a very small group of concerned scientists… Altogether, there were less than one hundred people, most of whom thought of themselves — probably correctly — as being involved in making one of the most fateful decisions of all time.
Another crucial figure in the debate was Dean Acheson, now Secretary of State. Truman listened to Acheson, Secretary of Defense Louis Johnson and the Joint Chiefs of Staff more than he listened to the scientists. The Joint Chiefs told Truman, without much staff evaluation, that a Soviet H-bomb would be “intolerable.” So, in as many words, did Acheson. None of the groups that participated in the debate studied the possibility that the result of the American initiative might be an arms race. The H-bomb appeared to its supporters to offer a return to nuclear superiority. On January 31, 1950, Truman announced in favor of proceeding to develop it.
Teller took the President's decision as a personal victory. Since at least the September afternoon of Oppenheimer's blunt advice to keep his shirt on he seems to have felt that his fellow physicist was personally attempting to impede him. Oppenheimer was chairman of the GAC, which had been asked in October to advise the AEC commissioners in the matter. Teller went to Washington to lobby past that committee. Among others he talked to Kenneth D. Nichols, now a general and with Groves' retirement the Army's resident nuclear weapons expert. One autumn Sunday morning on Nichols' front porch Teller argued so emotionally that Nichols finally challenged him: “Edward, why are you worrying about the situation so much?” “I'm not really worrying about the situation,” Nichols remembers Teller responding; “I'm worrying about the people who should be worrying about it.”
The GAC met on October 29 and 30 (Oppenheimer, Conant, Fermi, Rabi, Caltech president Lee DuBridge, metallurgist Cyril Smith, Bell Laboratories president Oliver E. Buckley, engineer Hartley Rowe, Glenn Sea-borg absent). It recommended in response to the Soviet achievement that the AEC look further into increasing the production of fissionable materials. It called for “an intensification of efforts to make atomic weapons available for tactical purposes.” It proposed building facilities to produce more neutrons for weapons research and development. And it strongly recommended pursuing an existing Los Alamos program to use small quantities of tritium in atomic bombs to “boost” those bombs to more efficient and more powerful explosions (an invention that was successfully tested in May 1951).
But the committee also recommended against pursuing “with high priority the development of the super bomb.” It based its recommendation on essentially two arguments: that at ten megatons a Super would be a weapon of mass destruction only, with no other apparent military use; and that it would not obviously improve the security of the United States. It would not do so, in particular, because the design then under consideration — the design reviewed at the Super conference in 1946, Teller's design, the classical Super — looked as if it would require large amounts of tritium, and pluto-nium and tritium, both created in nuclear reactors, would compete for existing plant capacity. Tritium was as well eighty times as expensive to make, gram for gram, as plutonium. The U.S. nuclear stockpile in late 1949 consisted of about two hundred atomic bombs. Slowing production of bombs that worked for a new weapon that might not work made no sense to the scientists and engineers of the GAC. As Oppenheimer summarized in later testimony, “The [H-bomb] program we had in 1949 was a tortured thing that you could well argue did not make a great deal of technical sense. It was therefore possible to argue also that you did not want it even if you could have it.”
The GAC members divided themselves into a majority and a minority to write explanatory annexes to their October 30 report. Conant drafted the majority annex, which Oppenheimer, DuBridge, Rowe, Smith and Buckley signed. It said that a Super “might become a weapon of genocide.” It said such a bomb “should never be produced… To the argument that the Russians may succeed in developing this weapon, we would reply that our undertaking it will not prove a deterrent to them. Should they use the weapon against us, reprisals by our large stock of atomic bombs would be comparably effective to the use of a super.” Rabi drafted the minority annex, which he and Fermi signed. It argued that the H-bomb question might serve as a springboard for new arms-control efforts. It described the Super as “a weapon which in practical effect is almost one of genocide,” then for good measure went even further in condemnation: “It is an evil thing considered in any light.”
Teller had been the leading proponent of this enormously powerful weapon since before Pearl Harbor. His Manhattan Project colleagues on the GAC — Oppenheimer, Conant, Fermi, Rabi, Smith — had encouraged him and even worked alongside him. Nothing was said then of evil, of genocide. They who had won high position building weapons used in the mass destruction of two Japanese cities now condemned another, more ingenious weapon. They who were scientists and understood that kindling a sustained thermonuclear reaction on the earth was a historic experiment in fundamental physics proposed the indefinite postponement of that experiment, thus probably handing the triumph gratuitously to the Russians. A courier delivered a copy of the committee's report to Los Alamos in advance of a delegation of interested congressmen who had already seen it. With AEC chairman David LilienthaPs approval, John Manley, associate director of the laboratory and executive secretary of the GAC, showed it to division heads, including Teller. “Edward was of course just completely aghast,” Manley recalls, “and his reaction was to offer me a bet that if we did not proceed immediately with a crash program on the Super, he would be a prisoner of war of the Russians in the United States.”
Teller explained his reaction to a
The reasons they gave just made me mad… The important thing in any science is to do the things that can be done. Scientists naturally have a right and a duty to have opinions. But their science gives them no special insight into public affairs. There is a time for scientists and movie stars and people who have flown the Atlantic to restrain their opinions lest they be taken more seriously than they should be.
It is sometimes difficult even for other scientists to remember that the atomic and hydrogen bombs were developed not only as weapons of terrible destruction. They were also, as Fermi once said, “superb physics.”
At the time of the Truman announcement, J. Carson Mark headed the Theoretical Division at Los Alamos. “Truman's words,” Mark comments, “didn't necessarily mean we did anything much different from what we had been doing because we really didn't know how to make a gadget that would work as a hydrogen bomb.” In February 1950, when Washington learned that for seven crucial years, from 1942 to 1949, Klaus Fuchs had passed along secret information to the Soviet Union, Truman turned to a special committee of the National Security Council for advice. The committee recommended that the President clarify his somewhat vague January 31 directive to put energetic development of a hydrogen bomb clearly on record. It emphasized at the same time that there was no obvious way to speed up the weapons test schedule and no guarantee of success. Truman promulgated the committee's report as official policy.
The necessary next step toward a workable thermonuclear weapon was elaborate mathematical simulation. Without a mathematical model of the evolution of thermonuclear burning from a fission trigger event, Mark explains, “no conclusive evidence was possible short of a successful stab in the dark, since a [test] failure would not necessarily establish unfeasibility, but possibly only that the system chosen [Teller's classical Super, for example] was unsuitable.” The Super problem, as the simulation came to be called, was the largest mathematical effort ever undertaken up to that time, “vastly larger,” writes Stanislaw Ulam, “than any astronomical calculation done to that date on hand computers.” It involved calculating the blooming of a thermonuclear explosion — its heating, its extremely complex hydrodynamics, its evolving physical reactions — in progressive increments of less than one ten-millionths of a second. The 1946 Super conference had called for such calculations, but until the further development of the electronic computer they simply could not be accompUshed in any reasonable period of time. Teller said as much in a September 1947 report, discussing choosing between his classical Super and an alternative design: “I think that the decision whether considerable effort is to be put on the development of the TX-14 or the Super should be postponed for approximately two years; namely, until such time as these experiments, tests, and calculations have been carried out.”
In late 1949, before the H-bomb decision, Los Alamos began detailed work preparing a machine calculation for the first primitive electronic computer, the ENIAC at the Aberdeen Proving Grounds in Maryland. After the H- bomb decision Ulam and the Theoretical Division's Cornelius J. Everett decided to go ahead by hand with a simplified version of the calculation. “We started to work each day for four to six hours with slide rule, pencil, and paper,” Ulam remembers, “making frequent quantitative guesses… Much of our work was done by guessing values of geometrical factors, imagining intersections of solids, estimating volumes, and estimating chances of points escaping. We did this repeatedly for hours, liberally sprinkling the guesses with constant slide-rule calculations. It
