Sleepwalking With the Bomb by John C. Wohlstetter

retaliatory forces. (McNamara’s policy prescriptions to the president, however, did not actually focus only on civilian destruction but also included options for targeting Soviet missiles and bombers.) On their side, the Soviets clearly did not believe that mutual assured destruction would be enough to deter a U.S. attack. They ran a massive civil defense program, building underground shelters that could keep millions alive after a strike, while U.S. civil defense efforts, even before MAD, were minimal.
    Soon after proclaiming the MAD policy, McNamara left office. Between 1967 and 2009, the U.S. reduced its stockpile of nuclear warheads from 31,255 to 5,133, an 84 percent drop. Russia, meanwhile, continued its own arms buildup for some two decades after the U.S. froze its arsenal—the U.S. call having been made based upon intelligence estimates that were egregiously optimistic. 5
    Despite this reduction, the U.S. could not stop developing new weapons. America was faced with the plausible prospect that Russian ICBMs, if not countered, would create catastrophic U.S. vulnerability. It was a risk that the U.S. could not prudently take, and thus a bipartisan domestic consensus supported ICBM development.
    Such a consensus required support for H-bomb deployment, because only the hydrogen weapon packed sufficient explosive power to enable a megaton warhead to fit on a missile. In the 1950s and early 1960s warhead delivery accuracies were measured in miles. Megaton yields were essential to destroy surface targets a few miles away. As missile accuracy improved, thermonuclear warheads could be shrunk in physical size, while ratcheting their yield down to a few hundred kilotons for U.S. warheads; for underground targets like missile silos, improvements in accuracy were vastly more important than increasing yield.
    The next step in this shrinking-warhead progression was the multiwarhead-missile platform, the so-called MIRV. MIRVs—“multiple independently targetable re-entry vehicles”—carry several warheads, each capable of hitting a different target. U.S. ICBMs carry three nuclear warheads. Russia’s ICBMs can carry many more—its giant SS-18 ICBM can carry up to 38.
    In 1962 the first “Single Integrated Operational Plan” (SIOP)—a comprehensive war plan covering all forces and how they would be used—was presented by military chiefs to President Kennedy. He and McNamara were horrified. It envisioned a massive nuclear strike to obliterate Russia, which surely would have led to reciprocal obliteration of the United States—reflecting the attitude of the leader of the Strategic Air Command, General Thomas Power, who quipped: “Look, at the end of the war, if there are two Americans and one Russian, we win!”
    Superpower leaders who held the power of final decision were not flippant. In his study of superpower nuclear affairs,
The Dead Hand
, David Hoffman notes that Richard Nixon was similarly aghast when he saw the 1969 SIOP, with some 90 variations on the apocalypse. In 1972 Soviet dictator Leonid Brezhnev played in a nuclear war game in which it was assumed that a U.S. first strike had killed 80 million Russians and demolished 85 percent of Soviet industry, as well as leaving the military with a thousandth of its original striking power. Asked to launch three dummy Soviet ICBMs as part of the exercise, a shaky Brezhnev asked his defense minister, “Are you sure this is just an exercise?”
    Ronald Reagan, whose political opponents portrayed him as a nuclear cowboy, was equally unnerved when, two months after assuming office, he received a detailed briefing on presidential nuclear command and control matters. In
Reagan’s Secret War,
longtime policy advisers Martin and Annelise Anderson quote from President Reagan’s diary entries:
The decision to launch the weapons was mine alone to make.
     
… The Russians sometimes kept submarines off our East Coast with nuclear missiles that could turn the White House into a