Crisis of Command
America’s Broken Civil-Military Relationship Imperils National Security
It didn’t take long for Congo’s transition from Belgian colony to sovereign state to turn ugly. Both the Soviet Union and the United States were keeping a close eye on the mineral-rich country at the heart of Africa when, on June 30, 1960, it gained independence under a democratically elected government headed by Prime Minister Patrice Lumumba. A charismatic nationalist, Lumumba led the only party in parliament with a nationwide, rather than ethnic or regional, base. Within days, however, Congo’s troops mutinied against their all-white officer corps (a holdover from the colonial era) and started terrorizing the European population. Belgium responded by sending forces to reoccupy the country and helping Congo’s richest province, Katanga, secede. The United States, declining the appeals for help from the new Congolese government, instead threw its support behind a UN peacekeeping mission, which it hoped would obviate any Congolese requests for Soviet military assistance. But Lumumba quickly came into conflict with the UN for its failure to expel the Belgian troops and end Katanga’s secession. After issuing a series of shifting ultimatums to the UN, he turned to Moscow for help, which responded by sending transport planes to fly Lumumba’s troops into Katanga.
That’s when the Eisenhower administration sent in the CIA. In the decades that followed, the dominant narrative in U.S. foreign policy circles portrayed the U.S. covert action in Congo as a surgical, low-cost success. Even the 1975 U.S. Senate investigation by the Church Committee, which was heavily critical of the CIA, concluded that of the five covert paramilitary campaigns it studied, the operation in Congo was the only one that “achieved its objectives.” Those who hold this view credit the U.S. government with avoiding a direct military confrontation with the Soviet Union and China while foiling the communists’ attempts to gain influence over a key African country. They acknowledge that the CIA contributed to the fall of Lumumba, who lost a power struggle with Joseph Mobutu, the pro-Western head of Congo’s army, in September 1960. But they maintain that even though the CIA plotted to assassinate Lumumba -- once even trying to get a recruit to poison his toothpaste or food -- it never did so, and had no hand in his eventual murder, in January 1961. They also recognize the agency’s contribution to the military defeat of Lumumba’s followers. As for Mobutu, who would go on to become one of Africa’s most enduring and venal leaders, proponents of the orthodox account argue that his faults became clear only later, many years after CIA involvement had run its course.
Over the years, many scholars and journalists have challenged parts of this orthodoxy, and public perception has begun to catch up. But their case has been hampered by the shortage of official documentary evidence. Recently, however, new evidence has become available, and it paints a far darker picture than even the critics imagined. The key sources include files from the Church Committee, which have been slowly declassified over the last 20 years; a 2001 Belgian parliamentary investigation into Lumumba’s murder; and a new volume, released last year, of Foreign Relations of the United States, the State Department series that presents a document-by-document record of U.S. decision-making. The new volume, on Congo, contains the most extensive set of CIA operational documents ever published.
We now know that even though the threat of communism in Congo was quite weak at the time of Congo’s independence, the CIA engaged in pervasive political meddling and paramilitary action between 1960 and 1968 to ensure that the country retained a pro-Western government and to help its pathetic military on the battlefield. So extensive were these efforts that at the time, they ranked as the largest covert operation in the agency’s history, costing an estimated $90–$150 million in current dollars, not counting the aircraft, weapons, and transportation and maintenance services provided by the Defense Department. The CIA had a hand in every one of Congo’s major political turning points during the period and maintained a financial and political relationship with every head of its government. And contrary to the conclusion of the Church Committee, Lawrence Devlin, the CIA station chief in Congo for most of the period, had direct influence over the events that led to Lumumba’s death.
Not only was U.S. involvement extensive; it was also malignant. The CIA’s use of bribery and paramilitary force succeeded in keeping a narrow, politically weak clique in power for most of Congo’s first decade of independence. And the very nature of the CIA’s aid discouraged Congolese politicians from building genuine bases of support and adopting responsible policies. The agency’s legacy of clients and techniques contributed to a long-running spiral of decline, which was characterized by corruption, political turmoil, and dependence on Western military intervention. So dysfunctional was the state that in 1997 it outright collapsed -- leaving behind instability that continues to this day.
In the beginning, U.S. covert action in Congo was exclusively political in nature. Washington worried that Lumumba was too erratic and too close to the Soviets and that if he stayed in power, Congo could fall into further chaos and turn communist. Allen Dulles, the director of the CIA, cabled the CIA station in Léopoldville, the capital, in August 1960: “We conclude that his removal must be an urgent and prime objective and that under existing conditions this should be a high priority of our covert action.” So the CIA station, in tandem with Belgian intelligence officials, subsidized two opposition senators who attempted to organize a vote of no confidence against Lumumba’s government. The plan was for Joseph Kasavubu, Congo’s president and Lumumba’s rival, to dissolve the government after the vote and nominate one of the senators as the new prime minister. The CIA also funded anti-Lumumba street demonstrations, labor movements, and propaganda.
But Kasavubu, encouraged by the Belgians, jumped the gun and publicly fired Lumumba two days before the vote was to be held. Lumumba responded by refusing to withdraw and continuing to dominate parliament, which would have to approve a new government. Devlin quickly found a solution to the stalemate in Mobutu, the 29-year-old army chief of staff. In two meetings, Mobutu told Devlin that he was moving troops to the capital and pleaded for U.S. help in acting against Lumumba. Devlin agreed to finance his efforts, subsequently telling CIA debriefers that, as the new Foreign Relations of the United States volume puts it, “this was the beginning of the plan for Mobutu to take over the government.” On September 14, Mobutu announced that he was suspending parliament and the constitution. He sacked Lumumba and kept on Kasavubu, but now Mobutu was the power behind the throne.
The CIA rushed to his side with more money, warnings about assassination plots, and recommendations for ministerial appointments. It counseled Mobutu to reject reconciliation with Lumumba and instead arrest him and his key associates, advice Mobutu readily accepted. Devlin became not just the paymaster but also an influential de facto member of the government he had helped install. His principal vehicle was the so-called Binza Group, a caucus of Mobutu’s political allies that got its name from the Léopoldville suburb where most of them lived. It included Mobutu’s security chief and his foreign and finance ministers. In the months after the coup, the group consulted Devlin on major political and military matters, especially those dealing with Lumumba, who was now under house arrest but protected by UN troops.
The group almost always heeded Devlin’s advice. In October, for example, Mobutu threatened to expand his power by firing President Kasavubu -- which would have deprived the government of its last shred of political legitimacy. So Devlin persuaded him to accept a compromise instead, under which Mobutu would work with a council of associates -- all paid by the CIA -- that would choose cabinet ministers for Kasavubu and control parliament. Devlin also convinced the Binza Group to drop a risky plan to attack Lumumba’s UN security detail and arrest Lumumba.
On January 14, 1961, Devlin was informed by a government leader that Lumumba, who had escaped from UN protection and been captured by Mobutu’s troops, was about to be transferred to the Belgian-backed secessionist province of South Kasai, whose leader had vowed to murder him. In his subsequent, January 17 cable reporting this critical contact to CIA headquarters, Devlin gave no indication that he had voiced any opposition to the plan. Given his intimate working relationship with Congo’s rulers and his previous successful interventions with them concerning Lumumba, Devlin’s permissive stance was undoubtedly a major factor in the government’s decision to move Lumumba.
But Devlin did more than give a green light to the transfer. He also deliberately kept Washington out of the loop -- an exception for a covert program that was being closely managed by the CIA, the State Department, and the National Security Council. On the same day that he was informed of Lumumba’s prospective transfer, Devlin learned that the State Department had denied his and CIA headquarters’ urgent request for funds to pay off a key Congolese garrison on the verge of a mutiny that threatened to restore Lumumba to power. John F. Kennedy was to take office in six days, and the State Department considered the request “one of high policy” that should wait for the new administration to decide.
Seeing his preferred method for preventing Lumumba’s comeback blocked, Devlin may have viewed the impending transfer as a promising Plan B. But he also knew that if he told headquarters about the plan, it would consult the State Department, which, given its response to his last request, would almost certainly have considered the U.S. position on the transfer a matter for the incoming administration. All of that meant that if Washington had been fully informed about the plot, it might well have tried to put the brakes on it through Devlin, the Binza Group, and their Belgian advisers. Moreover, Devlin knew that the Kennedy transition team was reconsidering the Eisenhower administration’s hard-line policy toward Lumumba. So even as he communicated with headquarters about other matters, Devlin withheld information about the planned transfer for three days, until the move was already under way. In a last-minute switch, Lumumba was sent to Katanga, the other Belgian-supported secessionist province, whose powerful interior minister had repeatedly called for his scalp. By the time Devlin’s January 17 cable arrived in Washington, Lumumba had been shot dead in Katanga.
Rather than end the struggle for control of Congo, Lumumba’s assassination only intensified it. In August 1961, the United States, under pressure from the UN and a pro-Lumumba state in eastern Congo, agreed that the Congolese parliament should reconvene to select a new national government. But the CIA used bribes to ensure that the new government was led by its ally Cyrille Adoula. While the resulting power-sharing deal did include some Lumumbists, as Lumumba’s supporters were called, the most important positions went to members of the Binza Group (with Mobutu himself remaining head of the army).
Once Adoula was in office, the CIA provided him with a public relations firm to help him bolster his image abroad and an adviser who wrote speeches for him. The CIA also bribed parliament, the Binza Group, a labor union, and an organization of tribal chiefs to back the new leader. Meanwhile, Devlin continued to behave like a member of the government. At the Binza Group’s behest, he persuaded Adoula not to make concessions to his Lumumbist deputy prime minister. When Adoula decided to fire Mobutu, Devlin convinced him to drop the idea. Adoula even asked Devlin to canvas political leaders in order to gauge his own parliamentary support. In November 1961, after only a year and a half on the job, Devlin cabled CIA headquarters that the agency could “take major credit for the fall of the Lumumba [government], the success of the Mobutu coup and considerable credit for Adoula’s nomination as premier.”
Adoula’s government didn’t perform as well as Washington had hoped: soldiers were forced to live off the land, avaricious officials looted the Treasury, and inflation sapped the incomes of everyone else. After Adoula removed nearly all his Lumumbist ministers and dissolved parliament in 1963, the Lumumbists returned to their home provinces and took to arms. By early 1964, their rebellion had swept across almost half the country. Alarmed by the insurgency, the Binza Group and Kasavubu decided to replace Adoula with someone they thought would deal with it more effectively: Moise Tshombe, the former secessionist leader of Katanga, whose breakaway government had murdered Lumumba in 1961. The CIA acquiesced to the change, adding tribal supporters of Tshombe and other key politicians to its existing payroll. It also added a major paramilitary thrust to its political program in Congo.
The agency endowed Tshombe’s new government with an “instant air force” to defeat the rebels, who were then receiving modest advisory and financial assistance from the Chinese. The unit, composed mainly of American planes piloted by Cuban exiles, enabled the advance of white mercenaries (predominantly South Africans and Rhodesians) who were leading the Congolese government forces. In August 1964, a National Security Council committee had signed off on a plan for 41 combat and transport aircraft and almost 200 personnel (Cuban air crews and European ground maintenance workers). In early 1965, the CIA added a small navy, also staffed by Cubans, to the mix to hamper shipments of military supplies to the rebels from neighboring Tanzania across Lake Tanganyika.
Washington was joining a particularly bloody conflict. When they seized rebel-held areas, the white mercenaries and government forces indiscriminately slaughtered the rebels and civilians they found there. Although there was no systematic counting of the casualties, it is estimated that at least 100,000 Congolese perished during this phase of the war. The insurgents killed about 300 Americans and Europeans whom they had taken hostage following the fall of Stanleyville, the rebel capital.
By the fall of 1965, the Congolese army and its foreign helpers had largely succeeded in regaining control of the country, but another threat loomed: increasing political competition between President Kasavubu and Prime Minister Tshombe. Both the U.S. government and the Binza Group feared that the conflict between the two men could cause one of the contenders to look for support from the more radical African regimes. As the crisis reached its apogee, Mobutu told Devlin that he was considering launching another coup, to replace both Kasavubu and Tshombe, or finding some other unidentified solution. On November 22, the United States responded by increasing CIA financing for Mobutu’s officers and giving Mobutu carte blanche to act as he saw fit.
Within three days, Mobutu bloodlessly seized power, a result that Devlin called “the best possible solution.” The CIA responded with still more money, which Mobutu used to pay off key officers, political leaders, and tribal chiefs. Throughout 1966 and 1967, the agency forwarded Mobutu intelligence about threats to his regime, uncovering a number of major plots (one of which ended with the public hanging of the alleged conspirators). And the CIA’s covert air force, along with overt transportation help from the Pentagon, helped Mobutu fend off two mercenary-led mutinies.
In October 1966, Mobutu threw out the U.S. ambassador for failing to show enough respect for his newly elevated status and stopped requesting his monthly CIA stipend. Two years later, Mobutu changed his mind and asked the CIA for more money -- which he got. By then, the CIA had wrapped up its paramilitary program and limited its political funding to four key people other than Mobutu. From the U.S. perspective, with no more legal opposition to control and no more Lumumbist rebels or pro-Tshombe mercenaries to fight, Congo could be transitioned to purely overt U.S. military and economic assistance.
Unfortunately, the full picture of the CIA’s involvement in Congo remains partly obscured. Concerned about protecting its sources and methods, the agency managed to delay the publication of the new volume of Foreign Relations of the United States for over a decade. And the version that was finally released takes an overly cautious approach to redactions, withholding four documents in their entirety, cutting 22 by more than a paragraph, omitting the financial costs of specific activities, and attempting to guard the identities of the CIA’s key Congolese clients besides Mobutu. Five decades after the events in question, most of these excisions seem hard to justify, especially given that historians, journalists, and even Devlin himself have already exposed the main actors’ identities.
Still, it is clear that the CIA programs of the 1960s distorted Congolese politics for decades to come. This is not to argue that in the absence of U.S. meddling, Congo would have established a Western-style representative government. But even in a region with plenty of autocracies, the country has stood out for its extreme dysfunction. Ever since the CIA’s intervention, Congo’s leaders have been distinguished by a unique combination of qualities: scant political legitimacy, little capacity for governing, and corruption so extensive that it devours institutions and norms. In the years following U.S. covert action, these qualities led to economic disaster, recurrent political instability, and Western military intervention. Finally, in 1997, rebels headed by a former Lumumbist and backed by military forces from Rwanda, Uganda, and Angola sent Mobutu packing, leading to a regional war that would kill more than three and a half million people over the next decade.
Of course, the main author of all of this misrule was Mobutu. But given that he would never have been able to consolidate control were it not for the CIA cash he distributed to his allies, as he himself admitted to the agency, the United States must bear some responsibility for what Mobutu wrought. Furthermore, the CIA’s predominant techniques -- corruption and external force -- constituted a tutorial on irresponsible governance. Weaned on the agency’s bribery, Mobutu and his associates never had to compete for the affection of the broader public and develop a real political base and had no incentive to put the state’s resources to good use. And because Mobutu could depend on the CIA’s paramilitary support, he felt no pressure to develop even a minimally capable military. In fact, even though he managed in the chaos of independence to be appointed army chief of staff, he was an incompetent military leader. By 1964, his army had, according to Averell Harriman, the U.S. undersecretary of state for political affairs, proved its “worthlessness,” being incapable of securing key territory without help from foreign mercenaries. What Mobutu was immensely talented at, of course, was the skill that the Americans had taught him: wheedling bribes. Twice, he even persuaded Devlin to reimburse him for army funds that he claimed to have used for unauthorized expenses or CIA objectives, arguing that if rivals discovered the misuse, they might charge him with corruption.
U.S. officials outside the CIA learned of Mobutu’s flaws early on. Following the 1965 coup, a State Department memorandum cautioned, “It is too early to discern where Mobutu will draw the line between corruption and the ‘normal’ use of payments and patronage to facilitate government operations.”
During the white mercenary mutinies of 1966 and 1967, U.S. cables and memorandums were scathing. A National Security Council memo to the White House chief of staff described Mobutu as “somewhat inept and his chances of pulling the Congo up by its bootstraps are indeed remote.”
The U.S. ambassador to Congo, Robert McBride, labeled Mobutu “irrational” and “highly unstable.” President Lyndon Johnson’s national security adviser at the time, Walt Rostow, called him “an irritating and often stupid” man who “can be cruel to the point of inhumanity.” In 1968, McBride sent a cable to the State Department that took note of the president’s new luxury airplane, plan for parks modeled on Versailles, thoughts of building a replica of Saint Peter’s Basilica in each of Congo’s three largest cities, and acquisition of a Swiss villa. McBride concluded,
I believe there is nothing which can be done to restrain these frivolous Presidential expenditures because Mobutu has apparently risen in souffle-like grandiloquence. I feel that to call his attention to the dangers of this type of thing . . . would be to incur instant wrath.
However, I felt a brief report should be made on [this] regrettable phenomenon because I believe it is the most serious problem facing Congo at present time and the fault is that of the President and the uncontrollable spending is emanating directly from him. Furthermore, it occurred to me this might have an effect on US policies towards the present regime in the Congo.
What McBride seemed not to realize was that eight years of covert action had done much to rule out any alternative U.S. policy, then or ever. The CIA had not only fostered a regime; it had stamped it “made in America” for future policymakers in Washington. As Mobutu’s government lurched from crisis to crisis, it continued to enjoy U.S. and Western financial and military help. Over the years, many in Congress and some dissidents in the State Department did urge the U.S. government to push for economic and political reforms in the country that Mobutu had renamed Zaire in 1971. Failing that, they said, it should distance itself from Mobutu and cultivate political ties with the opposition. When the Cold War ended, Congress finally cut off military and nonhumanitarian assistance. Yet even afterward, as the regime entered its death throes, U.S. officials could not bring themselves to abandon it and support the peaceful democratic transition proposed by the rising opposition.
Clinging to a longtime friendly dictator, even as his flaws become more risky for U.S. interests, is a well-known pathology of U.S. foreign policy. In the case of Congo, the relationship had been created and nurtured by CIA covert action. This endowed it with a special aura of intimacy, visible in the possessive language that U.S. officials used when referring to Mobutu. For Devlin, Mobutu became “almost our only anchor to the windward.” During the escalating battle between Kasavubu and Tshombe, Harold Saunders, a member of the National Security Council staff, wrote that Mobutu should be the one to resolve the conflict -- by military means if necessary -- because “he is already our man.” Ten years after the subsequent coup, Edward Mulcahy, the deputy assistant secretary of state for African affairs, testified in Congress, “We do have . . . a warm spot in our hearts for President Mobutu. At a time when our aid and advice were critical for the development of Zaire, he was good enough -- and I might say wise enough -- to accept our suggestions and our counsel to the great profit of the state.”
Like other such questionable commitments, the United States’ long support for Mobutu was rationalized as necessary because there was no alternative but chaos. In reality, Washington squandered opportunities to push for major reforms. After Congolese exiles from Angola unsuccessfully invaded Zaire twice in the late 1970s, the United States failed to use the leverage provided by the resulting Western military intervention to seek a more inclusive government. During the opposition ferment that swept Zaire in the 1980s, it refused to support the popular demand for a second party. Even when a strong democracy movement compelled Mobutu to make political concessions in the early 1990s, the George H. W. Bush administration prevented Herman Cohen, its assistant secretary of state for African affairs, from calling for Mobutu’s resignation after Mobutu reneged on his commitments. And although the Clinton administration banned visas for Mobutu’s associates, it also endorsed his laughable plan for “free elections.”
Covert action produced a Congolese government that largely supported U.S. foreign policy, but it burdened U.S. diplomacy in Africa for decades. In particular, the overthrow and murder of Lumumba and the support for Tshombe’s white mercenaries angered African nationalists and soured U.S. relations with many key countries, including Algeria, Ghana, Kenya, and Tanzania; these actions also antagonized liberation movements in Angola, Mozambique, South Africa, and Zimbabwe. The resentment and suspicion that the CIA’s program in Congo engendered subsided slightly as the agency’s involvement there declined, but they never disappeared, and they would resurface throughout the 1970s and 1980s whenever the West (and in particular the CIA) intervened in the region.
The root of the CIA’s intervention in Congo was an overhyped analysis of the communist threat. Congo scholars have long been skeptical of the notion that had Lumumba stayed in power, his government would have fallen under the sway of the Soviet Union or China. At the time, even some U.S. officials had doubts. In 1962, shortly after he retired as director of the CIA, Dulles admitted, “I think that we overrated the Soviet danger, let’s say, in the Congo.” The Kennedy administration’s initial policy paper, soon modified, advocated a broad-based government of “all principal political elements in the Congo,” to be followed by the release of Lumumba. Even at the height of the rebellion, in 1964, National Security Adviser McGeorge Bundy wrote Johnson, “What is very unclear is how deep the Chinese hand is in the rebel efforts. Harriman thinks it is pretty deep; most of the intelligence community thinks it is more marginal.” In November 1964, Michael Hoyt -- the U.S. consul in rebel-held Stanleyville, who had just been released from over three months of captivity -- informed policymakers that the leaders of the Lumumbist insurgency were “within the Congolese political spectrum” and that they were “essentially pragmatic and followed their own interests.”
The skeptics were right: Lumumba was never a communist, and he would not have yielded to foreign control. He and his supporters had cut their political teeth in the struggle against colonialism, and they found any form of external domination anathema. They were far more interested in nonalignment, and the foreigners they identified with were other African independence leaders, not Khrushchev or Mao. Lumumba and his followers also understood that the communist world could never replace the massive European investment and 10,000 Belgian technicians that served as the foundation for Congo’s Western-oriented economy. Even when they accepted Soviet military assistance to help reunify their country or contest their political exclusion, they continued to appeal for support from the United States, the rest of the West, and other African countries. Yet Washington refused to help.
Archives from the former Soviet bloc confirm that although Moscow was eager to squeeze every propaganda advantage it could from the West’s difficulties in Congo, it understood that Lumumba and his followers were no Marxists, and it hedged its support for them accordingly. Following Mobutu’s 1960 coup, Moscow meekly withdrew its airplanes and military advisers from the country and did nothing to help Lumumba. It provided little aid to his successors until Lumumba’s assassination and the capture of Stanleyville by white mercenaries outraged the rest of Africa. Even then, the Soviet Union dispatched arms but no advisers to teach the recipients how to use them. Soviet and Chinese military assistance were also constrained by the need to secure transport rights through neighboring African states, which was not always forthcoming.
In retrospect, it is clear that the U.S. officials directing Congo policy inappropriately projected their Cold War experiences in Europe, Asia, and Latin America onto Africa, where the conditions were completely different. In Congo, there had been no Soviet military occupation and no significant Marxist or communist party or cadres. Tragically, Washington spurned an alternative policy: engaging diplomatically with Lumumba and his successors as part of a broad effort to keep the Cold War out of Congo. Instead, it annointed Mobutu and other members of the Binza Group as Belgium’s heirs. Impatient and inexperienced as he was, Lumumba represented his country’s best hope for a successful postcolonial era. There is every reason to believe that working with him and other incipient democratic forces would have better served both the United States and Congo.