In this interview, Bulletin contributing editor Dawn Stover speaks with Fred Kaplan about his just-published book, The Bomb: Presidents, Generals, and the Secret History of Nuclear War (Simon & Schuster). Kaplan is a national-security columnist for Slate and the author of five other books, including The Wizards of Armageddon, a 1983 book on the origins of American nuclear strategy. He has a PhD in political science from MIT.

The Bomb offers an entertaining, detailed, behind-the-scenes look at how presidents, from Truman to Trump, and their advisers have grappled with nuclear weapons. In the end, many of them have been flummoxed by how to avoid using nuclear weapons for anything other than deterrence while simultaneously developing the nuclear warfighting plans that give deterrence its teeth. This “rabbit hole,” as Kaplan calls it, has been difficult for presidents to scramble out of.

Kaplan describes a July 2017 meeting in “the Tank,” the Joint Chiefs’ conference room at the Pentagon where Trump not only unloaded on Cabinet secretaries and generals who were trying to school him on military history and policy, but also questioned why he couldn’t have as many nuclear weapons as past presidents had. Kaplan also talks about the Trump administration’s first-strike war plan for responding to North Korean missile and nuclear weapons testing, the massive overkill built into US nuclear plans targeting the Soviet Union during the Cold War, the conundrum of “limited” nuclear war, and why John F. Kennedy—who saw only one way out of the rabbit hole—was the smartest president when it came to nuclear weapons.

Dawn Stover: Let’s start with our current president, who, before he was in office, told the New York Times that the biggest problem in the world, as far as he was concerned, was nuclear weapons and proliferation. But since he was elected, he has tweeted that the United States must greatly strengthen and expand its nuclear capability. What changed?

Fred Kaplan: I don’t know if anything systematically changed with Trump. He doesn’t really think deeply about many of these things. Other authors have reported various aspects of the now-famous meeting that Trump had, early on, in the Tank with all kinds of officials and generals. But one thing that I learned is that, at one point, they showed a chart of nuclear weapons over time. At our peak, the United States had more than 31,000 nuclear weapons in 1967, and now we have about one-tenth that number. Trump’s reaction was: “How come we don’t have as many nuclear weapons as we used to?”

It was explained to him that, well, there have been arms control agreements and we don’t really need these weapons anymore because we’ve built up conventional defenses and so forth. And he seemed to absorb that. But then I’m told that, about a week later in a meeting in the White House, he brought up this chart again, basically saying, “How come I can’t have as many nuclear weapons as some previous presidents have had?” He brought it up again one or two other times.

One theme of my book is that several presidents have faced crises in which they’ve had to contemplate the use of nuclear weapons, and through most of this history, presidents have actually delved very deeply into the logic of nuclear deterrence and nuclear war fighting. They’ve really absorbed where this could lead, and they’ve all decided to scramble out of this rabbit hole as fast as they can.

DS: It’s really striking how many of them changed their minds quite fundamentally.

FK: Right. But the danger with Trump is that he does not think deeply, and the most frightening thing about it might be that he could succumb to what used to be called the “clever briefer,” who could outline a superficially plausible course to use these weapons in a way that might improve our standing or to win a war.

DS: Is there someone in the administration like that right now? A clever briefer?

FK: You never know, it could be somebody from deep in. What we have right now in the administration are just people who, at least on senior levels, do whatever Trump wants them to do. And it also seems that Trump doesn’t want to get into a war, which is not to say that he might not find himself dragged into one.

DS: Let’s go back to that question Trump kept asking, why he couldn’t have as many nuclear weapons as previous presidents had. If he was sitting in the room with you today, how would you answer that? Why shouldn’t he have more?

FK: Back in 1967, we didn’t have as many conventional defenses, and it was before any arms-control treaties were signed. You don’t need that many now. You didn’t really need that many then. Nukes were the centerpiece of our defenses, and the Pentagon built more and more and more.

DS: How likely do you think it is that Trump might actually use a nuclear weapon at some point during his presidency?

FK: I think this is a very low-probability event at any moment in history, which doesn’t mean that it’s a zero-probability event. One thing I uncovered was about the incident that got a lot of people frightened and motivated me to write this book: Six months into Trump’s presidency, when he threatened to rain “fire and fury” on North Korea, he threatened to do that not if they attacked us, which would be another matter, but if they continued to make threatening remarks and continued merely to test missiles and nuclear weapons. I learned that this was in the wake of some very serious war planning that had gone on. Trump had demanded a new war plan to go against North Korea as a first strike, and this was quite serious.

That year the North Koreans conducted about 15 missile tests, and during each one there was a conference call among the various four-stars and commanders, the same kind of conference call there would be if there was warning of, say, a Russian missile attack on the United States. The secretary of defense was given advance authority to fire short-range conventional ballistic missiles at the test site in North Korea if the test looked like it might be provocative. That could destroy the missile site and possibly kill some North Korean leaders. Kim Jong-un for example, frequently likes to attend these tests. On two occasions, [then-Defense Secretary Jim] Mattis did fire two missiles from South Korea, not at North Korea but out into the Sea of Japan in parallel with a North Korean missile.

DS: As a demonstration of what we could do if we wanted to?

FK: Yeah, as a demonstration. And there are several military officers who were quite nervous about all this, because there were some people in the White House who thought that the United States could give just one punch, give Kim Jong-un a “bloody nose,” and he’d be so shocked he’d back off. But many military people feared that he might retaliate and this could lead to war. It was a tense situation and a much riskier plan than people realized at the time.

DS: So much of the strategizing around The Bomb seems to be based on guesses and assumptions about what people like Putin or Kim Jong-un are thinking and whether they’re bluffing. How much of nuclear war-planning really just comes down to human intelligence, and are we that good at it?

FK: There was a hearing that wasn’t covered very much at the time and has been forgotten since, around this same time as the “fire and fury.” People in the Senate who hadn’t really thought about nuclear war and nuclear weapons for decades, and there really wasn’t much reason to, suddenly realized, “Oh my God, the president has the power to launch nuclear weapons all by himself without any permission or review from anybody else.” Trump was seen as a wild card, and so there was a hearing and it was the first hearing held in Congress on presidential launch authority since the mid-1970s. And it’s strange because one Democratic Senator said, “Look, we’re holding these hearings because the president is unstable. He has poor judgment.” I mean, all but saying that he was crazy. And the interesting thing, if you go back and read the transcript of this hearing, is that none of the Republicans on the panel disputed this point. The retired general who was there, Robert Kehler, who had been strategic command commander just a few years earlier, was frustrated by this hearing because a lot of the senators were raising issues about whether the command structure could be trusted, but not taking any responsibility for it. He told them, “Look, Congress can change the authority if you want to.” But nobody was willing to do that, and he thought it was particularly dangerous to raise doubts about the reliability of the command structure without doing anything about it.

DS: I don’t think you mentioned this in your book, where you write about the meeting at the Tank, but it was reported that, in that same briefing, Trump asked why, if we have nuclear weapons, we can’t use them.

FK: Yeah, I’d read that as well. I couldn’t get that confirmed. But look, it’s a question that many people, new to the subject, embrace.

DS: It does seem like a question that really gets at this fundamental dilemma that you talk about in your book, which is that these weapons are meant to be brandished but not used. And yet you can’t credibly brandish them without making plans for how to use them.

FK: Well, that’s kind of an interesting feature of this. The [US] nuclear war plan up until the ‘60s was, if the Soviets or the communist Chinese invaded some area that was in our vital interests, say West Germany or West Berlin, not using nuclear weapons but crossing the line, the policy was to unleash our entire nuclear arsenal against every target in the Soviet Union, the satellite nations of Eastern Europe, and China, even if China had nothing to do with the war. And it was asked how many people this would kill, and the estimate was 285 million people.

So then what happens in the early ‘60s is that the Soviet Union starts to develop its own nuclear arsenal. And certain people say, “Well wait a minute, this is getting a little crazy. If they invade Western Europe and we clobber them with nuclear weapons, they can clobber us with nuclear weapons. So it’s a policy of suicide.” So some officials and strategists started thinking about ways to use nuclear weapons in a limited way, more like a military weapon—and in a way that might give incentives to the Soviets to also fire back, if at all, in a limited way, and at least try to end the war before catastrophe is unleashed. This made a certain amount of sense, although it was never really proved that the Soviets were interested in this kind of thing at all, or had the ability or desire to go along with this game.

But a certain dynamic was set in place. To make it an effective deterrent, you had to act like you really would use nuclear weapons and therefore you had to have plans in place to use them, and you had to have weapons that would allow you to use them. So as this evolved over time, nuclear deterrence and nuclear war fighting became almost indistinguishable—and that’s the rabbit hole that some presidents in times of crisis have tried to scramble out of. Once you accept a couple of premises, you can get caught down this rabbit hole very quickly, where it almost becomes an inevitable thing that you end up using these weapons, unless the president or his adversary makes a very deliberate effort to undo the logic chain that he’s locked into.

DS: Looking at all these different presidents as you have, do you have a sense about which president had the best handle on nuclear weapons? Who do you wish was in charge of the arsenal today?

FK: I think, just without question, President Kennedy.

DS: Why?

FK: His wisdom in this is still underestimated by a lot of historians. The thing that we have with Kennedy is not just documents but tape recordings. He taped, for example, all the deliberations during the Cuban Missile Crisis. It’s interesting because Kennedy came into office believing in the missile gap, this Air Force intelligence estimate that the Soviets were way ahead of us on missiles. In his first week in office with the Joint Chiefs of Staff, [Kennedy] said he wants to meet with them regularly. He’ll take their advice as the first thing. And then he went through a few crises over Laos, Berlin, and Cuba, and came to the conclusion that these guys weren’t as smart as he thought they were. And at the same time he also believed that a war with the Soviet Union would almost certainly escalate to a nuclear war.

After the Cuban Missile Crisis, he realized that the only thing that really could be done is to try to end the Cold War. He and [Soviet leader Nikita] Khrushchev took some concrete steps toward doing just that. That ended when [Kennedy] was assassinated in November of ’63, and a year later Khrushchev was ousted. The nuclear arms race really gets going after that, in 1964. It was a tragedy that’s even much greater than we thought.

I just want to elaborate on one point. Shortly after the Cuban Missile Crisis, Kennedy was meeting with Secretary of Defense Robert McNamara and General Maxwell Taylor, the Chairman of the Joint Chiefs, about the next year’s defense budget, especially for nuclear weapons. At one point, and this was on tape, [Kennedy] says, “I don’t know why we’re buying so many more nuclear weapons. It would seem to me that just having 40 missiles that could destroy 40 Soviet cities would be enough to deter them. I mean, when they had 24 missiles in Cuba, that was enough to deter me from doing a lot of things.” But then as the conversation continues, he says, “Well, I guess if deterrence fails, I guess I would want to go after their missiles and I guess I might need more than 40 weapons to go after their missiles.” What he does right there is to sum up the dilemma of nuclear strategy: On the one hand, to deter nuclear war, you want to impress the opposition that you will destroy them if they do anything aggressive. At the same time, things can get out of hand, and if deterrence fails, you don’t want to destroy them if they can destroy you in retaliation. So you have to come up with some limited plan that you might put in motion. And Kennedy didn’t like that situation, because even the limited attack might escalate to all-out war. The only way that he could see out of this was to end the Cold War. And we’re still in this same dilemma.

DS: You say there’s no escaping it. That’s how you ended your book. You don’t have any hope for abolition, even though the UN has passed a ban treaty?

FK: Well, it would have to be preceded by some upheaval in world politics that can scarcely be imagined now. Especially now that the genie is far out of the bottle and we’re not talking about just two or three powers, but more than a half dozen—and a dozen more that could make nuclear weapons if they wanted to. A good question is, why hasn’t there been any use of nuclear weapons since Hiroshima and Nagasaki? I think it’s a couple of things. First, nuclear deterrence really does work to some degree. I think one could list a few wars that might otherwise have happened had it not been for nuclear weapons, especially some wars between India and Pakistan.

But one thing I do in this book is to describe some very near misses—and what kept them as near misses instead of escalations to war were shrewd leaders and in some cases just blind luck. The alarming thing is that we can imagine a convergence of slow-witted leaders and very bad luck, and the combination of those two things could be disastrous.

DS: The advance materials for your book say the biggest surprise for you when you were doing your research was how much overkill was built into Cold War plans.

FK: Yeah, way more than we think. This wasn’t really revealed until the late ‘80s when there was a civilian in the Defense Department, named Frank Miller, who got permission to take a very deep dive into the SIOP, the Single Integrated Operational Plan, which is the nuclear war plan. [Miller was] looking at just what the targets are and how many weapons were aimed at each target, and what was the formula that determined how many weapons were aimed at which targets, and he and his team discovered some amazing things. I mean, there were something like 700 nuclear weapons aimed at Moscow, each with around the explosive power of 1 megaton. There was a Soviet airbase in the Arctic Circle that couldn’t even be used for three-quarters of the year because it was too cold, and there were 17 nuclear weapons aimed at this base. There was an anti-ballistic missile (ABM) site in Moscow that, as we discovered after the Cold War was over, couldn’t have shot down anything; there were 69 nuclear weapons aimed at this ABM site.

And the really revealing part of this was during the George H. W. Bush administration, when the US and the Soviets were negotiating an arms reduction treaty, Miller asked one of his people out at SAC [Strategic Air Command] headquarters near Omaha, “If we reduced the number of nuclear weapons to such and such an amount, would you still be able to accomplish your mission?” And the officer said, “Well that’s not the kind of question that we deal with. We take the weapons that we have and we assign them to the targets that we’ve listed.” There was a SAC commander named General John Chain in the ‘80s who once said at a congressional hearing, “I need 10,000 weapons because I have 10,000 targets.” People who heard that thought that he was either joking or just wasn’t very bright, but no, that was the mechanics of how this was done. It was completely out of control. It was a broken apparatus that just followed a completely circular logic where policy didn’t really even enter into things.

DS: Frank Miller turns out to be quite an interesting character in your book, because he’s the guy who later is so instrumental in pushing for the low-yield weapons that were just deployed by the United States.

FK: He had read all of the documents over the ages, where Secretaries of Defense were calling for limited nuclear options. He comes into the Pentagon, he hears the SIOP briefing, and there’s no mention of limited nuclear options. Frank wanted to whittle down the size of the nuclear arsenal and to make its targeting more rational, not because he was keen on nuclear disarmament or nuclear arms control, but more to make limited nuclear options truly limited.

One premise of that is that if you fire a nuclear limited strike, the Soviets or the adversary, whoever it is, if they respond at all, will also be restrained. But [Miller] had somebody in the Defense Intelligence Agency do an analysis of the Soviet air defense early-warning radar, and he asked the analyst how many missiles have to be in the sky before the Soviet radar just sees it as one big blob, a massive attack. It turned out the answer was 200, at the time. In other words, if we launched 200 missiles, the Russians would not be able to distinguish it as a limited attack. And at that time, the smallest limited nuclear option that we had involved firing 900 missiles. And so one thing that he did was to get a reduction in the plan so that you could fire, say, 20 missiles under certain circumstances. Before then, if a president ordered a limited nuclear strike, SAC would launch a massive strike.

DS: But do limited nuclear options actually make us safer?

FK: There are two views on this. On the one hand, maybe we can stop a nuclear war before it gets out of hand. On the other hand, if the president thinks he can get away with a limited nuclear strike, especially when using low-yield warheads, maybe he’ll do it. If there is too close a convergence between nuclear weapons and non-nuclear weapons, or nuclear war and non-nuclear war, he might feel more comfortable about slipping over the line.

DS: What are the scenarios where a high-yield conventional weapon just won’t do the job but a low-yield nuclear warhead could?

FK: I wrote about a war game, an exercise during the Obama administration, that hasn’t been reported anywhere before. The National Security Council ran a scenario where the Russians attack NATO and use a small nuclear weapon to try to thwart our conventional defenses. What do we do? And the deputies meeting decided that we should just continue responding with conventional weapons.

DS: That would be good enough?

FK: Yeah, one official made the point, “Look, we’re missing an opportunity here. The Russians used nuclear weapons for the first time since 1945, we could rally the entire world against them. This would just be an enormous setback from a global political perspective for the Soviet. But if we respond with nukes, we remove that.”

DS: The United States would be going low, too.

FK: Yeah. And if we do use nuclear weapons, what do we aim them at? How does this stop or win the war? Nobody could figure that out. And so that was the recommendation from the deputies group. When the principals meeting took it over—and these are the actual Cabinet secretaries and Chiefs of Staff—they couldn’t come up with any answer on where we would aim these nuclear weapons. But they roundly rejected the deputies’ view and concluded that if we do not respond to a nuclear attack with a nuclear counterattack, then our credibility would be destroyed. It’s long been a debate whether we should declare a no first use policy, and some people on the principals committee just thought it was bizarre that we might consider a no second use policy as well.”

DS: I thought it was interesting how Frank Miller was involved with this group of civilians that were instrumental in getting some deep reductions in the arsenal, and then later bringing this limited option into deployment. Does the effectiveness of just a few people, working within the system, suggest that treaties are overrated?

FK: Treaties are valuable in that they lock things down. The Joint Chiefs signed onto the Strategic Arms Reduction Treaty readily because they had witnessed Miller’s work and saw that we really don’t need all these weapons. But you’re right in this sense: When treaties are being negotiated and when they’re up for ratification in the Senate, which requires a two-thirds majority, the Joint Chiefs and Republicans have used that as a bargaining chip to get more weapons than they might otherwise have been able to get. Jimmy Carter was forced to accept the MX missile, which he loathed, as a tradeoff for getting ratification on SALT [Strategic Arms Limitation Treaty] II. Obama agreed to modernize all three legs of the triad as a tradeoff to getting New START. There are some people who think that a better way to do this is to just have what used to be called reciprocal unilateral reductions, and then lock that in with a treaty rather than go about it as a treaty.

DS: Because that wouldn’t need approval from Congress?

FK: Yeah, because you just take it out of the institutional framing and remove the power of people who can use it as a bargaining chip. Kennedy and Khrushchev did some of this after the Cuban Missile Crisis. The problem, of course, is that you have to have pretty good political relations between the powers who are doing this. And right now, I think it would be a terrible thing if Trump did not extend New START, because the relations among the US, Russia, and China right now are terrible, and without the restraints of New START, both sides could get wrapped up in another round of an arms race.

Editor’s note: This interview has been condensed and edited for clarity. Fred Kaplan will be speaking about “The Presidency and The Bomb” at the Chicago Council on Global Affairs on February 5, and @BulletinAtomic will be live-tweeting the event.