Spying on the Bomb: American Nuclear Intelligence from Nazi Germany to Iran and North Korea 
by Jeffrey Richelson.
Norton, 702 pp., £22.99, April 2006, 0 393 05383 0
Show More
Show More

Jeffrey Richelson is an expert on the American secret intelligence agencies, particularly on their peculiar devotion to spying without spies – their reliance on aerial or satellite imagery, intercepted communications, seismic and acoustic detection of nuclear bomb explosions, and other esoteric means of surveillance. Richelson’s politics are completely conventional. He sees the ‘West’, led by the United States, as being on the side of good in the world and its efforts to detect nuclear weapons in the hands of its adversaries as ingenious attempts to disarm what George W. Bush would call ‘evil-doers’. Richelson concludes his book by denying that ‘political leaders, including the president, dictated the content of the [National Intelligence] estimates to provide a “pretext for war” [in Iraq] or to “hoodwink” the American public.’ That proposition is not substantiated in this volume.

Spying on the Bomb is valuable nonetheless. Its 121 pages of source citations are a massive compendium of information on the nuclear weapons programmes of Nazi Germany, the USSR, China, France, India, Pakistan, Israel, South Africa, Iraq and Iran, as well as on the successes and failures of official US efforts to find out what each of these countries was doing. Richelson’s primary sources are highly classified CIA National Intelligence Estimates (NIEs), which he has been able to read in censored form in the National Archives, or through successful requests using the Freedom of Information Act. He does not go into detail on how he achieved this or where these documents can be found by other researchers today, but thanks to the FOIA, he obtained relevant documents from the CIA, the Defense Intelligence Agency, the National Reconnaissance Office, the State Department, the Air Combat Command and the Los Alamos National Laboratory. Most important, Richelson is a senior fellow at the National Security Archive in Washington, the pre-eminent American organisation researching the Cold War and US foreign policy, whose work is based primarily on an aggressive use of the FOIA. The National Security Archive has a well-deserved reputation for bringing suit to compel the government to divulge information that it would prefer to keep secret. Richelson is at the heart of this great tradition.

He has set himself too big a task, but he performs it as well as one could expect. No single human being could possibly master the massive files on nuclear weapons, from World War Two to Ahmadinejad’s Iran. In addition to the CIA estimates, Richelson relies primarily on standard works – John W. Lewis and Xue Litai’s China Builds the Bomb (1988), George Perkovich’s India’s Nuclear Bomb (1999), Seymour Hersh’s The Samson Option: Israel’s Nuclear Arsenal and American Foreign Policy (1991) – which he supplements with memoirs and establishmentarian or right-wing journalism. One won’t find references here to Bruce Cumings on North Korea’s bomb-making, or Wen-Ho Lee’s account of his alleged espionage for China from inside Los Alamos – or the racism of the FBI and the Department of Energy that lay behind Lee’s ordeal. But these are not fatal limitations.

Richelson is primarily interested in the means the US invented over the past sixty years to detect a nuclear weapons programme, from its developing stages up to and including testing, without having a human spy to witness it and tell us about it. He is frank about the numerous instances in which the whole detection apparatus failed, as it did in the first Soviet test of August 1949; the ‘Vela’ explosion in the South Atlantic in September 1979; the detonation by India of several devices, one thermonuclear, at Pokhran in May 1998; and Saddam Hussein’s alleged possession of weapons of mass destruction. Richelson presents each of these cases with his usual deep immersion in detail – sometimes failing to produce much in the way of insight or analysis.

During World War Two, the United States and Britain were very keen to find out what Germany was doing in the nuclear weapons field, not just because it was the enemy but because it was the home of many of the world’s greatest physicists, the country where Robert Oppenheimer and others had gone to do graduate work, and where fission had been discovered. In December 1938, Otto Hahn, Germany’s leading radiation chemist, and his student Fritz Strassmann, profiting from the experimental work of Irène and Frédéric Joliot-Curie and Enrico Fermi, had cautiously reported their success in splitting the uranium atom. In exile in Sweden because of her Jewish ancestry, Lise Meitner, Hahn’s former colleague and longtime collaborator, explained in the 11 February 1939 issue of Nature that what Hahn and Strassmann had discovered was the ‘fission’ of the uranium nucleus. In August, Einstein wrote from Peconic, Long Island to alert President Roosevelt to the threat of the new discovery. An explosive chain reaction of the isotope uranium-235, he wrote, might lead to ‘extremely powerful bombs of a new type’. Given the origins of these developments, Roosevelt and his military advisers urgently sought intelligence on what Germany would do to develop such weapons.

The first focus of analysis was to identify, locate and follow the work of leading physicists, above all Werner Heisenberg’s: Heisenberg, the winner of the 1932 Nobel Prize, was said to be the head of Hitler’s atom bomb effort. He had also been Edward Teller’s PhD supervisor at Leipzig in 1930. In December 1944, the OSS, the CIA’s wartime predecessor, ordered one of its agents in Switzerland, Morris ‘Moe’ Berg, to attend a Heisenberg lecture in Zurich and to carry a pistol. He was to listen closely to what Heisenberg had to say and if it led him to believe that Germany was close to creating an atom bomb, Berg was to shoot him, in the auditorium. As it turned out, Berg was barely able to understand what Heisenberg was talking about – strong interactions among particles – and did not kill him. Instead, he went to dinner with him and other faculty members, where Heisenberg helpfully remarked that Germany had already lost the war.

In the only instance in this long book in which Richelson openly reveals his own opinion, he adds in an endnote: ‘For what it’s worth, I do not believe that Heisenberg understood that an atomic bomb could be built but sought to sabotage, along with others, the German effort.’ If that’s all Richelson wants to say about this most controversial of matters, he probably shouldn’t have said anything. The eavesdropping on Heisenberg and other German atomic physicists who were held after the war at Farm Hall, near Cambridge, may be the reason for Richelson’s scepticism. In his conversations, Heisenberg seems seriously to overstate the amount of uranium necessary to build a deliverable bomb. Some historians believe he knew that he was being monitored and that what he said can’t be trusted. In any case, his obiter dictum does nothing to settle the many questions surrounding Heisenberg’s role in the Nazi effort. The only thing we know for certain is that Germany did not produce a chain reaction and did not discover an effective means of enriching uranium.

The wartime surveillance effort inaugurated the US government’s attempt to detect an atomic weapons programme through aerial collection of radioactive debris. One day in 1944, at Los Alamos, Leslie Groves, the director of the Manhattan Project, called in the young physicist Luis Alvarez to ask him if there was some technological means to determine if the Germans were operating plutonium-producing reactors. Alvarez invented a filtering device to be carried in the nose of an aircraft to detect xenon-133, a gas emitted by operating reactors. In the remaining days of the war, Allied pilots flew this filter over Germany, trying – and failing – to find xenon. In 1951 and 1952, intelligence officials revived Alvarez’s approach in their attempts to monitor the Soviets. They sought to detect another gas that doesn’t occur naturally in the atmosphere, krypton-85, which is released by the fission of uranium-235 in a reactor. Krypton-85 was considered a better marker because the amount of it released into the atmosphere is directly proportional to the amount of plutonium being produced in the reactor. Aerial attempts to find krypton are still being used in monitoring North Korea’s bomb programme.

The detection effort directed against the Soviet Union was very different from the one directed against Germany, if for no other reason than that the Soviet activities occurred after the attacks on Hiroshima and Nagasaki. The Americans had already made some progress in the aerial collection of debris, following the detonation of two bombs at Bikini Atoll in the Pacific in July 1946, and three blasts at Eniwetok in April and May 1948. They did not yet know that Klaus Fuchs, a German émigré physicist working in the British programme, had supplied the USSR with the design of its plutonium implosion weapon, which the Russians first tested on 29 August 1949, at Semipalatinsk in the Kazakhstan desert.

Well before the first Soviet test, the US was flying B-29s on weather reconnaissance missions to filter airborne dust moving east from any point in the Soviet Union. After 111 false alerts, blamed on natural phenomena, alert number 112 turned out to be the real thing. Despite the CIA’s failure to provide any advance warning of the test, the air force unambiguously detected the fall-out, and on 23 September, President Truman publicly announced that America’s nuclear monopoly had been breached.

The Cold War now entered its nuclear phase. Both sides decided to go for Teller’s ‘super’ – i.e. hydrogen – bomb, a decision that resulted in a spewing of radioactive fallout all over the globe. In November 1952, the US tested a ten-megaton weapon that totally obliterated the Pacific island of Elugelab, replacing it with a crater about two hundred feet deep and a mile and a half wide. The bomb proved to be worthless as a weapon, however: it weighed 50 tons, and was too heavy for delivery by an aircraft. Later tests brought down the weight and ultimately adapted it to missile launching. The Soviet response was Andrei Sakharov’s independent design for a thermonuclear weapon, which he tested at Semipalatinsk in August 1953. This frantic contest ultimately led to the Soviet detonation in October 1961 of a 57-megaton weapon at Novaya Zemlya, an island between the Russian mainland and the Arctic Ocean. This bomb was originally designed to be 100 megatons but Khrushchev ordered the yield halved because of fears of unknown effects and the likely massive fallout. Had a 100-megaton bomb been tested, the US’s KC-135 aircraft monitoring Novaya Zemlya would not have made it back; as it was, it returned with a scorched fuselage.

The nuclear arms race led to massive research efforts to keep track of what the Soviet Union was doing. Many in the US air force wanted to document the ‘missile gap’ that the Democratic Party used in 1960 to help get John F. Kennedy elected as president, and this led to the development of the U2 extremely high altitude reconnaissance aircraft, as well as Corona satellites equipped with the so-called KH (keyhole) camera. The first U2 flight over the Soviet Union took off in July 1956, from Wiesbaden; the first successful Corona mission came four years later. Photos from the U2 had a much higher degree of resolution than those from the early KH cameras, but on 1 May 1960, the Soviets shot down a U2, piloted by Francis Gary Powers, over Sverdlovsk. Thereafter, the US invested more in satellites, cameras and photo interpretation, which made spying from space the preferred form of espionage.

The ultimate repertoire came to include advanced satellites equipped with very high resolution cameras, plus infrared sensors for night-time imaging and meters to record the electromagnetic pulse emitted by an explosion. The US also built numerous (usually secret) stations around the world to listen for the sound waves created in the atmosphere by a detonation, to record seismic movements caused by nuclear explosions, to detect light flashes reflected off the dark side of the moon, and to observe blast-induced ‘dimples’ in the ionosphere. Whenever possible, the US listens in on the electronic communications and conversations of nuclear scientists and officials in target countries. They endlessly collect samples from the air using filter-equipped aircraft; they deploy balloons designed to loiter at between 33,000 and 66,000 feet collecting radioactive debris. (The wreck of one of these balloons near Roswell, New Mexico, a small town close to Alamogordo, is almost certainly the source of the myth that a UFO crashed there, which has made Roswell the UFO capital of the world.)

The intelligence specialists did other ingenious things. One was to modify a U2 so that it could fly off an aircraft-carrier and collect overhead images of the French preparations for numerous tests in Polynesia. I would have liked to see a photograph of this, to know how the huge wingspan of the U2 managed to fit onto the carrier’s flight-deck, but Richelson doesn’t provide one.

One of the remote sensing devices invented by the nuclear spies is known as a ‘bhangmeter’. It’s an optical sensor carried on satellites to detect the bright double-pulse of light that an atmospheric nuclear explosion gives off. This burst of light is produced by a short, intense flash from the fireball followed by a longer, less bright emission as the expanding atmospheric shock wave, which is opaque, obscures the fireball and then reveals it again as the shock wave starts to dissipate. No natural phenomenon is known to imitate this signature. According to one authoritative source, the device’s name is derived from the Hindi word bhang, denoting a form of cannabis. Some scientist or scientists naming the device evidently thought that people who believed in the optical detection of atomic explosions must have been stoned. Amazingly enough, however, no bhangmeter has ever produced a false positive throughout many years of use.

Bhangmeters played a critical role in the mysterious explosion detected in the South Atlantic on 22 September 1979, one of the most puzzling episodes in the history of nuclear espionage. The scientific gadgetry successfully identified a nuclear detonation but neither the detection devices nor the best scientific and analytical minds in the business could ever figure out who was responsible. It was in that sense both a success and a failure, encapsulating all the problems associated with reliance on technological rather than human intelligence. Richelson covers this event thoroughly.

Among the most capable of all American military satellites were six known as the ‘Advanced Velas’ (vela is Spanish for ‘vigil’). These were launched into orbit in 1967, 1969 and 1970 and had a nominal design life of seven years, even though the last one continued to function until 1984. They each carried two bhangmeters. In 1979, Vela 6911 recorded a double-hump flash somewhere in an area encompassing the southern tip of Africa, parts of the Atlantic and Indian Oceans, and a bit of Antarctica. Vela 6911 was a good satellite: it had previously picked up double-flashes some 41 times, all of which had been positively identified as nuclear explosions. However, by 1979, another sensor on board, which was supposed to locate the explosion, was no longer working, and the two bhangmeters recorded the flashes but did not agree on their intensity. The attempt to verify the Vela data and then determine whose nuclear weapon it was set off one of the most extensive air sampling and analytical efforts of all time. No fallout was ever recovered, and the bhangmeters provided the only evidence that something nuclear had occurred.

Speculation centred on South Africa, Israel, a possible Soviet evasion of the Partial Test Ban Treaty then in force, India and many other possibilities. Because of the Iranian hostage crisis and the Soviet invasion of Afghanistan, it was a bad time politically for the Carter administration, and would have been particularly bad if Israel had turned out to be the culprit. A blue-ribbon panel concluded that the flashes were caused by something other than a nuclear explosion: Luis Alvarez’s hunch was that a micrometeorite had collided in space with the satellite, letting light into the bhangmeters. The Defense Intelligence Agency, the Naval Research Laboratory and the Los Alamos National Laboratory each launched full-scale investigations, but none was conclusive.

Over the years a number of Israeli journalists – Eli Teicher and Ami Dor-on, for example, in their 1980 book suppressed by the Israeli government, None Will Survive Us: The Story of the Israeli A-Bomb, and the newspaper Yediot Aharonot – have argued that the test was a joint Israeli-South African venture. In April 1976, the South African prime minister, John Vorster, visited Israel, and it’s thought that he negotiated an agreement with Yitzhak Rabin to trade fifty tons of South African uranium for thirty grammes of Israeli tritium, which is used to boost the yield of a nuclear bomb. However, if that is true, F.W. de Klerk did not mention it in his historic address of March 1993, renouncing his country’s nuclear weapons programme. The 1979 incident is an enduring warning of how easily we can become captives of our seemingly objective means of observation and not learn what actually happened.

Other detection failures have resulted from analytical blindness, a lack of quality control in agencies such as the CIA, and political expediency. In reporting on China’s nuclear programme, for example, the CIA was often sabotaged by its own cultural arrogance and a belief that the Chinese were simply not up to it. In June 1955 the legendary intelligence analyst, Sherman Kent wrote that China’s attempt to build a bomb would take well over ten and possibly twenty years. He based his assessment on China’s inability to process uranium ore, as well as its having ‘almost no scientific tradition in theoretical and experimental physics’: China tested a 20-kiloton device in 1964 – less than ten years later. The US knew it was coming only because the Soviet ambassador to Washington had warned a former US ambassador to the USSR.

More seriously, the CIA was adamant in its belief that China’s first bomb would be a plutonium device because ‘the plutonium route to the bomb was the easiest’ and the Soviets, British and French had all taken that road. Nonetheless, evidence was starting to pile up that the Chinese were developing a uranium bomb. Donald Chamberlain, the head of the CIA’s Office of Scientific Intelligence, had misjudged the Chinese programme and, in the words of one CIA man, ‘got stubborn about it’. Images from a September 1959 U2 mission over Lanzhou revealed a 2000-foot-long building with the characteristics of a Soviet gaseous diffusion plant. Of course, overhead photography can’t see inside buildings and the evidence could not be conclusive. After the first test, the chairman of the Atomic Energy Commission, Glenn Seaborg, informed Lyndon Johnson that the Chinese bomb was a ‘uranium-235 device – an implosion design using uranium rather than plutonium’. The unexpected nature of the design, as well as the later discovery that the U-235 had come not from Russia or from the US via its European allies but from China itself, led to a major re-evaluation of China’s capabilities.

One of the key figures in the Chinese programme was Wang Ganchang, whom Richelson (like the CIA) insufficiently credits. He had received his PhD in 1933 under Lise Meitner in Berlin, and in 1948 studied at Berkeley’s Radiation Laboratory, sailing for home just as the Communists came to power in 1949. In 1960, Wang was working at the Joint Institute for Nuclear Research at Dubna with Bruno Pontecorvo, the Jewish-Italian atomic physicist who had defected to the USSR in 1950. Because of the Sino-Soviet dispute, he returned to China and devoted himself to developing China’s bomb.

In 1964, while doing research for an article on China’s test, I telephoned Emilio Segrè, a veteran of Los Alamos and my colleague at Berkeley, where he had been a professor of physics since 1946. I asked him whether he remembered Wang and he said he did. When I asked whether he thought Wang could have built a bomb, Segrè replied: ‘Of course. I taught him. After that it’s merely a matter of engineering.’

Other cases in which the CIA allowed ideological blindness to affect its analysis included its insistence that post-Soviet Russia had tested a nuclear device at Novaya Zemlya on 16 August 1997: the blast occurred sixty miles out to sea and the seismic tremor recorded by the CIA was actually caused by an earthquake. Similarly, the US allowed one inexperienced CIA analyst to insist that the aluminium tubes Saddam Hussein was importing into Iraq were for centrifuges to enrich uranium, in the face of truly expert information from the Department of Energy: false intelligence which made its way into Colin Powell’s speech to the Security Council on 5 February 2003, in which he set the stage for the invasion of Iraq by presenting what was thought to be ‘definitive’ American secret intelligence proving the existence of weapons of mass destruction in Iraq.

Political expediency clearly influenced the United States’s failure to monitor Saddam Hussein’s efforts to build an atomic bomb during the 1980s: the US was then supporting him in the war he started with Iran. Similarly, in late October 1986, in certifying that Pakistan ‘does not possess a nuclear explosive device’, the Reagan administration suppressed information from the Defense Intelligence Agency that Pakistan had detonated a high-explosive trigger for an implosion device and had succeeded in producing uranium composed of 93.5 percent U-235. It did so because Pakistan was the US’s indispensable ally in the guerrilla war it was sponsoring against the Soviet Union in Afghanistan.

Richelson does not leave us with a take-home message from his long and detailed account. He ends with nothing more than a suggestion for more of the same: ‘continued aggressive and inventive intelligence collection and analysis’ on Iran and North Korea’s nuclear programmes, ‘and those of other rogues, will be necessary to permit a clear understanding of the threat, and to guide decisionmakers in choosing what courses of action to take or avoid’.

The true significance of Richelson’s compilation lies, however, in its demonstration that the US economy is now dominated by ‘military Keynesianism’, whereby the growth rate of the economy is boosted by government spending on troops and weapons. But military Keynesianism short-circuits Keynes’s insistence that government spending should be cut back in times of nearly full employment. Governments have always found it politically hard to reduce military spending once committed to it, particularly when the munitions makers distribute their benefits as widely as possible and enlist the support of as many politicians as they can – as they have in the United States. In short, military Keynesianism leads to constant wars, or a huge waste of resources on socially worthless products, or both.

Between 1940 and 1996, the US spent at least $5.8 trillion on the development, testing and construction of nuclear weapons. By 1967, the peak year of the nuclear stockpile, it possessed some 32,500 deliverable bombs, none of which, thankfully, was ever used. But they perfectly illustrate Keynes’s proposal that in order to create jobs, the government should bury money in disused mines and then pay unemployed workers to dig it up. As Richelson has shown, military Keynesianism has involved not just the building of more atomic weapons than any nation could ever use, but also a fantastically expensive programme of satellites, cameras, specialised aircraft, monitoring stations around the world, and other secret activities aimed at detection.

On 17 May, a House of Representatives committee approved the Bush administration’s programme to build a new generation of nuclear warheads. It also increased the funding for what is known as the Reliable Replacement Warhead programme, from roughly $25 million to $52.7 million. Accompanying these sums are billions of dollars for new and untested technologies under the label of ‘ballistic missile defense’. Richelson has unintentionally documented not how these huge expenditures have made us safe, but the degree to which our economy has become dependent on them.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences