The problem had first been noticed by a young engineer in the Pentagon’s Office of Systems Analysis named Ivan Selin. Selin was Alain Enthoven’s deputy for strategic systems in the last two years of the Lyndon Johnson Administration, and a Whiz Kid of the highest order—fast, witty, fluent in seven languages, a brilliant computer engineer. One of his first tasks was to go down to SAC Headquarters and check out the SIOP. He discovered not only that the most limited option required firing a couple of thousand warheads, but that many of the military targets in the U.S.S.R.—especially the command-control facilities—were not too far from the downtown areas of the U.S.S.R.’s most populous cities. Even in the more limited options, there would be too much radioactive fallout, too many civilians killed. He worked with the JSTPS in Omaha to change several of the “aim points” and to remove certain targets from the limited options altogether. Still, that was not enough. If the idea behind counterforce was to signal restraint and to prompt reciprocal restraint from the Soviets, the Pentagon would have to devise new options that were much more limited in magnitude.
Selin received little encouragement from his superiors. Alain Enthoven, his immediate boss, fiercely loyal to McNamara, had embraced the tenets of assured destruction as an almost religious principle. Enthoven clearly believed that limited nuclear options were necessary if deterrence failed, but he did not want to change assured destruction as the declaratory policy, and certainly did not want to go mucking about in the SIOP trying to program more refined options. Like McNamara, Enthoven believed that once one trod down that road, the game was lost—the Air Force would find excuses to build more and more weapons, forcing the Soviets to respond, spurring an all-out arms race. The Whiz Kids of the early 1960s had learned that lesson after Ann Arbor, and they were not about to repeat the experience.
Selin received a boost in 1969. Administrations changed; Richard Nixon took over the White House; for one year, Selin became Acting Assistant Secretary of Defense for Systems Analysis. Nixon’s national security adviser was Henry Kissinger. Kissinger had long been a critic of massive retaliation. One of his first tasks, upon entering office, was to have Nixon order a study “reviewing our military posture and the balance of power.” The study was completed on May 8, 1969, and was labeled National Security Study Memorandum (NSSM) No. 3.
It was an interagency project, but the bulk of the analytic work was done by Ivan Selin and his staff at Systems Analysis. Among the deeds of NSSM-3 was to kill assured destruction. “Up to now,” it read, “the main criterion for evaluating U.S. strategic forces has been their ability to deter the Soviet Union from all-out, aggressive attacks on the United States.” However, a nuclear war may not take form as a series of “spasm reactions.” It “may develop as a series of steps in an escalating crisis in which both sides want to avoid attacking cities, neither side can afford unilaterally to stop the exchange, and the situation is dominated by uncertainty.” If during a future Berlin crisis, for example, the Soviets thought that the U.S. was about to start a nuclear war, the U.S.S.R., rather than being struck first, “might consider using a portion of its strategic forces to strike U.S. forces in order to improve its relative military position….” NSSM-3 conceded that these sorts of attacks “have no precedent in Soviet military doctrine or tradition,” making it “highly unlikely that such a situation would develop.” Even so, an American capacity for early “war termination, avoiding attacks on cities, and selective response capabilities might provide ways of limiting damage if deterrence fails.”
If the language sounded familiar to an analyst from the early McNamara years, it was no coincidence. Selin and his aides had thoroughly studied McNamara’s SIOP-63 guidance and were seeking to apply its basic principles to contemporary conditions. The military section of NSSM-3 was, in fact, composed in the summer of 1968, when Selin and his crew knew that there would soon be a new Administration. When Nixon took office, one of Selin’s aides, an analyst named Larry Lynn, was moved over to the NSC staff. When Kissinger assigned the NSSM-3 study, Lynn drafted it, using his work with Selin as its foundation.
On February 18, 1970, Nixon rhetorically asked in a foreign policy address to Congress (written largely by Lynn), “Should a President, in the event of a nuclear attack, be left with the single option of ordering the mass destruction of enemy civilians, in the face of the certainty that it would be followed by the mass slaughter of Americans?” The systems analysts in the Pentagon knew that Nixon was describing a dilemma that did not exist; that the SIOP, however imperfect, was far more flexible than he was suggesting. Nevertheless, such high-level concern about the issue prompted them to undertake a much more thorough study of options.
By this time, Ivan Selin had left government, but two deputies in Systems Analysis—a former Air Force civilian analyst named Archie Wood and an operations researcher named James Martin—persuaded the new head of Systems Analysis, Gardner Tucker, to take up the cause. Together, in early 1970, the three set up a middle-level interagency group, called the Strategic Objectives Study, to examine nuclear options. Martin had worked with Selin on the SIOP and was also inspired by the McNamara guidance of 1961–1962 composed by Ellsberg, Enthoven and Trinkl. Wood was a newcomer to Systems Analysis, but he had learned about limited options from the master of options, William Kaufmann, having been a student of his at MIT from 1963 to 1965, when Kaufmann was still very active in the McNamara Pentagon. Wood and Martin, then, set out in a direction that seemed destined to revive the old traditions of the RAND Corporation.
Two other influences reinforced this tendency. Shortly after the Strategic Objectives Study was established, Jim Martin received an unannounced visit by Albert Wohlstetter. By this time, Wohlstetter was a venerated guru in the strategic world, and Martin was highly impressed that he should be paying a call on a comparative neophyte like himself. Wohlstetter had heard about the project from his contacts around the government, and randomly dropped in on Martin several times over the next few months to chat about strategy. More than anything else, Wohlstetter emphasized that the U.S. should attack enemy cities only as a matter of last resort, that there should be a vast array of options available before taking that step—a point that Wohlstetter himself had picked up a decade earlier from Andrew Marshall and Herman Kahn.
Finally, this was the period when the RAND-Air Staff NU-OPTS project was running full steam. Jim Schlesinger and his two RAND aides, Frank Hoeber and Fritz Ermarth, had recently completed their memos rationalizing and elaborating upon NU-OPTS, and Martin and Wood read them excitedly. The papers had a profound impact on Martin’s thinking particularly. They described just the sorts of options that everyone had been looking for.
But the Strategic Objectives Study was headed nowhere. The legacy of McNamara and Enthoven had left the Systems Analysis Office with virtually no standing in the eyes of the Joint Chiefs. Any product with the endorsement of Whiz Kids would get no respect from the military. So Gardner Tucker persuaded Defense Secretary Melvin Laird to set up a high-level panel within the Defense Department to do the same sort of analysis that Martin, Wood and others had already been doing. To direct the project, Laird appointed Johnny Foster, Director of Defense Research and Engineering, the only high-level holdover from the McNamara regime but also a well-known nuclear scientist who, dating back to his days at the Livermore Lab, looked with favor on the development of new weapons systems and counterforce power.