The Cold War was fought not only with weapons and diplomacy but also with spies, satellites, and secrets. Behind the public diplomacy of Washington and Moscow lay a second competition: an intelligence war waged by the Central Intelligence Agency and the KGB, by Britain’s MI6 and the West German BND, by the GRU and Mossad, and by a cast of agents, defectors, and double agents whose careers shaped policy in ways their governments rarely admitted. The U-2 Incident of May 1960, when Francis Gary Powers was shot down over Sverdlovsk, is the single moment when the secret contest exploded into the open — collapsing the Paris Summit and pushing the superpowers toward the most dangerous phase of the conflict.
The intelligence establishments
The American intelligence community in 1947 was a patchwork. The Office of Strategic Services had been dissolved in October 1945, and its functions were scattered across the War and State departments. The shock of the Soviet atomic test in August 1949, the loss of China in 1949–1950, and the outbreak of war in Korea in June 1950 forced a reckoning. The result was the National Security Act of July 26, 1947, which created both the National Security Council and, on September 18, 1947, the Central Intelligence Agency. The CIA was given two jobs: coordinate the rest of the community and provide the President with independent analysis. The National Security Agency, founded in 1952, took over signals intelligence. The Defense Intelligence Agency, created in 1961, consolidated military intelligence. The National Reconnaissance Office, established secretly in 1961, ran the spy satellite programs.
The Soviet system was organized differently. Foreign intelligence, counterintelligence, internal security, border guards — the suppression of political dissent all sat under one roof: the Cheka, which became the GPU, then the NKVD, then the MGB, and finally, in March 1954, the KGB. The GRU — military intelligence — retained its own separate chain of command to the General Staff. The KGB’s combination of foreign espionage and domestic terror gave it a weight inside the Soviet state that the CIA, constrained by law from operating at home, never possessed. As Christopher Andrew notes in The Sword and the Shield (1999), the KGB’s files were simultaneously instruments of statecraft and dossiers on the regime’s own citizens.
Britain’s Secret Intelligence Service (MI6) and the West German Bundesnachrichtendienst (BND, founded 1956) were the most important Western partners. France’s SDECE and Israel’s Mossad ran their own parallel wars. The great majority of covert action in the early Cold War was an Anglo-American enterprise, with the British providing cover and in-country networks the United States could not build quickly.
Covert action: Iran, Guatemala, Cuba, and beyond
The defining feature of the early Cold War CIA was political action against governments the United States disliked. The pattern was set in 1953. Operation Ajax, run by Kermit Roosevelt Jr. of the CIA’s Near East Division in concert with MI6, overthrew Prime Minister Mohammad Mosaddegh of Iran and restored Shah Mohammad Reza Pahlavi. A year later, Operation PBSUCCESS toppled the reformist government of Jacobo Árbenz in Guatemala. The two operations established the CIA as an instrument of regime change, training thousands of case officers and giving the agency a mystique — and an arrogance — that culminated in disaster at the Bay of Pigs in April 1961.
The Cuban operation had been conceived under Eisenhower. A force of roughly 1,500 Cuban exiles, Brigade 2506, landed at the Bay of Pigs on April 17, 1961. The expected popular uprising never came, the promised air cover was canceled, brigade was routed within 72 hours. The political fallout was severe: it pushed Havana into Moscow’s arms, and eighteen months later the Soviets placed nuclear missiles in Cuba. As Tim Weiner writes in Legacy of Ashes (2007), the failure was less a tactical accident than a product of an organizational culture that had stopped telling its political masters the truth.
Other major operations followed: the Congo (1960–1965), where the CIA tracked and may have helped locate Patrice Lumumba; the secret war in Laos (1962–1975); the attempted assassinations of Fidel Castro, organized by the CIA’s chief of covert operations, William Harvey, in partnership with the Mafia; the CIA’s involvement with the Chilean military before the 1973 coup against Salvador Allende; and the long, illegal support for the Nicaraguan Contras that produced the Iran-Contra affair in 1986. By 1975, when the Church Committee opened its hearings, the catalog of abuses was long enough to require fourteen published reports.
Spies, defectors, and double agents
The other half of the intelligence war was human: traitors, heroes, and the difficult business of telling them apart. The British side suffered the worst penetration of any Western service. Kim Philby, Donald Maclean, Guy Burgess, Anthony Blunt, and John Cairncross — the Cambridge Five — handed the Soviet Union years of Anglo-American secrets, including, via Fuchs and others, the design of the atomic bomb. On the American side, Aldrich Ames turned himself over to the KGB in 1985, in exchange for $2.7 million, and gave up at least ten CIA sources, several of whom were executed. Robert Hanssen spied for the KGB and SVR from 1979 to 2001. On the Soviet side, Oleg Penkovsky, a GRU colonel, ran for MI6 and the CIA in 1961–1962 and gave Washington the intelligence that helped it read Khrushchev’s hand during the Cuban Missile Crisis. Yuri Nosenko, a KGB officer who defected in 1964, told the West that Lee Harvey Oswald was not a KGB agent — a claim the CIA’s Soviet division refused to believe for years.
Venona, the secret American program that decrypted Soviet cables from 1943 to 1980 (with results first published in 1995), reshaped the historiography of the early Cold War. It confirmed that the Rosenbergs’ network had been real, that the Cambridge Five were not a fever dream, and that Soviet penetration of the Manhattan Project was substantial. The cables also revealed that Alger Hiss, the subject of one of the defining trials of the McCarthy era, was probably an agent — although the Hiss case, like the Rosenberg case, still produces argument.
Technical intelligence: aircraft and satellites
The U-2, designed by Kelly Johnson’s Skunk Works at Lockheed, first flew in 1955. Its first Soviet overflight came on July 4, 1956. By the time Powers was shot down, the U-2 had mapped Soviet bomber bases, identified ICBM facilities at Plesetsk and Tyuratam, and given the President the photographic basis for the missile-gap debate. After 1960, the program continued but was politically constrained. The SR-71 Blackbird, which entered service in 1966, could outrun and out-climb anything the Soviets had, but it could not cover as much ground as the U-2 had.
The real answer to the U-2’s vulnerability was orbital. The Corona program, run jointly by the CIA and the Air Force, achieved its first successful recovery on August 19, 1960, three and a half months after the Powers shootdown. Over 145 missions, Corona returned more than 800,000 frames covering 1.5 million square miles of denied territory. The imagery resolved the missile gap (there was none), verified Soviet compliance with arms control, and — critically — provided the U-2 evidence of Soviet MRBMs in Cuba on October 14, 1962. By the early 1970s, film-return capsules had given way to digital transmission. The U-2 itself was rebuilt for tactical use; the SR-71 was retired in 1998; satellites had become the default.
Signals intelligence and codebreaking
Signals intelligence (SIGINT) deserves its own weight in the balance. The American-British BRUSA agreement of 1943 and the UKUSA agreement of 1946 stitched together a global listening partnership that survives in the Five Eyes arrangement. The NSA, with its enormous station network, processed everything from diplomatic cable traffic to telemetry. The British GCHQ performed the same role for London.
The crowning SIGINT achievement of the early Cold War was Venona. Soviet cable traffic from 1943 to 1945 had been encrypted with one-time pads. The Soviets, desperate for keys in the chaos of 1941–1942, reused some pages of the pads — a fatal error. American and British codebreakers, working at Arlington Hall and Bletchley Park’s successor, noticed the duplication in 1946 and began the long, painstaking process of recovery. By the late 1940s, they were reading traffic in volume. The program stayed classified until 1995; its disclosures are now the foundation of any honest account of Soviet espionage in the West.
Costs and consequences
The intelligence war cost lives and money on a scale the public rarely appreciated. The CIA’s MKULTRA program, run from 1953 to 1973, dosed unwitting American and Canadian subjects with LSD and other drugs — a fact that became public only during the Church Committee hearings. The FBI’s COINTELPRO disrupted civil rights, antiwar, and socialist organizations for two decades. The KGB ran its own equivalent: Operation INFEKTION, the AIDS-disinformation campaign of the 1980s, and the suppression of dissidents from Sakharov to Sharansky.
The end of the Cold War did not end the intelligence contest. The collapse of the Soviet Union in December 1991 produced a brief sense of victory; the appointment of Vladimir Putin, a former KGB officer, to the Russian presidency in 1999, and the return of large-scale Russian operations against Western targets in the 2010s, made clear that the contest had merely changed form. The agencies built to fight the Cold War — the CIA, the KGB’s successors SVR and FSB, MI6, the BND, and their partners — are still in business. The tradecraft has migrated online. The tradecraft itself has not.