CHAPTER 2 Image The Race Begins

Whoever won the 1960 presidential election, his term would mark the onset of the nuclear arms race as we still know it more than a half-century later. The 1960s were when the two superpowers would build and deploy nuclear weapons in large numbers, and when other, smaller but ambitious powers would join the game. Many, soon most, of these weapons would be missiles—nuclear-tipped rockets that could be launched from underground silos or undersea vessels, then streak across the heavens and plunge back down to the other side of the earth, exploding with monstrous waves of energy, heat, and toxins, all in a mere half-hour from blast-off to touchdown. There would be no time to deliberate, no time to hide. From this moment on, the shadow of the bomb would loom forever present.

When the election’s winner, John Fitzgerald Kennedy, took the oath of office on January 20, 1961, he carried away the dubious prize of getting first crack at handling the new era’s threats and challenges. His presidency seemed to offer promising prospects for all factions of the United States armed forces. Kennedy had campaigned on warnings of a “missile gap,” echoing intelligence estimates that the Soviets were far ahead in the race to build long-range nuclear missiles and that, therefore, the U.S. Air Force needed more money and more missiles to catch up. He had also absorbed the studies showing the growing vulnerability of land-based missiles and bomber bases, and thus championed the Navy’s Polaris missiles. And, like many of the Democratic Party’s national security experts, he’d criticized John Foster Dulles’s massive-retaliation policy and called for more conventional weapons and troops for the Army; he even brought General Maxwell Taylor out of retirement to serve as his special military adviser.

Finally, Kennedy touted himself as a man of “vigor,” boasting not only his youth—at forty-three, he was the youngest man ever elected president, succeeding Eisenhower, who, at seventy, was stepping down as the oldest—but also his eagerness for military action, pledging in his Inaugural Address to “pay any price” and “oppose any foe” in the pursuit of liberty’s survival. At his first meeting with the Joint Chiefs, five days later, he assured the nation’s top officers that he valued their advice and would solicit it routinely.

Kennedy’s secretary of defense was another matter. Robert Strange McNamara was just one year older than Kennedy (he too was the youngest man ever to take his job) and no less vigorous, though in ways that the Chiefs found unsettling. During the Second World War, McNamara had worked in the Statistical Control Office of the Army Air Forces, as part of a group recruited from the Harvard Business School to apply new management techniques to the planning of bombing raids over Europe and Japan.

Toward the end of the war, the group’s revamping of fueling routes gave the B-29s in Curtis LeMay’s XXI Bomber Command 30 percent more mission flight-hours, which is to say, 30 percent more time to bomb targets. The feat bolstered McNamara’s sense that the generals didn’t know everything about their business—that a numbers-smart civilian, like him, could help in ways that the high brass hadn’t been trained to comprehend.

After the war, he and his teammates sold themselves as a group to Ford Motor Company, where they rationalized production practices, swept the firm’s finances clean of debt, and earned for themselves the nickname “whiz kids.” McNamara had risen to the top of Ford Motor, as the company’s youngest president ever, before coming to work for Kennedy, and now he hired a new set of whiz kids to overhaul the Pentagon.

He found them at the RAND Corporation, the Air Force–funded think tank in Santa Monica: young economists mainly, who had written books and monographs on how to analyze defense budgets, weapons systems, and war plans rationally—assessing, for example, how to pick the right weapon, and discard the wrong ones, by calculating which of them would destroy the most targets for the same amount of money (or destroy the same number of targets for the least amount of money). These were the sorts of calculations that McNamara had performed during the war, but the RAND analysts were applying them across the gamut of national security issues. From a cursory flip through the Pentagon budget, McNamara saw a lot of redundancy. The Air Force had three bombers in production or development; the Air Force and Navy, between them, planned to produce four different types of cruise missiles; the Army, which had lost the ICBM competition to the Air Force, was developing two different models of anti-ballistic-missile systems. Certainly some of these projects were expendable.

Previous defense secretaries had set limits on how much the generals and admirals could spend, but none had restricted what the top brass could buy; no previous secretary had thought it proper to question their professional military judgment, to tell them which weapons they could build and which they’d have to give up. McNamara was determined—he was keyed up—to assert his wisdom over theirs: to force big changes. And the whiz kids from RAND would help him figure out how.

Two weeks into Kennedy’s presidency, seized in this frame of mind, McNamara flew to Omaha, then took the ride to SAC headquarters, to hear General Tommy Power brief him on the nuclear war plan.

He came away as shocked and appalled as he’d ever felt in his life.

It was the briefing for SIOP-62, the same stem-winder that Power’s staff had delivered to McNamara’s predecessor and a roomful of officers and aides two months earlier. But McNamara, the supreme numbers cruncher, saw through the glossy graphs and complicated charts; he instantly discerned the connection between the high “damage-expectancy” curves (the hundreds of targets that needed to be hit with a probability of more than 90 percent) and the “requirement” for more weapons.

He told Power that the plan seemed excessive: the number of targets struck him as staggeringly high; and hitting hundreds of them with two, three, in some cases four nuclear weapons would kick up “fantastic” levels of radioactive fallout.

The Navy’s notion of deterrence held that we needed only enough weapons to destroy the largest cities in the Sino-Soviet bloc. But in the SIOP, just 200 of the 4,000 targets were “urban industrial” complexes. The rest were military targets. Three hundred of those were air defense sites, most of them in the USSR’s “satellite” nations of Eastern Europe, which needed to be destroyed with nuclear weapons in order to plow a “corridor,” allowing Air Force bombers to fly toward their targets inside Russia without getting shot down along the way. McNamara was already close to concluding that ballistic missiles were better than bombers, precisely because they couldn’t be shot down and were, therefore, more reliable as deterrents. Now he saw another benefit of cutting the number of bombers: it would mean he could cut the number of targets and, in the event of war, inflict less damage on Eastern Europe, whose populations despised their Soviet overlords and certainly bore no responsibility for the Kremlin’s actions.

If the briefing’s substance jolted McNamara out of his customary cool, Power’s style, especially his vulgar jokes, sent him over the top. One of SIOP-62’s targets was a huge air defense radar in Albania. The bomb sent to hit it would release several megatons of explosive power, enough to kill hundreds of thousands of Albanians, even though the tiny country was politically independent of Moscow.

“Mr. Secretary,” General Power said with a chuckle, “I hope you don’t have any friends or relations in Albania, because we’re going to have to wipe it out.”

McNamara fumed with contempt.

Back at the Pentagon, McNamara recited the horrors of Omaha to some of his top aides, including a couple of the RAND whiz kids. They told him about their colleague Bill Kaufmann and his counterforce study. McNamara told them to have Kaufmann come brief him.

The briefing took place on February 10, one week after the debacle at SAC. McNamara was fascinated by Kaufmann’s concept: limiting the initial attack to the Soviets’ military targets, holding back a secure reserve force to threaten their cities, and brandishing the force as a bargaining chip to persuade the Kremlin to halt the aggression that prompted our attack in the first place. Any idea that seemed to rationalize conflict would have appealed to McNamara; counterforce held a special allure, as a saner alternative to SAC’s war plan.

McNamara expressed two qualms about Kaufmann’s idea. First, he didn’t quite see how this war ended. What if the Soviets didn’t get the signal that we were sending with the small dose of destruction and the threat of more to come? What if they simply weren’t inclined to play by the rules we were trying to lay down and, instead, retaliated with full force?

Kaufmann said that he’d mulled over the same concern and, frankly, didn’t have a lot of confidence that the strategy would work; it was quite possible that, once nuclear weapons were used, escalation to all-out war would prove inevitable. But, he reasoned, if there was even a small chance of controlling the spiral, it was a chance worth pursuing. Meanwhile, if the Soviets retaliated anyway, the fact that we’d destroyed a large portion of their missiles and bombers would mean their attack on us would wreak less damage.

McNamara’s second qualm was that he didn’t see how this strategy would help him put a cap on SAC’s voracious appetite for more missiles and bombers. The Navy’s notion of attacking only Soviet cities had its strategic drawbacks, but at least it set a cut-off point on how many weapons were needed. Counterforce seemed to have no cut-off point: if we targeted only Soviet military sites, we would have to keep buying more weapons, as long as the Soviets kept buying more weapons.

Kaufmann had no answer to this one. Neither, for the moment, did McNamara.

On March 1, McNamara issued a thirteen-page memo, asking ninety-six questions, each one assigned to a specific group or individual—some to the Joint Chiefs, some to an assistant secretary of defense or to one of the service secretaries—with unreasonably tight deadlines. Some of the colonels and generals had taken to calling the whiz kids “McNamara’s Band,” so they called this memo the “96 Trombones,” a play on “76 Trombones,” the title of a song in a hit Broadway show called The Music Man about a charlatan bandleader. (McNamara later cut the number of questions to ninety-two, but the label stuck.)

The first two questions on McNamara’s list were direct follow-ups to Kaufmann’s briefing. The first: “Prepare a draft memorandum revising the basic national security policies and assumptions, including the assumptions relating to ‘counterforce’ strikes.” The second: “Prepare a ‘doctrine’ which, if it is accepted, would permit controlled response and negotiating pauses in the event of thermonuclear attack.”

Project No. 1, due May 1, was directed to the chairman of the Joint Chiefs of Staff, Army General Lyman Lemnitzer, and the assistant secretary of defense for international security affairs, Paul Nitze. Project No. 2, due April 17, was assigned only to Lemnitzer.

Nitze delegated the first assignment to his deputy, Henry Rowen, one of the RAND whiz kids that McNamara had brought on board. Rowen, in turn, handed it to a RAND colleague, now working for him as a consultant, named Daniel Ellsberg. The previous year, Ellsberg had conducted a highly classified study of the military’s nuclear command-control network. He was one of the very few civilians ever to immerse himself in the nuts and bolts of the nuclear war-making machinery, and he emerged from his probe with deep alarm over the rigidity of the system: not only how easily a small conflict could escalate into all-out nuclear war, but how the system had been designed to ensure escalation.

It was only logical that Ellsberg would find much to like in his colleague Bill Kaufmann’s counterforce briefing. A decade later, Ellsberg would transform himself into an antiwar activist and win fame (or, in the eyes of his former colleagues, infamy) for leaking the Pentagon Papers, the Defense Department’s Top Secret history of America’s intervention in Vietnam. But for now, he was a Cold War hawk; he thought that war with the Russians was not just possible but likely. A few years earlier, he’d decided not to enroll in RAND’s quite generous retirement plan, because he didn’t think he’d live long enough to reap its benefits—he’d be killed in a nuclear war, along with millions of others, well before the plan came due. Counterforce struck him not only as a more credible deterrent than SAC’s current war plan, but also as a possible way out of total annihilation.

Ellsberg’s response to McNamara’s question, which he finished on April 7, his thirtieth birthday, drew heavily on Kaufmann’s briefing. He noted the suicidal consequences of SIOP-62’s “spasm” attack, “in which we fire off everything we can” all at once. He proposed instead a “doctrine of controlled response” to Communist aggression, giving the president the option of hitting only Soviet or Chinese military targets, refraining from hitting Soviet cities—refraining from hitting China altogether if that country wasn’t involved in the war—and holding back a secure “reserve” of nuclear weapons, which the president could threaten to launch against Soviet cities if the Kremlin didn’t back down. The explicit goal here was not merely to end the war quickly, before it spiraled out of control, but to end it in a way that didn’t leave the U.S. military inferior to the Soviet military after the first volley of attacks: to give the United States a means of “prevailing”—not just surviving but, in some meaningful sense, winning—a nuclear war. To Ellsberg, and to many of his colleagues, counterforce was both a more humane approach to nuclear war and a shrewder strategy.

The Joint Chiefs’ response was skeptical, bordering on hostile. The Chiefs were coming to view McNamara’s whole approach as a violation of their space and a danger to national security. They accepted that the president and his cabinet had the right to make broad policy on war and peace—that’s what civilian control of the military was about—but the details of tactics and strategy should be left to the professionals: should be left to them. And the whiz kids, with their incessant intrusiveness, did little to mollify their anxiety. Alain Enthoven, the chief whiz kid, a thirty-one-year-old from RAND with the appointed rank of deputy assistant secretary of defense, particularly annoyed them. Once, Enthoven ended an argument with a three-star over the nuclear war plan by bellowing, “General, I have fought just as many nuclear wars as you have.”

In this light, the Chiefs saw the “96 Trombones” exercise as another annoyance. And the first two of McNamara’s ninety-six questions, they saw as a stab at the heart of the war plans—of their province.

General Lemnitzer replied with a brief memo, speaking on behalf of all the Chiefs, dismissing the whole notion of “controlled responses” and “negotiating pauses” in a nuclear war. First, he argued, it was impractical. U.S. nuclear forces were too vulnerable to risk holding back a large number of them once a war had begun; a Soviet strike would disable this so-called reserve force, so, as a practical matter, we had to use it or lose it. Second, there was a conceptual problem. The alleged advantages of “limiting” our attack were plausible only if the Soviets cooperated—that is, if they limited their attack too—and that premise, he wrote, “does not now appear realistic.”I

Finally, Lemnitzer argued, the act of even advocating this “controlled response” doctrine “could gravely weaken the current deterrent posture.” Though he didn’t bother to elaborate, his point was clear: to the Chiefs, and to SAC, the appeal of nuclear weapons—the crux of their power to keep the enemy at bay—was precisely their massive destructiveness. Talking about restraining this power, much less touting restraint as a new “doctrine,” might tempt our enemies to flex their muscles without fear of a devastating counterpunch.

Still, Lemnitzer—who had briefly replaced Maxwell Taylor as Army chief of staff in the final year of Eisenhower’s presidency, before his subsequent promotion to chairman—had witnessed the internecine warfare sparked by SIOP-62, and he was well aware of that plan’s excessive “requirements” for weapons and damage. In the final lines of his memo to McNamara, he acknowledged that it was “desirable” to examine the possibility of adjusting the SIOP, so it could respond proportionately to a “less than large-scale” attack, and he told McNamara that he had ordered a strategic council within the JCS to analyze the idea “as a matter of priority.”

With that concession, McNamara felt that he’d wedged his foot in the door—that he’d won a say in the planning of nuclear war.

In fact, though, he was kidding himself.

By the summer, the Joint Chiefs of Staff worked up the guidance for a new war plan—SIOP-63—based, in part, on McNamara’s desire for more options and restraints. But crucial details of this guidance were couched in deliberately ambiguous language, which had the effect of blunting, in some cases reversing, McNamara’s reforms.

The guidance for SIOP-63 was written by a few of McNamara’s whiz kids and the strategic plans division of the Joint Staff, the large cadre of colonels and generals in the Pentagon who do the spade work for the Chiefs on specialized issues. The guidance split the SIOP’s existing list of targets into three categories: Soviet and Chinese nuclear weapons and their infrastructure; other Soviet and Chinese military targets that were located outside cities; and military, industrial, and governmental buildings that were inside cities.

It then outlined five discrete U.S. “attack options.” Under Option 1, the president would attack the first types of targets—the Sino-Soviet nuclear arsenal. This would involve launching only some of America’s nuclear weapons, holding back the rest, and threatening to use them against other targets if necessary—a clear reflection of Kaufmann’s counterforce idea, as embodied in Ellsberg’s response to McNamara’s “96 Trombones.”

Under Option 2, the president would attack the first and second types of targets, still refraining from hitting cities. Options 1 and 2 were both explicitly preemptive attacks: the United States would launch a nuclear first strike in response either to a Communist invasion of allied territory or to intelligence reports that the Soviets were about to launch a nuclear attack against the United States.

Options 3 and 4 assumed that the Soviets had already launched a nuclear first strike and that the president responded with the same “controlled” attacks described in Options 1 and 2.

Finally, Option 5 was the all-out attack, launched either preemptively or in response to an attack—the unloading of the entire U.S. nuclear arsenal against every target in Russia, China, and Eastern Europe.

Curtis LeMay was now in the Pentagon, serving as the Air Force chief of staff. He tore into this guidance, intent on blocking even the slightest modifications of SAC’s total-destruction philosophy. Any substantial reduction in “our strategic nuclear offensive operations,” he wrote in a memo to his fellow chiefs, would incur an “unacceptable risk” not only to a credible deterrence strategy but also to “the survival of this Nation.”

To LeMay, all the options except Option 5—the all-out attack—would leave too many crucial targets untouched. As an example, he noted that the other four options all called for nuclear attacks on Soviet air bases, including their repair depots, but that they held back from attacking the factories supplying those depots with spare parts because the factories were in nearby cities. This omission, LeMay wrote, “would permit the enemy to reconstitute significant portions” of his nuclear forces “in a minimum period of time” and thus “enable him to continue his attacks on the U.S. and its allies.”

The landscape evoked by LeMay’s critique was astonishing even by the Chiefs’ standards. Hundreds of nuclear warheads would have exploded across Russian and Chinese territory, destroying every air base, missile site, naval port, and dozens of other garrisons and facilities—yet, to LeMay, this attack would inflict too little damage, to the point where the very survival of the United States would be imperiled, because a spare parts factory was left standing, as a result of which the Soviet and Chinese air forces might be up and flying in next to no time.

General Lemnitzer overrode LeMay’s protest; the five options were allowed to stay. But LeMay objected, above all, to the fact that the plan imposed restraints of any sort on SAC’s freedom of action—that it took seriously McNamara’s demand for “controlled responses” and “negotiating pauses.” And the other Chiefs, including Lemnitzer, sympathized with LeMay on that point.

Over the next several months, SAC and the Joint Strategic Target Planning Staff drew up a new SIOP that incorporated the five options. When Power and his staff briefed McNamara on the results, he thanked the commander for a job well done—just what he and the whiz kids had ordered.

But, perhaps because he heard so many of his own buzz words in the briefer’s presentation, McNamara didn’t scrutinize the fine print or ask as many probing questions as he had at his briefing on the first SIOP. Had he displayed the same critical curiosity, he might have learned that no one, either in the Joint Chiefs or at SAC, took Options 3 or 4 at all seriously. In their minds, if the enemy had already launched a nuclear attack, it would make no sense for the United States to hold back; any weapons held in “reserve” would probably be destroyed before SAC had a chance to launch them. The only “option” would be to use them or lose them, and given that choice, the military would always choose “use them.” The idea behind Option 3 in particular—attacking only Soviet and Chinese nuclear weapons sites after a first strike against us—struck SAC officers as so preposterous, they didn’t bother rehearsing it.

In other words, SIOP-63, like SIOP-62, was, basically, a nuclear first-strike war plan.

Even the first two options—the ones that called for attacking only military targets, sparing Communist cities from destruction—wound up, in SAC’s hands, departing drastically from McNamara’s principles. The final draft of the guidance noted that damage to Soviet and Chinese civilians would be minimized “to the extent that military necessity permits.” Similarly U.S. nuclear responses would be controlled, enemy cities would be excluded from an attack, and a secure reserve of weapons would be retained “to the degree practicable.”

What “military necessity” permitted, or what was deemed “practicable,” were matters to be decided by SAC and the Joint Strategic Target Planning Staff, not by McNamara and his whiz kids or even by the Joint Chiefs of Staff. And they decided that these notions were not remotely practicable or consistent with military necessities.

Finally, the “damage-expectancy” numbers—the requirement that certain targets be damaged with 90 or 95 or even 97 percent probability, which provoked such agitation in the debates over the first nuclear war plan—remained the same in this revised plan.

As in the first war plan, the Navy’s delegates to SAC led the protest against these requirements. The plan, they said, called for too much damage against too many targets, “far in excess” of the Joint Chiefs’ requirement. For all practical purposes, they concluded, SIOP-63 was not much different from SIOP-62. It was still, by and large, a Single Integrated Operational Plan—an only slightly less rigid blueprint for an all-out nuclear first strike that would kill hundreds of millions of people.

McNamara was more prone to compromise with the Chiefs than his stiff bearing might have suggested. He and his whiz kids canceled nuclear weapons that were clearly redundant or outmoded, but when it came to the military’s core programs, he met them much more than halfway.

This was particularly true of the new Minuteman intercontinental ballistic missile. First tested soon after Kennedy entered office, the Minuteman was powered by a solid-fuel rocket, meaning it could be launched straight out of its underground silo, carrying a 1-megaton warhead from the continental United States to any target in the Communist world in thirty minutes. The Air Force wanted to build 2,300 of these missiles over the next six years; McNamara sliced the request to 1,000. This may have seemed a drastic cut—it certainly was in General LeMay’s eyes—but several officials in the White House urged Kennedy to cut the number further to 600 ICBMs.

Kennedy’s national security adviser, McGeorge Bundy, a former professor and dean at Harvard as well as an eminent figure at the Council on Foreign Relations, had assembled in the West Wing of the White House a staff of fellow intellectuals—economists, political scientists, and physicists—who he’d hoped would form the nucleus of critical thinking on national security issues. But McNamara insisted on controlling that sphere from the Pentagon, and Kennedy, who believed in giving his cabinet officers a wide lane, assented to the wishes of the defense secretary. Still, Bundy didn’t come to Washington to serve as a mere bureaucratic coordinator and note-taker, which is how most of his predecessors had defined the job; he was keen to set and engage in the debates.

His deputy, Carl Kaysen, who’d been a colleague of Bundy’s at Harvard, wrote two memos—the first for Bundy, then a more concise rewrite for Kennedy—arguing that McNamara’s proposal was excessive and could provoke a nuclear arms race. The notion that the United States needed 1,000 ICBMs, Kaysen wrote, seemed to be based on “the old estimate of the probable development of Russian missile strength.” If we look at “the newer estimates,” he went on, “we find we can accept a significantly smaller force.”

Kaysen was referring to the fact—awkward to discuss but enormously pertinent all the same—that, in the first few weeks of the new administration, the “missile gap” proved to be a myth. It was an awkward fact because Kennedy had built much of his election campaign on the charge, which he drew from Air Force Intelligence estimates, that there was a missile gap—that the Soviets were outbuilding us in ICBMs and that, as a result, a moment of “maximum danger” would soon come when they could launch a first strike with such devastation that we could not effectively retaliate.

In August 1960, three months before Kennedy’s election victory, the Eisenhower administration launched the first photoreconnaissance satellite, called the Discoverer. The camera on board the Discoverer took pictures of such high resolution that a skilled analyst could identify objects as small as one meter in diameter. The first photos were dropped down to earth and processed in November. The satellite had traversed the entire length of the Soviet Union, along every road and railroad track—any passageway that could support a gigantic missile and its launching gear. Before this, Air Force Intelligence had claimed that the Soviets had 200 ICBMs; Army and Navy Intelligence had put the number at 50. In fact, the Discoverer photos revealed that they possessed just four long-range missiles.

McNamara learned about the pictures during his first week in office. He and his deputy, Roswell Gilpatric, a former undersecretary of the Air Force and a true believer in the missile gap, walked up to the Air Force Intelligence photo shop on the fourth floor of the Pentagon and spent hours gazing at the images with the analysts, who explained what they were seeing and its implications. There was a missile gap—but, they said, the gap was very much in America’s favor.

No one inferred from this that the Russians would now sit still. But the intelligence agencies scaled back their projections for the future. Before Discoverer, they’d estimated that, by 1965, the Soviets would have 750 ICBMs and, by 1967 they’d have 1,000. After Discoverer, they cut the estimates in half—to 340 and 525, respectively.

And yet, in calculating how many ICBMs the United States needed to build, McNamara was still assuming the old intelligence numbers. If the Soviets were going to possess half as many missiles, Kaysen argued, the United States could get by with half as many too. This would be the case, he wrote, whether the goal was “a survivable second-strike capability” (as the Navy proposed) or an ability to attack the Soviet nuclear force while retaining “a reserve force… capable of threatening Soviet cities” (as the counterforce strategy required).

But the main issue, as Kaysen saw it, wasn’t merely one of waste. In setting the size of our nuclear arsenal, he wrote, “we must always consider the possibility of interaction” with the Soviets. If we built many more missiles than we needed for deterrence, the Russians might take that as a sign that we were striving for “a full first-strike capability”—and, in response, they might order a crash buildup of their own, setting off an arms race.

To Kaysen and some of his colleagues in the White House, this was a pivotal moment. The debunking of the missile gap meant not only that the Soviets had no ability to launch a first strike but also, perhaps, that they had no such desire. General Lemnitzer had recently told President Kennedy, in reply to a query, that the Soviets could not pull off a successful strike against U.S. nuclear forces for at least the next few years. The CIA had observed, in a recent report, that the Soviets seemed to have adopted a “finite deterrence” posture, possessing just enough weapons to deter a nuclear attack, not to instigate one. If McNamara’s proposal for a “large increase” were approved, Kaysen asked, “how great is the risk that we will push the Russians over to another strategic concept?”

At a meeting with Bundy and Kaysen, McNamara acknowledged that this analysis was sound. But, he said, he couldn’t recommend any fewer than 1,000 Minutemen without provoking a rebellion from the Chiefs. He reminded the two White House officials that the Air Force had wanted 2,300 missiles; he’d managed to cut that request by more than half.

McNamara’s air of triumph on this point too would be short-lived, as the Air Force figured out a way to maneuver around him. For now, though, the president sided with McNamara. Again, it came down to Kennedy’s preference for giving the secretary of defense the leeway to run his own department.

There was one question that Kennedy knew he couldn’t delegate. It was the ultimate question, which only he as commander-in-chief could answer—what would he do in the event of “general war”? Would he actually push the nuclear button?

Ten days into his presidency, Bundy wrote a memo, laying out the decisions that Kennedy would soon have to make on issues spanning “the whole spectrum from thermonuclear weapons systems to guerrilla action and political infiltration.” On the question of nuclear war, he would need to decide whether to emphasize “strike first” “counterforce” strategies or a “second strike” “deterrent” posture. “The matter is of literally life-and-death importance,” Bundy wrote, “and it also has plenty of political dynamite in it.”

The same day, David Bell, the budget director, wrote a memo noting that the three military services disagreed on these strategies. The Air Force advocated “counterforce deterrent,” requiring more heavy bombers and ICBMs; the Navy touted “finite deterrent,” accomplished by a limited number of submarines; and the Army coined the phrase “credible deterrent,” defined as more conventional forces—tanks, troops, artillery, and so forth—to minimize the chance that a small conflict might escalate to nuclear war in the first place. The implications were twofold: first, disputes over nuclear strategy would spawn battles over budgets; second, the military services, in this sense, seemed no different from other government bureaucracies—their vested interests drove their policy preferences, their top officers weren’t necessarily founts of ultimate wisdom.

In his first substantive meeting with the Joint Chiefs, on February 6, Kennedy asked about the nuclear war plan. Three days earlier, General Lemnitzer had flown with McNamara and a few of his aides to Omaha for General Power’s briefing on SIOP-62. If we launched the first strike, Kennedy asked, would that wipe out the Soviets’ ability to strike back?

Lemnitzer replied that it would not. At least a few of the Soviets’ missiles or bombers would survive the attack, and they could hit back hard.

That told Kennedy a lot about his so-called “options” in a nuclear war. There was no silver bullet, no magic escape valve from catastrophe.

Within a few months, Kennedy found himself embroiled in a crisis that could easily boil over into nuclear war. It involved a face-off over the Cold War’s most volatile hot spot, a patch of vital land that the Soviets had been threatening for years and were now threatening more ominously than before—a threat that, Kennedy’s advisers were saying, could be met only with nuclear weapons. And in the middle of this crisis came an idea, which morphed into a detailed alternative war plan, suggesting that there might be a silver bullet after all—that the United States might be able to pull off a disarming nuclear first strike against the Soviets if the Kremlin made a move. For several weeks in the summer and fall of 1961, it seemed that such a strike might be the only alternative to losing West Berlin.

I. There was another vulnerability, of which McNamara seemed only dimly aware: if the weapons themselves survived a first strike, the communications links between the president and the weapons—the links allowing him to order a launch—might be severed. This was a grave concern to SAC officers: it was one reason, quite aside from the cult of LeMay, that many of them saw little alternative, in the event of nuclear war, to firing as many weapons as quickly as possible.