Yudkin and the NU-OPTS group back in the Pentagon were enthusiastic. However, Rationale for NU-OPTS merely described a concept, not an operational plan. What types of targets should be hit—military, economic, some mix? Where should the targets be—on the battlefield, in the U.S.S.R.? Should the target have some military significance, or did it matter? Would the mere act of setting off a nuclear explosion—anywhere—send a sufficiently clear signal to the Soviets? How many bombs should be dropped? How many would be enough to get the signal across but not so many as to make the Soviets see the strike as the opening volley of an all-out assault, prompting them to respond in kind?
Over the next several months, RAND and the Air Force embarked upon a major study. Air Force analysts worked on the technical side, compiling three volumes of highly complex calculations, figuring the optimal routes for the bombers, the probabilities of penetration and damage in various scenarios and circumstances. Schlesinger, aided mainly by two young RAND colleagues, Francis Hoeber and Fritz Ermarth, elaborated the politico-military angle, thinking about what the NU-OPTS strategy should accomplish and how best, in principle, to go about accomplishing it. All of the scenarios involved a conventional invasion or land-grab by the Soviets—in northern Norway, in the Middle East, in Central Europe—followed by the first use of nuclear weapons, on a very small scale, by the United States.
In the end, the project failed. The whole business was too uncertain, risky, hinged too much on dubious assumptions. If bombers were to be used in the strike—and the source of most officers’ interest in NU-OPTS was the rationale it might provide for a new bomber—the only way to penetrate Soviet air defenses with high confidence was through surprise. Yet the NU-OPTS concept critically depended on telling the Soviets in advance what we were about to do and what sorts of limitations we were deliberately imposing upon our actions. Or, if we did tell them ahead of time, thus forgoing the advantage of surprise, extra bombers would have to fly the mission to compensate for those that might get shot down, those bombers would have to be escorted by jet fighters, other fighter-bombers would have to accompany them to destroy air-defense sites ahead of the attack. In short, a whole armada would be involved, as many as 100 airplanes in most cases, so many that the original intent underlying NU-OPTS would be lost. The ambiguities of such a raid would be too severe; the Soviets would have no way of distinguishing it from an all-out attack, and the hope of mutual limitation—of controlled nuclear war-fighting—would be destroyed.
Aside from this operational defect, there were conceptual problems that, months of close analysis notwithstanding, remained unsolved. It was not so difficult to identify particularly vulnerable Soviet targets that might be attacked. But then what? Would the Soviets really be coerced? What if they didn’t back down? What if they fired a similar—or much larger—salvo back at the U.S. or Europe or the Middle East? What if they did nothing? How would we end the war on favorable terms?
The dissent to NU-OPTS began at RAND, among several members of the social science division who were only peripherally involved in strategy and not at all caught up in the traditions passed down by Kaufmann, Kahn, Marshall and the others in the strategic community. The debate over NU-OPTS became closely entangled with the debate then raging at RAND over the Vietnam War. Those who thought that the U.S. was bound to win the war also thought that the U.S. could prevail in a limited strategic nuclear war with the Soviets, and those who were convinced that Vietnam was a losing proposition, on the grounds that the Viet Cong and the North were ignoring our finely wrought signals, felt the same way about NU-OPTS. If the Soviets lacked the ability or desire to play along with our tit-for-tat game, they could simply threaten to raise the stakes of destruction, twisting the game to their terms.
The Air Staff officers who originally supported NU-OPTS were also growing increasingly wary. It just didn’t seem practical. Military officers tend to be conservative; they want to have high confidence that a military operation will work before they endorse it. And high confidence was the one element most conspicuously missing from any close consideration of NU-OPTS. Most of them felt, like Schlesinger, that the NU-OPTS strategy was “less miserable” than other options. For that reason, they worked hard, redoing the calculations, searching for better targets, more efficient bombing techniques, more favorable scenarios, trying to make it work. It was 1970, two years into the project, before they gave it up. It was frustrating to see the project fall apart. Its failure meant that, in practical terms, there might be nothing left except massive destruction after all, that maybe nuclear war was “unthinkable.”
Their frustration was short-lived. For just as NU-OPTS showed sure signs of failure, new technology was coming to the forefront that, on the face of things, seemed to reinvigorate the notion of a controllable nuclear-war-fighting force. The technology was the Multiple Independently targetable Re-entry Vehicle, the MIRV. During the ABM debate, it was MIRVs that the Administration feared the Soviets would soon be putting on top of their SS-9 ICBMs, thus making the U.S. Minuteman ICBMs vulnerable to a counterforce first-strike. In June 1970, the U.S. started to put its own MIRVs on a new generation of Minuteman missiles, the Minuteman III. And to many officers and civilians inside the Pentagon, the purpose was the same—to revive counterforce as a realistic option.
Very much behind the scenes, MIRV had been labored into life in the early-to-mid 1960s just as McNamara was beginning to abandon the premises of the counterforce philosophy.
It began life as a figment inside the flighty imagination of a physicist at the RAND Corporation named Richard Latter. Latter had been one of the handful of those at RAND who had helped set up the Theoretical Division of the Livermore weapons lab to work on the H-bomb in the early 1950s, and was a perpetual promoter of a U.S. nuclear buildup to defend against the Soviet threat. In 1962, Latter fiercely opposed the atmospheric nuclear test-ban treaty that the Kennedy Administration was negotiating with Moscow, and worried intensely that someday the Soviets could launch a knockout punch against the Minuteman ICBM force.
At a cocktail party that spring at the Dupont Plaza Hotel in Washington, D.C., Latter sat chatting with an old friend from Livermore named Roland Herbst and an Air Force colonel who supported the treaty. The colonel pointed out to Latter that we had more missiles than the Soviets, that we could build more if they built more, and that we had dispersed the ICBMs and hardened them in underground silos to such a degree that even if the Soviets detonated a 100-megaton bomb in between two silos, it wouldn’t destroy both. Therefore, Latter should not worry about Minuteman vulnerability. This was essentially the argument that Robert McNamara and his chief scientist, Harold Brown, had been making before Congress.
Latter objected. The number of missiles doesn’t mean a thing: the Soviets could simply put lots of warheads on each missile, just as we had done with the Polaris A-3 “Claw”; but they could do it differently, devising a separate aiming mechanism for each of the warheads so that they could be released against widely separated targets. Thus we might be equal in missiles, but their warheads would outnumber our missiles, making Minuteman vulnerable. The colonel didn’t know what to say; nobody had ever talked about anything like that before. Latter had mostly made it up off the top of his head, but the more he thought about it, the more plausible it sounded. Back at RAND, he wrote a report on how such a multiple-warhead missile might work and why it would make Minuteman vulnerable to attack.