, and therefore too important to entrust to functionaries.”77 CIA analysts assessed in 1964 that “[t]he Pakistani President knows that the strongest card he holds is the US communications facilities in Peshawar. He knows that they are of great strategic value to the United States and to the free world.”78 “Paks [sic] assume Peshawar is of very substantial importance to us,” the U.S. ambassador to Pakistan wrote two years later, “because they have been assured over the years that this is so by a formidable list of high US officials.”79 That knowledge translated to leverage. As former Foreign Minister Abdul Sattar later wrote, “The base … was not only the most concrete and strategic benefit the US derived from the alliance, but exponentially increased Pakistan’s importance in the eyes of the strategic community.”80 By maintaining personal control over the ongoing discussions with Washington over its intelligence facilities, Ayub also helped to solidify his own position. As a military ruler who had seized power in a coup, his ability to keep the Pakistani armed forces well-supplied was critical to his continued political survival.
America’s intelligence-collection operations were focused entirely on the Soviet and Chinese nuclear and missile programs, which were targeted toward the United States and its allies, not Pakistan. Pakistani Ambassador to the United States Agha Hilaly claimed his government’s “technical people” assured his foreign ministry “the product was of no importance” to Pakistan.81 Ayub too, wrote to Johnson that “[t]he Peshawar facility on our soil in no way contributes to our security, useful though it may be to your country.”82 There’s no question that the United States benefitted more from the intelligence collected at Badaber than Pakistan ever could. For Pakistani policymakers, the secret programs’ appeal lay in the enhanced U.S. support that the country was receiving, rather than the programs themselves. Sattar makes no pretense about Pakistan’s motives in agreeing to host clandestine U.S. intelligence collection facilities. He wrote that Ayub granted the base with an eye to Pakistan’s economic and military needs. “Also, Pakistan was interested in the delivery of B-57 bombers.”83
Not only did Pakistan find little value in “the product,” but Badaber actually became somewhat of a liability in 1960 after Soviet air defenses shot down a U-2 launched from its runways. Although Khrushchev was probably blustering when he warned that Moscow would retaliate if Pakistan allowed any future such flights, Ayub faced sharp domestic criticism nonetheless for permitting it. After Khrushchev noted he had Peshawar “circled in red on his maps,” Ayub ever after insisted that Pakistan faced potential Soviet nuclear attack for its continued collaboration with Washington.84 Soviet and Chinese representatives also raised the base as a problem when Ayub sought to reduce his dependence on Washington by expanding Pakistan’s diplomatic ties.85
The U.S. clandestine presence benefited Pakistan almost exclusively through Washington’s side payments, rather than any direct payoffs. Many intelligence-liaison relationships involve service-to-service training and technology transfer. However, training and technology-sharing was limited at Badaber. U.S. intelligence agencies employed hundreds of Pakistanis in support jobs and trained at least 100 Pakistani designees who moved on to operational jobs in Pakistan’s military and intelligence agencies.86 Nonetheless, U.S. policymakers limited Pakistan’s access to sensitive programs and technologies. Pakistani air force pilots flew missions collecting Soviet rocket telemetry, for example, but had no access to the equipment needed to interpret the intelligence they collected.87 Moreover, the collection packages on these aircraft were useless against Pakistan’s key rival, India.88 During the late 1950s and early 1960s, while other U.S. aid was flowing comparatively freely, this lack of access to the installations and their technologies remained only a minor grievance. When the side payments slowed and eventually stopped under Johnson, however, the grievances assumed greater importance. U.S. diplomats in 1965 described Pakistani leaders as “increasingly outraged” at their exclusion from installations established on Pakistani soil.89 Pakistani diplomats denounced base security procedures, such as blindfolding Pakistani workers passing through secure areas, as national insults.90 The shift in Pakistani reactions supports Sims’ argument that side payments are critical to maintaining a clandestine alliance based on asymmetric interests.
One might expect Pakistan’s dependence on U.S. support to maintain its economy and military would outweigh Washington’s interest in a handful of secret collection sites. Indeed, many in the White House believed just that. “As the weaker power on the subcontinent,” National Security Council South Asia expert Robert Komer insisted, “Pakistan needs the U.S. connection more than we need it.”91 All the same, policymakers in the CIA, State Department, and Department of Defense argued against risking the relationship. They warned that the value of intelligence that America was collecting was too high, that withdrawal would drive Ayub closer to China, and that the Southeast Asia Treaty Organization and the Central Treaty Organization would collapse should Pakistan pull out.92 Regardless of the truth of these arguments, Pakistani leaders bargained as if they held the upper hand.
For Pakistani policymakers, the secret programs’ appeal lay in the enhanced U.S. support that the country was receiving, rather than the programs themselves.
Soon after U.S. operations at Badaber began, Ayub cited increased Indian and Russian incursions into Pakistani airspace to press Washington to supply F-104 aircraft, which “Pakistan felt very strongly” should be provided for free.93 After the 1960 U-2 incident, Pakistani Foreign Minister Manzur Qadir visited Washington to warn that the United States needed to “provide some kind of demonstration” of why it was better to ally with the United States than to remain neutral. The best demonstration, Qadir suggested, would be to “increase Pakistan’s military preparedness in her own territory.” Over the next several months, the U.S. embassy in Karachi reported continuous pressure from Pakistani leaders for additional military aid. U.S. Ambassador William Rountree wrote in July, for example, that providing the Pakistani air force with the requested F-104s — over and above those agreed to before the U-2 incident — “would do much to strengthen their determination not to weaken as [the] result [of] recent developments.”94 In the end, the United States delivered the aircraft.
The Pakistani government increasingly interfered with U.S. intelligence operations to drive home the point that the United States could not operate there without Pakistan’s goodwill. In 1961, Pakistan evicted U.S. Navy fliers monitoring Soviet missile launches at Kapustin Yar and other test ranges, claiming they had flown over sensitive airspace. To restore the program, U.S. negotiators agreed to Pakistan’s demands that the missions be flown by Pakistani air crews in aircraft that they were familiar with. The Defense Department and General Dynamics, which built the aircraft, trained a squadron of 40 Pakistani pilots to take over the mission, supplying the air force’s new 24th Electronic Intelligence Squadron with RB-57Fs, a variant of the B-57 bomber.95 U.S. and Pakistani negotiators built on this agreement in 1963, opening additional operational jobs at Badaber to hundreds of Pakistani military personnel (all trained for their new positions at America’s expense). U.S. diplomats reported that this accord removed a “point of friction.”96
How much aid did Pakistan gain through side payments in exchange for permitting U.S. intelligence programs to operate in its country that it would not have otherwise received? Too many documents remain classified to say for certain, particularly since the United States also provided aid through multiple other programs. Nevertheless, the timing of certain aid packages suggests that the quid pro quo aid was substantial. Pakistan gave the go-ahead to modify Badaber’s runways to accommodate U-2 flights almost immediately after the National Security Council approved a $784.7 million three-year military aid package, nearly a year before reaching the formal agreement to lease the base.97 After the United States and Pakistan began negotiating the 10-year lease for Badaber, Pakistani leaders pressed Washington to deliver a long list of new weapons systems in 1958, including “early delivery of a light bomber squadron.”98 Is it coincidental that the new requests came as Washington negotiated for rights to Badaber? The month after the lease was signed, the New Republic criticized an aid program run amok, noting that a program originally budgeted in 1954 for $25 million “has now passed $522 million in hardware alone … [to] round out the $1.5 billion aid total for Pakistan.”99
Between alliance obligations and intelligence-liaison side payments, Pakistan received nearly $1 billion in military assistance from the United States between 1954 and 1964.100 Whatever the breakdown was between liaison side payments and other alliance and bilateral obligations, National Security Council staffers considered the whole to be largely “rent” for Badaber.101 “We should … be clear,” the State Department’s Policy Planning staff warned, “that in Pakistan eyes [sic], military assistance figures as part of the price we pay for Peshawar.”102 Komer wrote, “Our access to very valuable real estate in West Pakistan was of course a bilateral exercise paid for by US aid.”
U.S. policymakers also grew increasingly convinced during the Kennedy and Johnson administrations that the intelligence facilities — and not Pakistan’s membership in the treaty organizations — represented Pakistan’s key value as an ally. “Despite its allied status,” Komer wrote to the chairman of the Joint Chiefs of Staff, “Pakistan makes very little real contribution to SEATO or CENTO. What we do get from Pakistan are some highly important facilities, but little else.”103 Writing to Johnson, Komer judged “they’ve really given us nothing aside from some admittedly valuable real estate (if so, it’s the most expensive we’ve ever bought).”104 Komer was among the strongest critics of Pakistani “blackmail,” but he was hardly alone.105 Both Kennedy and Johnson complained about how fears that Pakistan would shut down Badaber constrained their freedom in dealing with India.
Pakistani leaders also linked the liaison program with Washington’s overall South Asia policy. Sattar later described leasing Badaber to Washington as “an important factor in restraining Washington from selling modern weapons systems to India.”106 In April 1961, Kennedy nonetheless provided military support to India in its growing confrontation with China. Ayub retaliated by shutting down the CIA’s Tibetan operation in East Pakistan.107 Agency officers could neither launch their resupply missions from Kurmitula to support fighters in the field nor bring back the Tibetan trainees as they completed their training in Colorado.108 Ayub only relented after Kennedy pressed him during his July 1961 state visit to Washington. In return, Ayub extracted a pledge from Kennedy that the United States would notify Pakistan in advance of any future arms deliveries to India and come to Pakistan’s defense if India were to use those arms against Pakistan.109
Ayub’s accusations that the United States had betrayed Pakistan were useful in extracting the new commitment, but they also reflected genuine concern. As a senator, Kennedy had argued that Washington should build ties with India. “If India collapses, so may all of Asia.”110 Pakistani diplomats in Washington warned Ayub that the new president’s advisers also favored “free and democratic” India over “theocratic and military” Pakistan.111 Ayub pressed Vice President Johnson for reassurances during his May 1961 visit to Karachi but found his response to be “general.” Kennedy’s invitation to both Nehru and Ayub to visit Washington — rather than singling out Ayub first, as the closer ally — so alarmed Ayub that he moved his visit up six months.112
U.S. policymakers also grew increasingly convinced during the Kennedy and Johnson administrations that the intelligence facilities — and not Pakistan’s membership in the treaty organizations — represented Pakistan’s key value as an ally.
To Ayub, Washington’s attempts to strike a new South Asian balance represented a betrayal of its moral obligations and the 1959 Baghdad Pact pledge of mutual support. Ayub warned Kennedy during his July 1961 visit to Washington that “the Pakistani people would force his country out of the pacts and alliances and everything,” should the United States provide India with military aid.113 At a minimum, Ayub argued, the United States should force India to settle the Kashmir dispute in return for aid.114 Kennedy’s limited response to India’s seizure of the Portuguese colony of Goa in December 1961 increased Ayub’s skepticism that the United States would come to his aid in a crisis. If Washington wouldn’t stand by its NATO ally, Portugal, when it was invaded, how could it be trusted to stand by Pakistan? Ayub again cut off support for the CIA’s Tibet program — this time for good.115 Ayub later wrote in his memoirs that he had become convinced that Kennedy and the United States were “out to appease India.”116
Over the next year, as India and China moved toward war — and Kennedy did indeed send military support to New Delhi — Ayub increased his pressure on U.S. intelligence facilities. He withheld permission to expand the Badaber base and withdrew promised logistic and other support to the facility.117 Secretary of State Dean Rusk told a closed session of the Senate Committee on Foreign Relations in May 1962 that the administration was “concerned” about Ayub’s attitude. “This has moved over into a lack of cooperation on some very sensitive military problems where it is beginning to hurt us,” Rusk said. “They are being extremely difficult.”118 Komer and Tom Hughes, the State Department’s chief of intelligence, warned that trading “baksheesh” for restored intelligence support “merely convinces the Paks [sic] they’ve got us where they want us.”119 Nevertheless, State Department officials judged that “[t]he value of our special relationship with Pakistan … is such that we cannot contemplate withdrawal.”120 Kennedy abandoned a joint U.S.-U.K. plan to provide supersonic jets to India after CIA analysts warned that Ayub would retaliate against Badaber if the sale went through.121
The Beginning of the End: Missed Signals and Bad Reads
Successful alliance bargaining requires negotiators on both sides to have a reasonable understanding of the other side’s interests and perceptions. As this understanding eroded between the United States and Pakistan, the bargaining necessary to maintain the relationship grew increasingly difficult.
When Kennedy appeared less supportive of Pakistan than his predecessor, Ayub launched a series of overtures to the Communist bloc. These moves likely were initially meant to signal to Washington that Pakistan had other options. Over time, however, Pakistani leaders came to view their country’s improving relationship with China as an important strategic gain in their confrontation with India. In 1961, Pakistani leaders arranged a $30 million oil exploration project with Moscow.122 By mid-1963, Pakistan negotiated a series of deals with Beijing that involved resolving longstanding border disputes, exchanging most-favored nation status with China, and providing Beijing with its first non-communist civil-aviation agreement. The same year, Pakistan upgraded its interest section in Havana to embassy status and signed barter agreements with a half-dozen Warsaw Pact states.
The gestures were small, but they challenged U.S. efforts to isolate Moscow and Beijing and spurred questions regarding Pakistan’s commitment to anti-communism. CIA analysts judged that Ayub was seeking greater foreign policy independence within the larger framework of his alliances with the United States, but would continue his “repeated requests for demonstrations of US confidence and support.”123 U.S. diplomats in Pakistan agreed, writing that the moves were “undoubtedly” meant to pressure Washington.124 Ayub hinted at a quid pro quo, telling visiting Under Secretary of State George Ball that Pakistan “would not hesitate” to abandon its overtures to Moscow and Beijing “if there were a power prepared to underwrite” its needs.125 Ayub’s later comments in his diary suggest he was indeed seeking to use ties with Moscow to pressure Washington. “It is in our interest that our relations with the Soviets gain depth,” reads one entry. “We can then develop greater leverage with the USA and India.”126
Johnson, who had hosted Ayub at his Texas ranch in 1961, came to the presidency considering Ayub a friend. The new president expressed “the greatest of confidence in Ayub” to CIA Director John McCone and National Security Adviser McGeorge Bundy in his first briefing on South Asia as president.127 Nevertheless, Pakistan’s announcement on Nov. 28, 1961, that Ayub would host Chinese Premier Zhou Enlai for a state visit convinced Johnson that he needed to be “forceful.”128 Johnson dressed down Pakistani Foreign Minister Zulfiqar Ali Bhutto during a Dec. 2 condolence call, warning that the invitation threatened the White House’s chances of pushing future aid through Congress.129 Days later, the National Security Council shelved plans to offer Pakistan a $40 million three-to-five-year Military Assistance Program package. The package, a follow-on to Eisenhower-era pledges that had now been met, was put on hold until Pakistan could “fulfill the basic requirements of its alliance relationship with us” and back off its relations with China.130 The move that Ayub had intended to use to gain greater leverage with the United States instead undermined Washington’s perception of him as a reliable ally.
As the Vietnam conflict increasingly dominated Johnson’s attention, his impatience with Ayub grew. The United States sent roughly a third of its aid to the subcontinent — more than any other region — yet seemed to exercise decreasing influence there. Pakistani overtures to China threatened to undercut congressional support for Johnson’s overall foreign aid program, including to Vietnam. Alone among the Southeast Asia Treaty Organization states, Pakistan refused to make even a token gesture of support to the U.S. effort.131 Johnson told the U.S. ambassador to Pakistan, McConaughy, in July 1964 that he doubted that the United States was “getting much for its money.”132 Bundy and his deputy, Komer, rushed to remind Johnson of the importance of Badaber, but they agreed that “the price was high.”133
Ayub may have planned the Beijing and Moscow visits to convince Johnson that Pakistan had alternatives if the United States didn’t boost its support, but the maneuvers backfired.
Over the next year, Pakistan continued to move closer to Beijing and the Soviet Union. In March 1965, Ayub made a state visit to Beijing, the first of any Pakistani leader. U.S. diplomats thought the visit represented a “significant consolidation” of Sino-Pakistani ties.134 Weeks later, Ayub visited Moscow, two short weeks before a scheduled state visit to Washington. That May, U.S. intelligence sources reported that Pakistani military officers were paying frequent visits to China and suggested that Pakistani and Chinese leaders may have reached a “top secret” understanding.135 Meanwhile, the government-controlled Pakistani press amplified an anti-America campaign. If Goa was a turning point for Ayub, Johnson’s watershed moment was here, with Ayub seeming to publicly turn his back on Washington and the anti-communism effort.
Ayub may have planned the Beijing and Moscow visits to convince Johnson that Pakistan had alternatives if the United States didn’t boost its support, but the maneuvers backfired. Johnson rescinded his late April invitation for Ayub to visit Washington, fearing the juxtaposition of the three visits would endanger congressional support for his foreign aid bill.136 That summer, Johnson ordered all new aid to Pakistan and India frozen until Congress passed the Fiscal Year 1966 aid bill. “The postponement was designed to show Ayub,” National Security Council staff member Harold Saunders wrote, “that American aid was far from automatic, and to be a forceful reminder that his relations with Communist China and other U.S.-Pakistani difficulties could endanger his country’s economy.”137 Johnson was determined to “make Ayub come to us and to play hard to get on all fronts until he does so.”138 Johnson and his advisers were convinced that Pakistan was too dependent on U.S. support to risk a permanent breach. The trick was to convince Ayub that, unlike his predecessors, Johnson would follow through on threats to cut aid despite the associated risk to Badaber. U.S. News and World Report observed in May 1965 that the Johnson administration was now refusing to pay “blackmail,” giving Ayub the “shock treatment” and leaving Pakistan in the “doghouse.”139
Rather than getting the message, an angry Ayub increased pressure on U.S. intelligence facilities, presuming that Washington would ultimately acquiesce. The U.S. ambassador to Pakistan reported that “Ayub seems to feel our postponement move challenges him, and strikes at [the] self-respect of country by seeking to penalize Pakistan publicly for pursuing ‘independent’ foreign policy.”140 In March 1965, U.S. diplomats warned that Pakistani leaders, “in a mood of emotionalism and frustration,” might abandon their western ties.141 The next month, Foreign Secretary Aziz Ahmed demanded an accounting for all equipment and buildings installed at Badaber since 1959, as well as full and complete access to “every sector” of the facility.142 Ahmed further demanded a date by which the United States would close down the smaller intelligence facilities that had been established based on handshake agreements.
Meanwhile, Pakistan began to send operatives into Indian-held Kashmir in a doomed bid to spark a popular uprising. As the resulting clashes spiraled into war with India, Ayub demanded U.S. support. He insisted the 1959 Mutual Defense Assistance Agreement impelled Washington to come to Pakistan’s aid. Pakistan expected “not only moral and diplomatic but tangible military support from allies,” Ayub’s foreign minister told a Sept. 5, 1965 press conference.143 Johnson refused Ayub’s and Bhutto’s demands for U.S. support. “I made up my mind last April,” the president told Cabinet members, “we simply were out of business with Ayub and [Indian Prime Minister Lal Bahadur] Shastri until we sign a contract. We are now in a position to tell them to quit fighting or else we will do no more business with them.”144 Rather than meeting Pakistani demands to provide military support, or at least pressure India to withdraw, Johnson cut off U.S. arms deliveries to both nations, while supporting U.N. efforts to broker a settlement. The move had strong bipartisan support in Congress, where 143 members voted to cut off aid — including economic and food assistance — to both warring parties.145
The CIA warned officials at Badaber to prepare for a possible Pakistani move to seize the facility.146 Instead, Pakistani soldiers shut down the smaller intelligence sites in Lahore and Karachi.147 Pakistan also withheld the permission needed to evacuate U.S. personnel from Lahore — which was within range of Indian artillery — and Dhaka.148 New bureaucratic problems proliferated, from customs seizures of Americans’ personal mail to grounded logistics flights.149 “The pattern seems … to point to a deliberate Pak [sic] effort to show us they have cards too, and as pressure to get us to resume aid,” Komer wrote to Johnson.150 The U.S. ambassador to Pakistan agreed, writing, “We estimate prospects for reprisals … would be in direct proportion to [the] growth of GOP dissatisfaction with the US position … on Kashmir.”151
Johnson refused to budge. He supported U.N. ceasefire negotiations but would neither support Pakistan materially nor press India to withdraw. Johnson maintained the embargo after the fighting ended in late September 1965, closing U.S. Military Assistance Advisory Groups in both the Pakistani and Indian capitals. Ayub visited Washington one final time in December 1965 to press America to restore its aid. Johnson told him that the cost of close relations with the United States would be abandoning any “serious relationship” with Beijing and recognizing that Washington “was not going to let Pakistan tell us how to handle India.”152 Ayub left empty-handed, except for a handful of small development loans offered to help him save face.
The Final Collapse of the First U.S.-Pakistani Clandestine Accord
Little research has been done on how and when nations decide to end intelligence-liaison programs. The assumption seems to be that these relationships end naturally when the benefits no longer exceed the costs. This assumption is almost certainly wrong, given the complexity and politics of assessing the cost-benefit equation. It is more likely that liaison programs survive only as long as they maintain constituencies within partner governments. This would suggest that some programs might end with a change in organizational or national leadership, regardless of their merit, while others continue long after the costs come to outweigh the benefits. As Allen Dulles once observed, “Most intelligence operations have a limited span of usefulness. The most difficult decision is when to taper off and when to stop.”153
In the Pakistani case, Ayub’s attempt to teach Washington the costs of neglecting Pakistan’s interests instead eroded the consensus that Badaber was invaluable and irreplaceable. During the early 1960s, the CIA, State Department, and Defense Department had questioned Kennedy’s more ambitious initiatives toward India because of the perceived risks to operations at Badaber.154 By 1964, intelligence community concerns about Ayub’s long-term intentions made it easier to shift the consensus. If U.S. intelligence assets in Pakistan were already at risk, it made sense to consider alternatives. Advances in U.S. satellite technologies also shifted the cost-benefit analysis. By 1965, National Reconnaissance Office spy satellites were successfully collecting imagery and electronics intelligence that could previously only have been captured from strategically located sites such as Badaber.155
Ayub’s moves against Badaber caused the intelligence community to determine in a May 1965 report to Johnson that continued U.S. tenure was “at risk”156 The report crystalized a new consensus that it was time to consider ways to reduce America’s dependence upon Badaber. The U.S. Intelligence Board, made up of the heads of the intelligence agencies, recommended that Washington develop alternatives to the facilities in Pakistan.157 Rusk wrote to Bundy in July that “the whole question of our intelligence relationships with Pakistan must be subject to the most searching examination.”158 Defense Secretary Robert McNamara agreed, stating that the United States shouldn’t “let a little blackmail affect us — whether people, installations or any sort of thing.”159
Neither Johnson nor the intelligence community initially viewed their search for alternatives to Pakistan as more than a contingency plan and a means of reducing Pakistan’s leverage. Johnson had ordered an “orchestrated U.S. Government effort” to convey America’s dissatisfaction with Pakistan after Ayub’s early 1965 visits to Beijing and Moscow.160 Komer seized on contingency planning for finding alternatives to Badaber as one tool of persuasion. “If we cut back a little on personnel or new construction [at Badaber],” he wrote Bundy, “it would shake Pak [sic] confidence in what they regard as their hole card. They’d really be shaken if they thought we were getting ready to pull out.161”
Bundy, in turn, wrote to the director of the CIA, Vice Adm. William F. Raborn, McNamara, and Rusk, proposing that they suspend nonessential activities and temporarily reduce personnel strength at Badaber to “convey the impression” that Washington was “taking positive measures to reduce its dependence on US intelligence facilities in Pakistan.” As a final point, Bundy urged them to explore developing alternative facilities in Iran as “insurance.”162 Two weeks later, Bundy convinced Johnson to cancel planned improvements at Badaber, which might “emphasize to the Pakistanis our dependence on the Peshawar complex.”163 U.S. officials at Badaber made the adjustments, but there is no evidence that anyone in the Pakistani leadership considered the changes to be meaningful.
Neither Johnson nor the intelligence community initially viewed their search for alternatives to Pakistan as more than a contingency plan and a means of reducing Pakistan’s leverage.
Johnson also endorsed the Bundy-Komer plan to develop Iranian sites as a contingency for losing Badaber.164 Raborn initially argued that Washington should pursue further negotiations with Pakistan, a “bonded ally.” He dropped his objections after Komer warned him that Johnson would no longer respond to Pakistan’s “blackmail tactics.” Similarly, U.S. Ambassador to Iran Armin H. Meyer warned that boosting the U.S. intelligence presence in Iran would create “tailor made targets for anti-Shah and/or anti-American elements,” potentially jeopardizing the Shah’s hold on power.165 State Department officers, coordinating with Komer, counseled Meyer that he “misunderstood” the situation and was expected to comply with Johnson’s directives.
By May 1966, Washington had both identified contingency collection sites in Iran and won the Shah’s agreement to host any relocated operations,166 although U.S. officials still hoped to maintain facilities in Pakistan. Indeed, U.S. officials, including Johnson and diplomats in Pakistan’s capital, spent considerable time working to convince Ayub to allow Washington to continue operating Badaber and reopen the six auxiliary facilities in Lahore, Karachi, and Rawalpindi.167 Shifting from Badaber would involve losing a substantial investment in the form of infrastructure that the United States had already built and would increase America’s dependence on the mercurial Shah. The back-up facilities in Iran also lacked the access to China’s Xinjiang Province that the Pakistani location had provided. Meanwhile, the nascent U.S. satellite program was better positioned to complement Badaber operations than replace them. All the same, Johnson and his advisers had broken with the earlier consensus that Badaber was irreplaceable and deliberately reduced America’s dependence upon Pakistan.
Viewed objectively, both the United States and Pakistan still stood to gain from continuing their relationship. Although both had developed new alternatives, those alternatives were second-best. Just as the United States would have preferred to continue collecting intelligence from Pakistan, the Pakistani military would have preferred continued access to U.S. equipment, both for reasons of familiarity and quality. Reaching that point of greatest mutual benefit, however, would have required more trust, flexibility, and empathy than either side appears to have been able to muster. As a result, each side miscalculated the leverage it had over the other while underestimating the other’s alternatives. Indeed, the collapse of U.S.-Pakistani negotiations over the Badaber lease illustrates the difficulties states engaged in intense and emotional bargaining face in accurately assessing leverage.
For Johnson, Ayub’s growing ties with Beijing represented a personal betrayal. How, he asked Ayub’s representatives, could Pakistan — a member of both the Southeast Asia Treaty Organization and Central Treaty Organization anti-communist alliances — become friends with communist China while American boys were dying in Vietnam?168 Johnson feared that Ayub’s ties to Beijing and flirtations with Moscow threatened his overall ability to win funding from Congress for his overseas programs, including Vietnam. The harassment of U.S. personnel at Badaber and other sites also represented a breach of Pakistan’s agreements, not to mention Ayub’s personal assurances of friendship. Johnson also resented Ayub’s refusal to lend any support to U.S. operations in South Vietnam, making Pakistan the sole member of the Southeast Asia Treaty Organization not showing the flag there.169
Ayub, too, saw Johnson as having betrayed both personal and national commitments. In November 1965, Pakistan’s finance minister confided to U.S. diplomats that Ayub suspected that the CIA was working to undermine and oust him.170 Ayub wrote in his diary about alleged CIA plots, including a supposed plan to restore East Pakistan to India.171 He obsessed over the supposed risks Pakistan faced from hosting Badaber, recalling Khrushchev’s threats after the U-2 incident.172 Johnson had, Ayub believed, the power to prevent Pakistan’s disastrous loss to India, either by intervening directly or by forcing India to back down. Not only had Johnson failed to do so, but the U.S. arms embargo undercut Ayub’s standing with his military base.
By withholding aid, Johnson thought to force Pakistan to act according to what the United States viewed as its treaty obligations. He expected Pakistan to support Washington’s anti-communist agenda, allow declared U.S. intelligence collection operations to continue unhindered, and work to deescalate tensions with India. Instead, Ayub and his advisers viewed Johnson’s demands as impinging on their country’s sovereignty. Ayub had sought to convince Washington it needed to restore aid and accept his “independent” foreign policy by demonstrating how much the United States needed Pakistan’s support. Instead, he spurred U.S. policymakers to find alternatives to the facilities they had been using in Pakistan. Both governments expected the other’s dependencies to ultimately produce concessions. Neither government recognized that the other was willing to accept imperfect alternatives rather than concede core principles.
The 10-year U.S. lease for Badaber expired in July 1969. The deadline played a growing role in bilateral relations over 1967 and early 1968 as Pakistan pressed for renewed military aid to replace equipment lost in the 1965 war.173 CIA analysts predicted that Ayub would allow the facilities shuttered during the war to be reopened “if the U.S. refocused on delivering aid and repairing the friendship.”174 Ayub himself suggested that the sites — with their highly classified technologies — could be reopened if they were run by Pakistani personnel.175 Instead, Washington moved the critical Atomic Energy Detection equipment to Badaber, making no effort to reopen the secondary sites.176 As the deadline for renewing the lease for Badaber approached, the State Department directed the U.S. ambassador to avoid raising the subject in his discussions with Ayub. “The temptation to ‘buy’ an assured future for the Peshawar facility with one or two hundred tanks is very real,” one cable acknowledged, “but [it] would probably result in intolerable pressures from [the Government of Pakistan] for more and more hardware.”177
On Dec. 21, 1967, the State Department advised the ambassador that Washington wanted to renew the lease on Badaber — “though not at any price.”178 The cable counseled the ambassador to make clear to Pakistan that the issues of base renewal and the military aid embargo were separate and unrelated. Washington analysts expected Pakistan leaders to drive a hard bargain but assumed Ayub would ultimately renew the lease. Ayub wrote in his diary in March 1968 that he had decided to close Badaber.179 In a later entry, he cited the cut off of military aid and the perceived U.S. “betrayal” in not coming to Pakistan’s aid during the 1965 war as chief reasons for ending the liaison relationship.180 In April, his foreign minister made the decision official, notifying Washington that the country was terminating the lease.181
Was the decision a ploy to force Johnson to rethink the military aid embargo or an attempt to curry favor with Soviet Premier Alexei Kosygin, who was scheduled to visit Pakistan days later? Was it a response to a series of “loaded questions” that a CIA source reported that Zhou asked Foreign Minister Sharifuddin Pirzada the previous December? Possibly all of the above. Ayub had offered to close the base during his visits to Moscow, in return for arms and a shift in the Soviet position on Kashmir (neither of which occurred).182 Ayub’s thinking may well also have been affected by two heart attacks that he suffered only weeks before the diary entry. Heart attacks are often linked with depression and Ayub’s advisers confirm he was struggling emotionally and physically.183 All the same, Pakistani leaders were well aware that neither Moscow nor Beijing could meet their needs.
Perhaps the most important factor precluding a negotiated settlement, however, was Johnson’s refusal to restart military aid.
The U.S. ambassador described the announcement as a “logical way … to initiate discussion.”184 The Washington Post drew on anonymous sources to report that “the United States does not regard the termination notice as official,” describing the CIA and National Security Agency as “very eager to hold on” to Badaber.185 Ayub mistakenly believed Washington expected the notice of cancellation, suggesting that he may have meant his announcement to be the opening of new negotiations.186 His advisers continued to assure him that Badaber remained “vital to United States security,” suggesting he believed he had more leverage than he actually commanded.187 Several Pakistani officials told U.S. officials the closure was open for negotiation, while others relayed messages through Iran and Turkey that there was room for discussion so long as the United States was willing to reopen the weapons pipeline.188 U.S. diplomats described a “vital tug-of-war” within the top levels of the Pakistani government over the closure.189 Hilaly brought long lists of desired arms to a State Department meeting a few short weeks after the termination notice.190
Shortly after Ayub informed Washington he would not renew the lease, Pakistan’s foreign minister announced the secret base termination order in an open National Assembly session. This public acknowledgement of the facility both angered Johnson and limited the scope for compromise.191 Perhaps the most important factor precluding a negotiated settlement, however, was Johnson’s refusal to restart military aid. Diplomats and officials at the Defense Department and CIA proposed a series of compromises and intermediate steps the United States could take to address at least some of the Pakistani military’s post-war resupply needs. But Johnson would not budge. He believed the United States to have the high ground and refused to give further military aid without concessions on China and guarantees that future intelligence operations in Pakistan would not be harassed. There is no sign that Johnson questioned the continued value of Badaber, but his primary focus was on Vietnam.
Johnson’s firmness reflects his frustrations with Pakistan and his preoccupation with Vietnam, but he also had an advantage that his predecessors had lacked. Washington had become less dependent upon Badaber than Ayub realized. By 1966, the National Reconnaissance Office was launching imagery satellites roughly monthly, providing an alternate source of intelligence on Soviet nuclear strike capabilities.192 A CIA review found that the Peshawar facilities were “substantially less important now” and that many intelligence functions could be moved to Iran, as noted above.193 The technicians did report that there were “no substitutes” for some targets that could only be reached from Pakistan, but judged the intelligence that could be collected only from Pakistan to be “desirable” but not “vital in itself.” Nevertheless, the CIA recommended a quid pro quo to keep the facility. The State Department, conversely, advised Johnson that his decisions regarding economic aid and military supply to Pakistan should be made purely “on the basis of broader United States interests,” rather than to shift Pakistan’s position on Peshawar.194
While both sides probed for the possibility of a compromise, the dialogue soon degenerated into a fraught debate over the U.S. withdrawal schedule and what equipment America would leave behind for Pakistani use.
Both the United States and Pakistan profited from their first national security collaboration. The covert relationship raised Pakistan’s importance on the world stage, providing it with leverage and high-level access it could not have gained through overt alliances such as the Southeast Asia Treaty Organization and the Central Treaty Organization. During the 15 years between the first serious U.S.-Pakistani security discussions in 1954 and Badaber’s 1969 closure, Pakistan received enough military and other aid to consolidate its military and launch its economy. While Pakistan couldn’t prevent America’s outreach to India, especially during the 1962 Sino-Indian war, U.S. leaders clearly calculated Pakistan’s response into their India policies and provided less lethal aid than they otherwise might have
Meanwhile, the United States gained strategically from the information it collected on Soviet and Chinese nuclear and missile activities. Both Pakistan and the United States also profited from acknowledged collaborations, such as the Southeast Asia Treaty Organization and the Central Treaty Organization, and collaboration in international forums. These interactions cannot be fully isolated from one another. Even as Johnson’s advisers pressed him to “get tough” with Ayub on Badaber negotiations, they worried that the two treaty organizations might not survive a Pakistani withdrawal.195 Pakistan may not have participated in the Vietnam effort, but its continued engagement in the two pacts added to the groups’ symbolic value. None of these factors, however, was sufficient to save the clandestine relationship.
This case highlights the importance of side payments in preserving asymmetric liaison relationships. Even more, though, it illustrates the role of implicit assumptions in driving behavior within those relationships. True, the U.S.-Pakistani relationship collapsed when it became clear that Johnson would not lift sanctions against Pakistani military resupply. The quality, amount, and type of side payments that Washington provided provoked continual debate throughout the lifespan of the relationship. What proved most important, however, were the assumptions that each side brought to negotiations. To Eisenhower and his successors, the act of joining anti-communist pacts represented a commitment to shun Beijing and Moscow. To Pakistani leaders, those same pacts represented a pledge to protect Pakistan’s interests and territorial integrity — even from the country’s own folly. To Washington, strengthening ties with New Delhi and supporting it in a conflict with China was consistent with the anti-communist stance that had first motivated its relations with Pakistan. To Pakistan, these overtures not only strengthened a deadly foe but betrayed the very idea of an alliance.
Each state also misjudged its leverage over the other. Ayub’s early successes in winning new concessions from the Eisenhower administration clearly led him to expect that the technique would continue to work with Eisenhower’s successors. Rather than reassess his expectations when these tactics no longer produced the results he wanted, Ayub presumed he needed to apply additional pressure to remind Washington of its dependence. Johnson’s advisers, too, presumed that Pakistan was more dependent upon the United States for its economy and military and that, ultimately, it would respond to pressure. Each country also discounted the alternatives available to the other. Johnson knew that Beijing could never provide the amount or quality of military support that Washington had to offer. What he misunderstood was that his arms embargo merely convinced Ayub that Pakistan needed a more reliable supplier, even if of secondary quality. Ayub, a confidant of the Shah, surely knew of the CIA intelligence-collection sites in Iran. There is no evidence, however, that he considered the possibility that the United States would rather shift its intelligence programs there than concede to his demands. Ayub also seems not to have realized that U.S. satellite advances might offer alternatives to the airborne collection platforms operating from Badaber.
The key to successfully optimizing gains from any bargaining scenario is understanding the limits of one’s leverage as well as having at least a basic grasp of the other side’s motivations. In the U.S.-Pakistani case, each country recognized that the alliance was in jeopardy. At the same time, the countries’ leaders saw the solution to the problem to be forcing the other to acquiesce. Each believed he held the upper hand and would ultimately prevail once the other was forced to recognize his dependence. Neither appears to have challenged these assumptions as the confrontation continued. Indeed, both Ayub and Johnson so completely identified their positions with questions of national sovereignty and pride that they undercut any possibility of compromise.
Such impasses can and do occur in traditional, publicly acknowledged alliances, but this case suggests that miscalculation may be a greater problem in secret relationships. The personal involvement of national leaders combined with receiving limited input from experts, the legislature, and the public offers additional opportunities for misunderstandings to occur. The fate of Kurmitula and the six unapproved bases in Lahore and Karachi suggests, too, that unwritten agreements give small state hosts additional leverage, because they have greater freedom to shut down the programs at any moment. The parallel example of Kamina Airfield, where Zairian leader Mobutu allowed the CIA to base its covert operations, would seem to support this argument. The CIA operated out of the airfield for six years, based only on Mobutu’s say-so, allowing Mobutu to bypass normal diplomatic channels to successfully press Reagan for ever more aid. This included successfully winning an “unprecedented” U.S. intervention with multilateral creditor institutions.196 The limited work to date on these secret and unwritten agreements, however, precludes a more general assessment.
These cases suggest the need for more work in investigating not only the negotiations involved in formally documented alliances but also in those pacts that are based on secret handshake agreements. There has been substantial work investigating how states use information to bargain in conventional alliances, but little focus to date on how these models apply to clandestine relationships. This review of the first decade of the U.S.-Pakistani intelligence liaison suggests that we have much to learn about how states gather and use information about their partners, particularly in cases where classification and compartmentation may limit input from experts. The misunderstandings between Ayub and Johnson may have been shaped by their individual personalities, yet they also reflect questions of how states update their assessments of their bargaining positions, something that remains relevant for today’s leaders.
Diana Bolsinger is the director of the University of Texas at El Paso’s graduate programs in Intelligence and National Security and a doctoral candidate at the LBJ School of Public Policy at the University of Texas at Austin. Before retiring from federal service, she served in a variety of analytic and policy support positions in the U.S. intelligence and foreign policy communities. An earlier version of this article was the winner of the 2020 Bobby Inman Award for Student Scholarship.