The Twilight of the Bombs

by Richard Rhodes

  The country has claimed that its intentions in developing nuclear explosives were benign, that a mining country might have use for such technology, and that it took its inspiration from the U.S. Plowshare Program. Plowshare, championed by the Lawrence Livermore National Laboratory director Edward Teller, sought to develop peaceful nuclear explosives (PNEs) for mining, harbor dredging, and underground gas and oil liberation. The U.S. Atomic Energy Commission conducted thirty-five PNE tests between 1957, when the program began, and its termination in 1973. Teller and his lab were “proselytizing the PNE8 worldwide” during that period, a nuclear official told the journalist Mark Hibbs. The only important difference between a peaceful and a military nuclear explosive, however, is its target. Significantly, South Africa began secretly supplying Israel with uranium yellowcake in 1961,9 with a first shipment of ten tons, and by 1972 it was negotiating the first of many secret agreements with its Middle Eastern ally.

  Waldo Stumpf, a metallurgical engineer who was the general manager of the South African Atomic Energy Corporation (AEC) from 1987 to 1989, characterizes his country in the 1970s as beleaguered, increasingly isolated both politically and technologically, and essentially backing into weapons development:

  During the 1970s10 and especially during the latter half of that decade, the international security situation around South Africa deteriorated markedly. This was mainly due to its own racially-based internal policies but was also exacerbated by Portugal’s withdrawal from its African colonies of Mozambique and Angola and the uncertainties about the true intentions of the Warsaw Pact countries and especially the Soviet Union, in the light of their openly declared expansionist policies in Southern Africa. The strong buildup of Cuban surrogate forces in Angola from 1975 onwards and which eventually peaked at 50,000 foreign soldiers reinforced a strong perception within the Government of international isolation should South African territory be under threat.

  Increasing international restrictions on the supply of conventional arms against South Africa, primarily due to its internal policies, also made the argument that the country virtually had no alternative but to develop its own nuclear deterrent to counter an external threat probably convincing to the Government of the time.

  Stumpf cites11 as particularly egregious affronts the Carter administration’s 1976 decision to block further shipments of fuel for Safari-1 (while pocketing the payment that the South Africans had made), and the U.S. Congress’s passage of a nonproliferation act in 1978 that forbade the transfer of nuclear technology to countries not party to the NPT. “This pressure by the USA,”12 Stumpf writes, “was viewed very negatively by South Africa.… These actions strained US-South African relations in the nuclear field severely at that time.” South Africa was denied a seat on the IAEA board of directors in 1977, Stumpf complains further,13 and barred from participating in an IAEA conference held in India in 1979 despite the fact that India itself had tested a “peaceful” nuclear explosive in 1974.

  The problem with all these complaints is that South Africa had been pursuing nuclear-explosives development since at least the early 1960s, and probably even earlier, in the mid-1950s, like many other countries (e.g., Sweden, Norway, Yugoslavia, and Switzerland) in the years before the NPT provided a treaty-based alternative to nuclear arms. “South African scientists14 demonstrated the feasibility of uranium enrichment at the same time that the country began to experience increased threats from enemies in the region and throughout the world,” two historians of the South African nuclear program pointed out. “The concept of ‘encirclement,’ or perceptions of heightened threats from all quarters—at home, in the region and abroad (the laager complex)*—figured prominently in the minds of senior South African political and military officials.” The historians continued:

  Senior South African politicians,15 members of the ruling Afrikaner elite, and scientists, engineers, and military officers involved in defense research and development shared a sense that they had been abandoned by the United States after it intervened in Angola in the mid-1970s. The sense of betrayal and abandonment was fueled further by escalating violent opposition at home, increased pressures from international anti-apartheid opponents, and a recognition by leaders of governments, the political Afrikaner elite, and the wider defense-establishment elite that time was not on their side. As both a strategic rationale and a rationalization, former apartheid leaders viewed weapons of mass destruction as a form of insurance and political leverage to guarantee Western involvement in the event that the apartheid regime found itself “up against the wall facing an overwhelming communist threat” at home and in the region.

  By 1976, South Africa was preparing a site at Vastrap Weapons Range, a military testing ground in the Kalahari Desert about four hundred miles west of Johannesburg, for an underground nuclear test. Its enrichment plant had not yet enriched enough uranium, but the bomb-design team had constructed a first massive uranium gun bomb. The design was similar to the Little Boy gun bomb that was exploded over Hiroshima. Little Boy fired a hollow, cylindrical uranium “bullet” up a cannon barrel where it seated itself around a uranium target post to form a cylindrical supercritical mass. The target in both devices was surrounded with a tungsten tamper to reflect neutrons back into the core and to hold the explosion together for a few milliseconds longer, allowing the chain reaction to proceed through a few more exponential generations and thus produce a higher yield. Little Boy had yielded 13.5 kilotons; the yield of the South African design was estimated to fall between 10 and 18 kilotons. The South African target assembly is reported to have been spherical rather than cylindrical, however, with a blind hole into which a solid uranium bullet seated itself.

  The gun assembly system, while technically the most straightforward method of detonating a nuclear weapon, is still difficult to design. “In comparing a gun device16 with an implosion device,” a U.S. expert told Hibbs, “‘If you had to rely on the open literature, you would get the impression that the gun bomb was easier.’ But engineering problems in the gun device would ‘later arise which would not be apparent on the basis of the design itself.’” Hibbs explored the engineering problems with a spokesman for Armscor, the South African government’s weapons manufacturer:

  LITTLE BOY CROSS SECTION (COURTESY OF JOHN COSTER-MULLEN)

  In fact, when engineering work got started, South African experts soon ran into trouble. “Many areas of development and production were problematic during the early years of the program,” Armscor said.

  South Africa had problems in assuring that the velocity for the HEU projectile in the gun barrel—which experts said would be about 700 meters [2,300 feet]/second—could be repeatedly attained with accuracy and reliability. If the speed is not carefully controlled, one U.S. expert said, “injection of neutrons into the highly supercritical system can happen too late. If so, the ‘bullet’ will feed on the target and burst into pieces.” Timing problems, another weapons expert said, could be solved only by “experimental investigation. You set it up, fire it over and over, and record what happens with cameras and timing equipment. Then you look at the result and find out where the problems are.”

  “Projectile velocity was a critical factor” in the early days of the program, Armscor said. “Special propellants and igniters were developed and matched to give the desired accuracy.” Doing this was very challenging, since “the team had to develop everything in-house due to the security requirements of the program.”

  Armscor also had problems with the “symmetry requirements when the projectile is injected.” Experts clarified that this pointed to difficulties in getting the uranium “bullet” to travel predictably and seat properly into the uranium target during firing tests. Some problems were caused by variations in temperature. South Africa had to “qualify” the design “to make sure that the ‘bullet’ didn’t wobble in the barrel and that the device would work in conditions of heat and cold.”

  South Africa also experienced problems in demonstrating the reliability of ar
ming devices. These problems, according to one U.S. expert, are “common to all nuclear weapons development efforts.”

  In 1975, the South Africans began drilling three deep boreholes17 in the ground at Vastrap, abandoning one because of bad geology, completing the other two in 1977. They planned first to cold-test a uranium gun with a depleted uranium core (which would therefore not chain-react) to assess the performance of its nonnuclear components, instrumentation, and telemetry. Then, when their enrichment plant had produced enough HEU, in late 1977 or 1978, they would test a second gun at full yield. Since India had tested Smiling Buddha, its so-called peaceful nuclear explosive, underground in 1974 without irreparable damage to its international relations, the South Africans did not expect any large outcry.

  On 3 and 4 July 1977, however, a Soviet Cosmos 922 surveillance satellite photographed the test-preparation activities at Vastrap. The film returned from that mission alarmed Soviet intelligence sufficiently that it launched a Cosmos 932 low-orbit close-look satellite on 20 July. That film convinced the Soviets that a South African test was imminent, and on Saturday, 6 August 1977, coincidently the thirty-second anniversary of the atomic bombing of Hiroshima, the Soviet embassy in Washington hand-delivered a message to Jimmy Carter’s White House from Soviet leader Leonid Brezhnev asking for Carter’s help in preventing a South African test. The appeal was an exemplary but far from unique example of Cold War cooperation between the two supposedly implacable enemies.

  Within hours, the White House ordered U.S. aircraft and satellite surveillance of the Kalahari site. The next day, wrote the analyst Jeffrey Richelson, “an unmarked light aircraft18 flew over the borehole sites, the instrumentation and cable trenches, the area that would house the instrumentation trailers, the office and the accommodation block.” The South Africans failed to identify the plane, Richelson added, which “belonged to the U.S. military attaché’s office in Pretoria and was equipped with suitable cameras.” In the days ahead at least one, and possibly two, U.S. reconnaissance satellites—one capable of returning imaging data in near-real time—were repositioned to keep watch over Vastrap.19

  Tass, the Soviet news agency, broke the story on Monday, and the next day South Africa began scrubbing the site even as it issued contemptuous denials of the Soviet Union’s “romantic notion about20 a Kalahari Test Site.” Brezhnev wrote to the British prime minister, James Callaghan, on Monday as well, pointing out that Britain had “the necessary channels21 and possibilities to exercise direct restraint in respect of [South Africa].” Callaghan, surprised and out of the loop because British intelligence had missed the call, spoke to Carter on 14 August, suggesting that the Germans might be involved. Carter told Callaghan drolly, the conversation notes report, “that although he knew22 that the Germans were involved, he had considered the US and the French were the two main culpable parties.” Carter responded to Brezhnev that day as well, agreeing with the Soviet assessment. He sent an ambassador to France with satellite photographs to win the French government’s support.

  “The United States and other23 governments exercised concerted pressure,” Richelson concluded, “which reportedly included France’s threat to break diplomatic relations and terminate its assistance in constructing the nuclear power plants that it had sold South Africa—a sale for which the French had taken some flak. The United States followed by sending to Pretoria a precise statement of the assurances it wanted. ‘We were pretty severe in private,’ one U.S. official noted.” South Africa capitulated on Sunday, 21 August, and Carter announced its prevaricated pledge at a press conference the following Tuesday: “South Africa has informed us that they do not have and do not intend to develop nuclear explosive devices for any purpose, either peaceful or as a weapon.” He knew better, having seen an interagency assessment on 18 August that identified a “long-standing [South African] program24 to develop a nuclear weapon” but could postulate no circumstances under which the South Africans could be forced to desist. Though he had just asserted that South Africa had no program, Carter went on to say that the United States would “continue to monitor25 the situation very closely.”

  A problematic consequence of doing so appeared two years later, in September 1979. In the interim South Africa had filled the Vastrap boreholes with sand, capped them with concrete, and dismantled and abandoned the site. It had continued producing HEU and had designed and built a one-ton gun bomb to replace the three-ton behemoth of the Vastrap era. According to analysts Helen Purkitt and Stephen Burgess, the U.N. Security Council’s passage of a resolution in 1977 calling for a mandatory South African arms embargo had perversely emboldened corporate South Africa, which thereafter celebrated the day “as the onset of26 the growth of the South African military-industrial complex.” (The share of its weapons budget that South Africa spent domestically increased from 29 percent in 1977 to 85 percent by 1982.)27

  Most significantly, the prime-ministership had passed from B. J. Vorster to P. W. Botha, a right-wing white supremacist who for a decade previously had been minister of defense and who continued to hold the defense portfolio as prime minister. Upon taking office in 1978, Purkitt and Burgess reported, Botha had articulated a brutal program of “total strategy” to defend the Afrikaner regime, including “assassinations, torture, and smuggling28 as well as forgery, propaganda, and subversion. All were defined as legitimate weapons against the ‘total onslaught’ of communist and black nationalist forces.” Botha also gave “unconditional support to the development of advanced weapons projects, including weapons of mass destruction.… Covert nuclear warheads and an arsenal of different types of missile systems came to be viewed as essential force multipliers for a military stretched thinner and thinner by operational demands at home and along the border, which included South West Africa (Namibia) and Angola.”

  As part of his “total strategy,” Botha assigned Armscor, the South African Defense Force, and the Atomic Energy Board to develop a nuclear-weapons-research program; the committee thus formed was authorized to look beyond heavy gun bombs to “studies of implosion29 and thermonuclear technology” that might produce weapons more adaptable to delivery by missile. In the summer of 1978, Armscor took over production of nuclear weapons, inheriting from the Atomic Energy Board the one-ton gun bomb, charged with about 60 kilograms of 80-percent-enriched uranium.

  On 22 September 1979, an aging and ailing U.S. Vela satellite orbiting sixty thousand miles above Earth picked up an intense double flash of light in the South Indian Ocean. No natural phenomenon is known to produce such a double flash; its only known source is a nuclear explosion.* The time between the peaks of the double flash gives a direct measure of the yield, and in this case the yield measured between two and four kilotons, too low for one of South Africa’s uranium-packed gun assemblies, which in any case hardly needed testing.

  GRAPH OF NUCLEAR EXPLOSION DOUBLE FLASH

  President Carter, National Security Adviser Zbigniew Brzezinski, Secretary of State Cyrus Vance, and other officials gathered at the White House within twenty-four hours of the flash report to hear from CIA and Defense Intelligence Agency representatives that the Vela signals almost certainly indicated a nuclear test. Carter wanted two questions answered: Was it really a nuclear test? And, if so, whose was it?

  A panoply of investigations followed. Multiple missions of military aircraft—twenty-five sorties totaling 230 flight hours between 22 September and 29 October—searched the remote ocean regions southeast of the southern tip of Africa and collected samples of air and dust. Data was reviewed from U.S. military satellites, as were recordings of underwater acoustic signals picked up by U.S. navy hydrophones. An all-sky camera at Japan’s Showa Base, in Antarctica, revealed a suddenly brightened patch of aurora tens of kilometers in diameter northward of the base at exactly the time of the double flash.30 The Arecibo radio telescope in Puerto Rico detected an upper-atmosphere electromagnetic disturbance moving southeast to northwest early on the morning of 22 September. Corroborative data would come lat
er from New Zealand—radioactive fallout in rainwater samples collected between the beginning of August and the end of October—and from western Australia, where low but measurable levels of radioactive iodine-131 in sheep thyroids were registered.

  All such measures that were directional pointed toward a pair of isolated islands, Prince Edward and Marion, two seamounts located twelve hundred miles southeast of the southern tip of Africa and fifteen hundred miles north of Antarctica in a region of wind and bad weather where ships rarely sail or aircraft fly and where clouds shield the islands from visual satellite coverage an average of 288 days per year. Significantly, they were South African possessions.

  While its investigations were ongoing, the Carter administration kept the Vela information secret. By 23 October, at a meeting of the National Security Council’s Special Coordination Committee (SCC), the State Department was prepared to report that “the Intelligence Community has high31 confidence, after intense technical scrutiny of satellite data, that a low yield atmospheric nuclear explosion occurred in the early morning hours of September 22 in an area comprising the southern portions of the Indian and Atlantic Oceans, the southern portion of Africa, and a portion of the Antarctic land mass.” If this information became public, the report continued, “most observers will assume that South Africa tested a nuclear device.” The debate then focused on what to do about it.

  Far from advancing the United States’ antiproliferation efforts, the State Department argued, this discovery of a clandestine test actually threatened those efforts. If South Africa were called out and further isolated, it “might then support nuclear weapons32 programs in other politically isolated states, such as Israel and Taiwan.” Public exposure would lead to a more stringent United Nations arms embargo that would probably go beyond what “Western members of the Security Council could accept.” It would “come at a bad time for efforts to achieve settlements in Rhodesia and Namibia.” It could interfere with “efforts to deter proliferation elsewhere, e.g., Pakistan and India.” Perhaps more to the point, the State Department acknowledged, “the South Africans have the capability33 to retaliate against sanctions with some effect.… The UK, for example receives something more than 50 percent of its uranium from South Africa.… The West Germans look to South Africa for nearly half their uranium, the Japanese would view with alarm any major dislocation in the world uranium supply market, and a number of other countries would be affected to varying degrees.” Not mentioned in the NCC discussion was the fact that the following year, 1980, was an election year in the United States, and that the Carter administration already had its hands full dealing with a revolution in Iran; the day before the NCC meeting, Carter had reluctantly admitted the deposed Shah, Mohammad Reza Pahlavi, into the United States for treatment of pancreatic cancer, a decision that led directly to the long hostage crisis that began on 4 November and would plague Carter’s final year as president.