At a meeting of the Air Force Scientific Advisory Board on October 21, von Neumann talked of the possibility that the accuracy requirement for Atlas might be relaxed, but he was still tentative on this matter. “There is no doubt,” he said, “that this subject should be examined much more carefully.” Bruno Augenstein’s briefing before the committee in late December describing the precise guidance system that could be used, portraying the precise trade-off curves between CEP and the number of weapons necessary, provided the more careful analysis and reinforced the direction in which von Neumann was headed from the start.
Augenstein’s formal RAND research memorandum, RM-1191, titled A Revised Development Program for Ballistic Missiles of Intercontinental Range, was delivered to the Air Force on February 8, 1954. Collbohm wanted above all to beat the Teapot Committee to the punch. He did. Two days later, von Neumann filed his report with the Air Force Chief of Staff. It concluded that the ICBM was feasible, that twenty launching sites with a stockpile of 100 missiles could be established by June 1960 if the requirements for the warhead’s weight and accuracy were substantially modified, and that modifying them would in no way reduce the missile’s effectiveness.
On March 11 and 14, the Air Force Council met and “agreed that it is possible to broaden CEP requirements immediately” and noted that doing so “will simplify the problems of developing adequate guidance systems for all strategic missiles.” The Council ordered that the Atlas “should be given a high priority” and “extraordinary action be taken to accelerate the revised program… so as to achieve the early establishment of an optimum intercontinental ballistic missile system.” On May 6, General Hoyt Vandenberg, Air Force Chief of Staff, approved the Council’s recommendations.
In July, Trevor Gardner directed General Schriever to take charge of the newly created Western Development Division of the Air Research and Development Command. The organization was so highly secretive that its very initials, WDD, were classified beyond top secret, and its headquarters were located in an abandoned parochial schoolhouse on the edges of Los Angeles, with frosted and heavily barred windows, locked doors, a chain-link-fenced parking lot, and an armed guard sitting inside. Simon Ramo and Dean Wooldridge banded together to form the Ramo-Wooldridge Corporation (later TRW, Inc.), whose sole task was to provide technical assistance to the ICBM program.
The age of the strategic missile, “the hydra-headed monster,” as those working on it liked to call it, had begun.
Albert Wohlstetter was still briefing his overseas-base study in the fall of 1953 when he heard about Bruno Augenstein’s analysis on the feasibility of the ICBM. Wohlstetter’s first reaction was minor panic. He had spent nearly two years working on the vulnerability of SAC’s bombers. First, the hydrogen bomb came along and threw off his calculations about how vulnerable his proposed basing system might be, and now the intercontinental ballistic missile threatened to make his whole study irrelevant.
His worries were alleviated, however, by two considerations. First, bombers would still comprise the vast bulk of both sides’ strategic arsenals for years to come. Second, the ICBM would make SAC vulnerable in ways completely different from those that Wohlstetter was describing in the overseas-base study. Furthermore, the protective schemes proposed in that study would be inadequate against ICBMs, which reach their targets so quickly that in a surprise attack there would be virtually no warning time at all.
In short, the ICBM would allow Albert Wohlstetter the perfect opportunity to be commissioned a follow-on study to the overseas-base study, an update given the new technological situation. Wohlstetter was making the vulnerability of SAC his own little cottage industry.
Wohlstetter had also acquired a singular reputation with RAND management. The base study was a great success. He had delivered it ninety-two times—ninety-two, beyond most people’s comprehension, much less endurance. And it worked. General Tommy White pushed it; the Air Force Council endorsed it; RAND saved the world—that was how management felt. Wohlstetter wanted this new project, on SAC vulnerability in the new age of ICBMs, to be grand and comprehensive. More than twenty RAND technicians and consultants ended up working on specialized studies within the project—on early-warning lines, on infrared detection of missiles, on fallout estimates, on the practicality of bomb-alarm systems, on the hardening of aircraft shelters, on possible anti-missile defenses. It was a high-priority RAND project, and Wohlstetter could pick his team as he wished, even raiding other projects while doing so.
Just after the start of the project, in January 1954, Wohlstetter and Fred Hoffman, his assistant, who had also worked on the overseas-base project, laid out their basic assumptions and a few tentative conclusions in an internally circulated RAND document called Defending a Strategic Force After 1960. Their first observation: “The defenses programmed, or recommended, to protect SAC in the Fifties will be entirely ineffective against an [ICBM] which would deliver bombs with essentially no warning,” and the ICBM “may be feasible for the Russians by the end of this period.” If the United States cannot protect SAC from a Soviet ICBM threat, they continued, reflecting a principle that went back to Bernard Brodie’s 1946 essay, “then our advertised capability for retaliation will be fictitious.” SAC could not “hurt the Russians very much.”
A few rudimentary calculations revealed to Wohlstetter and Hoffman that the Soviets could destroy at least 80 percent of SAC bombers with a force of about 150 missiles, less than one-third the number of missiles that Air Force Intelligence was predicting the Russians would have by 1960. It was the same problem that they faced in the overseas-base study—a surprise attack that could decimate SAC and thus wipe out the deterrent force.
But the ICBM had limitations, and if SAC were to be protected, there must be a way to exploit them. These limitations were poor accuracy and a payload too small to deliver—at the time, anyway—more than a couple of megatons of explosive power. The solution, then, was to put the bombers in blast shelters. This could be done, they reasoned, by placing them on top of hydraulic elevators that could sink into a pit shaped like the contours of the airplane itself and then sliding massive concrete doors over the pit. Unsheltered, a bomber will be destroyed by a 500-kiloton bomb exploded as far away as three miles. A bomber placed in a shelter hardened to resist fifty-five pounds per square inch (psi) of blast overpressure, however, will not be destroyed by 500 kilotons unless the bomb lands 3,600 feet from the plane. A much tougher shelter still, one hardened to resist 200 psi, is destroyed only if the bomb lands within 600 feet. The destructive power of a bomb depends on its explosive yield and its accuracy. The ICBM would already be quite inaccurate; and hardening the target to resist the force of blast would have the same effect as reducing its explosive yield considerably. As a result, Wohlstetter and Hoffman noted in their initial calculations, a Russian attack with 300 ICBMs directed against SAC bases would probably destroy only 30 percent of the bombers if they were hardened to resist 55 psi, and only 2 percent if they were hardened to 200 psi.