Scott D. Sagan on the Nuclear Necessity Principle

This interview was recorded in February 2017, but Carnegie Council is posting it on Friday August 4 to commemorate the anniversaries of the atomic bombings of Hiroshima and Nagasaki, which occurred on August 6 and 9, 1945, respectively.

RANDALL PINKSTON: Hello. I'm Randall Pinkston. Welcome to Ethics Matter, sponsored by the Carnegie Council for Ethics in International Affairs.

Our guest today is Dr. Scott Sagan. He is the Caroline S. G. Munro Professor in the Department of Political Science and a senior fellow at the Center for International Security and Cooperation and the Freeman Spogli Institute for International Studies at Stanford University.

Welcome, Dr. Sagan.

SCOTT SAGAN: Thank you.

RANDALL PINKSTON: You are the author of several books and articles on nuclear weapons. I think your résumé runs about nine pages. I counted them. You focus on issues such as targeting, deterrence, and the ethics of using nuclear weapons.

In recent months—maybe even recent years—we've heard a lot about presidential executive orders. Today our audience is going to be introduced to a presidential directive that I suspect not a lot of Americans know about, the presidential guidance on the employment of nuclear weapons. We'll get to that in a moment.

But first let's talk about your most recent work. You are one of the co-authors of "Ethics, Technology, and War" and "The Nuclear Necessity Principle: Making U.S. Targeting Policy Conform with the Ethics and Laws of War." Both of those articles—as you know since you wrote them—appear in the Fall 2016 issue of Daedalus, the quarterly publication of the American Academy of Arts and Sciences.

The bottom line, you say, is that there is a difference between how America talks about nuclear deterrence and how it is practiced. Let's explore that perspective, beginning with one of your quotes. You write, "Technological innovations and political developments are changing the nature of warfare, posing complex challenges to traditional standards, which fall under the influence of international law and the just war doctrine."

Before we talk about the innovation and the changes, would you just please review, for those of us who maybe never learned it or forget, exactly what is the just war doctrine, and would you talk about those three main components of just war?

SCOTT SAGAN: Just war doctrine is a set of philosophical principles that most states—not all—aspire to. They include the principle of non-aggression. The United Nations is based on the idea that aggression is wrong and that you can only fight in self-defense or for international peace and security. It includes the idea of distinction, what is often called "noncombatant immunity." We know that civilians die in war, but it is illegal under the Geneva Conventions; and that is a byproduct of just war doctrine, that civilians cannot be deliberately targeted in war.

RANDALL PINKSTON: So you distinguish them from combatants.

SCOTT SAGAN: Yes. And you know that noncombatants may die because of an attack on a military target, but that's collateral damage. That is not to be used. It is permitted because you sometimes can't help it when you attack a military target. But it is not by design. It is not your purpose.

RANDALL PINKSTON: So if you have an enemy inside a housing compound and this person was responsible for, let's just say for example, the World Trade Center attacks, you go after that person and in the process you also collaterally—

SCOTT SAGAN: You might. That's where the other principles come in. Other principles include the principle of proportionality, that you have to weigh the evil of killing noncombatants against the good, the importance of the military target. While you know that civilians—noncombatants—may die, under the Geneva Conventions an attack should not be disproportionate, that is, you shouldn't attack a minor target when you know that there are going to be large-scale killings.

Now the exact proportion numbers are never defined by international law, and different militaries, or even different commanders, will make different judgments, but they all have to by law take into account proportionality. It is not enough just to say, "I'm not deliberately targeting a civilian." You also have to say, "I understand that civilians might die as collateral damage," but it should not be disproportionate to the importance of the target.

RANDALL PINKSTON: There's also a component that deals with when you go to war how you conduct your military campaign and then what you do afterward. How does that factor into the just war doctrine?

SCOTT SAGAN: Just war doctrine traditionally is divided into these three areas:

• Jus ad bellum, which is the justice of going to war, for example, only in self-defense, or in perhaps preemption if it is done as anticipatory self-defense, against an enemy who is about to attack you and you have no other alternative.
  
  
• Then there is jus in bello, justice of the fighting itself. What are the laws of armed conflict? Once you're actually in a conflict, what do you have to do? Those principles would include this noncombatant immunity principle, the principle of distinction—you have distinguish between military targets and civilian targets—the principles of proportionality.
  
  
• And there's a third one here, which is often called "risk acceptance" or "feasible precaution." Michael Walzer is really the dean of just war doctrine. In his book Just and Unjust Wars he gives an example of a young soldier in the First World War who hears activity going on in a cellar of a house that he is passing by in what had just been a German-occupied village. His instinct originally was to pull out a hand grenade and toss it in because this might be a German sniper coming out to kill him. Instead, he opens the door and shouts out, "I've got guns on this building. Come out with your hands up!" He took some risk upon himself, and it turns out there were just civilians down there.

What Walzer says is that proportionality and the principle of noncombatant immunity are not enough. We also want our soldiers to do something active to try to protect civilians, to take even some risk to themselves, to show some restraint in their actions even at some risk to themselves.

Those would be the three main components, I would say, of jus in bello.

RANDALL PINKSTON: That last component brings to mind the My Lai situation, where infantry members are confronted with children who may have blown up their friend—they are giving the children candy, and the children have a hand grenade hidden in their pockets and take out the American soldiers, and this happens more than once. So the American soldiers encounter a village, and they don't know who is friendly and they don't know who is the enemy, and so the just wipe out the whole thing.

SCOTT SAGAN: I think that was a war crime. The individual who stopped it, Hugh Thompson, and his gunner, I think were real heroes that day because Thompson hovering up ahead looks down and sees U.S. GIs lining up women and children outside the village of My Lai and just executing them. He lands his helicopter between the villagers and the U.S. soldiers and threatens that he's going to shoot on our soldiers if they don't stop.

The commander, Lieutenant Calley, said, "Get out of the way, I've got my job to do," and Thompson said, "No." Thompson was doing the right thing. Calley was doing the wrong thing.

There was a cover-up. But once the public knows about it, Richard Nixon, unfortunately, gives clemency to Calley because the public thought Oh well, he's our guy. It's war; this happens. That was wrong.

It is only many years later that Thompson is seen to be the hero that he was. He is given the Army's highest medal for bravery outside of combat, and he is the one who then every year goes to talk to the plebes at West Point in their first week, saying, "You're a warrior, you're not a murderer. We want to train you to follow orders but also to know what is an illegal order. If your commander ever does what Lieutenant Calley did and tells you, 'Go out there and wipe out that village. Everybody's a VC. Women and children are VC'—'they're Viet Cong' is what Calley said—that's just wrong. That's against the law."

I think we should feel much better today about the kind of training that our soldiers get than what they got during the Vietnam War period.

RANDALL PINKSTON: Let's talk about that presidential directive dealing with nuclear weapons. In 2013 former President Obama—he was president at the time—issued new guidance for the employment of nuclear weapons that directly correlated with a key component of the just war doctrine and the Law of Armed Conflict. Let me quote you the operative phrase from the president's statement. Mr. Obama said: "All U.S. nuclear plans must apply the principles of distinction and proportionality, and seek to minimize collateral damage to civil populations and civilian objects."

You and your co-author Jeffrey Lewis argue that President Obama's doctrine should result in a new doctrine, which you call "the nuclear necessity principle." What is that? How would it work?

SCOTT SAGAN: Let me explain. Let me take you back a little bit to the beginning to help understand how there's been a tradition in the United States, in my judgment, of obfuscating, of using euphemisms about nuclear weapons.

Harry Truman, when he announced the dropping of the bomb, said it was dropped on Hiroshima, a military base, so that we could avoid as much as possible noncombatants. Later, when he realized what this meant and the destruction involved, he privately said that he didn't want to drop the third bomb that he could have because he didn't want to kill all those women and children.

So Harry Truman understood what nuclear weapons can do if they're dropped on a city. Hiroshima did have a military industry along the edge of it, but actually we moved the target from the industrial site to the center of the town so that it would kill more people. We were deliberately trying to kill lots of people to shock the Japanese government into surrendering.

RANDALL PINKSTON: Did President Truman know about the moving of the target?

SCOTT SAGAN: No. He knew that Hiroshima, along with a number of other cities, had been left unscathed by the U.S. conventional bombing attack precisely because we wanted to blow it up in as dramatic a way as possible once we got the bomb so that we could compel the Japanese government into surrendering.

RANDALL PINKSTON: Isn't there a line of philosophical thought that says that use of a weapon in a war is appropriate if it's going to force the civilians into putting pressure on their government to lay down their arms?

SCOTT SAGAN: There are people who argue that.

RANDALL PINKSTON: And I'm not—I'm just saying. I read it in your paper.

SCOTT SAGAN: I understand. There are people who argue that, and certainly it is, I think, a reasonable argument to make about threatening to do things that you would not necessarily want to do. I think we would not want to kill lots of innocent civilians today under any circumstances, certainly not deliberately. We have signed agreements that said—the principles of distinction—that we do not deliberately target noncombatants.

What Obama did by signing that executive order on the employment of nuclear weapons was to tell the military that "even with nuclear weapons I want you to follow the principles of noncombatant immunity. Do not aim deliberately at civilian structures."

Back on this point that we sometimes use euphemisms. We've had a tradition in the United States that goes back a long ways of saying "we don't target populations per se." That was a Nixon administration-era statement. But under that doctrine, what we were doing as the guidance was to tell the military for the sake of deterrence—this is during the Soviet Union/Cold War period—that "you're supposed to target the ability of the Soviet Union to recover from a nuclear war." Because if they can't recover, that's going to make sure that they don't start it. So that was the logic of it.

RANDALL PINKSTON: That meant we were going to target their fertilizer plants and food bins.

SCOTT SAGAN: We were going to do all those kinds of things.

RANDALL PINKSTON: Starve them to death.

SCOTT SAGAN: We had this huge number of weapons and a huge number of targets, and we were doing studies to say, "Well, how quickly could they recover?" We claimed that we weren't targeting civilians per se, but our policy in terms of the way it was being implemented was to starve them to death. To me that's the same thing. That shouldn't be considered to be an ethical way of targeting.

So what Obama is saying—here is a president who really cares about just war doctrine, who uses the Nobel Peace Prize speech not to just talk about peace but also to talk about just war. "There's evil in the world," he says; "There are times where we have to go to war to combat that evil and to respond to that evil, and we will follow the principles of just war doctrine." So now he's putting his money where his mouth is by having the policy really ingrained, not just in the Laws of Armed Conflict that the military is now trained to understand, but also with our nuclear weapons policy.

What Jeffrey and I are trying to do here is to say: "Is it enough to say, "Well, we will not deliberately target civilians"? Is it enough to say, "We will use the principles of proportionality"? Shouldn't we also say, "We should not use nuclear weapons against any target in which a nuclear weapon isn't necessary to destroy that target"? Because otherwise you could say: "Well, we want to target that plant or that military base. If it's absolutely necessary for its destruction to use a nuclear weapon, then you'd have to calculate whether it's proportional." But if you could use a conventional weapon for the same purpose and it would be roughly equivalent, why would you want to use the nuclear weapon at all?

There have been some who have responded to this saying, "This is just common sense. Of course we should add that." There are others who say, "No, because this would weaken deterrence, because we want the collateral damage that could come from a nuclear attack against a legitimate target, the collateral damage of civilians nearby, We want that for the sake of deterrence."

RANDALL PINKSTON: You want to put the fear of a nuclear weapon in the enemy.

SCOTT SAGAN: But if you're using collateral damage for deterrence purposes it's no longer collateral, it's now part of your purpose.

So that's where I think the debate is today. I think it's an important one.

RANDALL PINKSTON: The necessity principle would mean what?

SCOTT SAGAN: That you should never use a nuclear weapon unless it's necessary for the destruction of that target. For example, some underground command post, a military target that has been deeply buried, or one that's mobile and we know its rough location but not its exact location. All of those might be, in my judgment, considered legitimate military targets and ones in which nuclear weapons might be necessary. But I do not think that we should respond or use a nuclear weapon first against targets that we could destroy with conventional weapons.

Let me give one possible exception, and that is the question of reprisal. This is something that I think scholars and soldiers need to do a lot more work on. If a country like North Korea uses a single nuclear weapon against us, how should we respond? The ethical thing to say would be to attack the guy who ordered it or attack the military base from whence the weapon that was used against us came. It could be that a reprisal, a reciprocal response, might be the best thing for deterrence. But that is deterrence without retribution. That is trying to think through What's the best way to end this war?

That's where I think the ethical debate about nuclear weapons should be. It should be on those very limited uses. I don't think we should be thinking about using weapons on a large scale when they're not necessary.

RANDALL PINKSTON: In a sense, when you talk about using nuclear devices and applying rules of proportionality, or even necessity, I'm wondering if it's even possible, because a nuclear weapon by its very nature is designed to destroy, obliterate, not only the target—what do you call it?—the blast zone, the radiation zone?

SCOTT SAGAN: Right.

RANDALL PINKSTON: Hiroshima and Nagasaki killed more than 200,000 people, half of them on the day the bomb dropped.

SCOTT SAGAN: The blast effect killed 100,000. Nagasaki was below 200,000, but Hiroshima, over time with the radiation effects and the fire effects, it gets to be 200,000.

RANDALL PINKSTON: So any nuclear weapon is by definition going to cause collateral damage, is it not?

SCOTT SAGAN: It depends on the target. Any nuclear weapon on a military base inside a city clearly would. But there are legitimate military targets at sea, in the desert, or a remote base somewhere where collateral damage would not be great. You'd probably still have some, and then you'd have to weigh under proportionality these questions.

But I think dropping deliberately on a city or on a military target next to a city—if possible, we don't want to even have to measure proportionality. We would rather just use the necessity principle to say, "We could hit that target with conventional weapons."

RANDALL PINKSTON: Are you familiar with—I'm sure you are—something called a daisy cutter? I was told that that is like one step below a nuclear weapon. Is that correct?

SCOTT SAGAN: It depends on how big a step you're talking about. Daisy cutter or other fuel-air explosives can be very deadly. They are usually still not as deadly as a nuclear weapon. But you can make nuclear weapons go very low in yield or very high in yield, even design them to have "dial-a-yield."

So one good thing that the Obama administration's rules, their employment policy, meant is that people now are dialing down the size of weapons, saying, "We don't need as large weapons because we have to"—following that principle—"take into account proportionality and take feasible precautions to minimize casualties."

RANDALL PINKSTON: I mentioned the daisy cutter because I was in Jalalabad, Afghanistan, in 2001 and thought that I was feeling an earthquake. Some of my colleagues who had military experience and who also had friends on the other side of the wire, as it were, said, "No, no. That wasn't an earthquake. That was a daisy cutter." We were about 15 miles from Tora Bora. That thing shook the earth. Whatever it blew up, it did a lot of damage.

SCOTT SAGAN: It is not clear to me that attacking a cave in Tora Bora with a daisy cutter—I'm not sure that would be a violation of just war doctrine. I would think, on the contrary, you're trying to go after a deeply buried target.

RANDALL PINKSTON: Osama bin Laden and the remnants of al-Qaeda.

SCOTT SAGAN: Osama bin Laden and his crew. And it's not an area with lots of villages; this is a very remote location. So I think proportionality and the principles of distinction about not deliberately targeting a village. I think you could still follow that and use a daisy cutter against Tora Bora.

RANDALL PINKSTON: You also talk in your article about the decisions that military commanders have to make to comply with the rules that are established for the employment of nuclear weapons. You say that it's difficult for them to know exactly what to do because they are dealing with civilians who have vague and sometimes not-too-logically thought-out goals to employ a military tactic, a weapon. How does that work in your necessity doctrine? What is the military commander who gets the order supposed to do?

SCOTT SAGAN: My view would be that the war plan should be designed with that principle in mind, so that if there are still targets that require a nuclear weapon for their destruction, the weapons are limited to those rather than saying we're going to have a large-scale first use or second use, and that the military commanders following those rules should have—far fewer nuclear weapons would be necessary; far more conventional weapons would be necessary under that principle.

Now, there may be some critics who would say that by making war less costly you're making it more likely. I think we have to accept that criticism. That is, by saying we're going to reduce the role of nuclear weapons in U.S. national policy, it might make it more tempting for someone to use conventional weapons. But I would rather have that temptation than have the temptation of using nuclear weapons when you don't need them.

RANDALL PINKSTON: You say that U.S. military officers want to follow the laws of war, seeking to be just warriors and not illegal killers, but they are in a nearly impossible position because of the expanded definition of "military optics," the creation of loopholes, the exceptions to rules. Any of those, or a combination of those, could produce, you say, tens of millions of noncombatant deaths.

SCOTT SAGAN: Yes.

RANDALL PINKSTON: Tens of millions.

SCOTT SAGAN: Yes. These are huge weapons, and if you're having a thousand of them go off in Russia or in China, there could be tens of millions of deaths. And I don't want the U.S. military to ever have to do that.

But we have loosened the rules. In the Laws of Armed Conflict—like all laws—you have two different kinds of rules: one you can call rules, and one you can call standards. Rules are rules that are more tight. Say, for example, if you're driving, that you can't drive above 65 miles an hour.

RANDALL PINKSTON: That's a rule.

SCOTT SAGAN: That's a rule.

A standard is I'm going to give this standard that you have to interpret, like, "Don't drive recklessly." That's a standard.

Most of what we call the Laws of Armed Conflict are standards. They require interpretation on the part of an officer. I would like to see some of those standards tightened up and become more like rules.

An example that is very relevant here is the question of what is a military target and what is a civilian target. Following the Laws of Armed Conflict, the lawyers, civilian and military, in the Pentagon have rules that they say—they have written these out, and they're not classified—that as an officer you can only target a military target. However, if you see a civilian target that you believe could be used in the future for military purposes, we will deem that to be a legitimate military target. The example that they give is a civilian airport.

In my judgment that's too loose of a rule. That is a very loose standard. You should make it a more tightened rule.

RANDALL PINKSTON: You just mentioned the civilian and military planners in the Pentagon. What is the role of civilians in establishing targets and deciding the tonnage, the throw-weight? What is the role of civilians versus military? Who controls our nuclear arsenal, I guess that is really what I'm asking.

SCOTT SAGAN: I think there has been a sea change in the degree of control on the part of civilians. During the Cold War, when I worked inside the Pentagon, civilians had relatively loose control. The president would give a two-page guidance, and it would go to the secretary of defense who would then add a little more guidance language, and then it would go to the Joint Chiefs of Staff and the Joint Chiefs of Staff would add more, and then it would go to the Strategic Air Command. It would go from a two-page document to a massive war plan. It took lots of interpretation and the civilians were often cut out of the loop.

RANDALL PINKSTON: Did it ever go back to the president for final sign-off?

SCOTT SAGAN: It went back to the president for a briefing, but it was a briefing without a whole lot of detail and without a lot of staff to explain it and to go over "this is why this and this." So you had times where we were literally attacking Moscow with 70-80 weapons. It got pretty crazy.

RANDALL PINKSTON: One city.

SCOTT SAGAN: Yes.

RANDALL PINKSTON: With 70 or 80 nuclear weapons?

SCOTT SAGAN: Yes. So it got pretty crazy because the military was doing what they're really good at, which is following the procedures that they have. They had that you have to have high damage expectance against this target, against that target. We don't know the conditions, whether we'll be shooting first or second, so therefore you want to put more weapons on in case the ones you were going to have weren't launched. And it got out of control.

Under Reagan and then under George H. W. Bush there was, I think, a really important effort for civilians to take better control over this. With a cooperative Strategic Air Command—they are the strategic command—they got so that they are actually now much more integrated. So civilians know more and are following things more.

It doesn't mean that necessarily you have to agree with everything that the civilians are doing, but it is no longer in my judgment a runaway military, a military that doesn't have sufficient guidance. It has the guidance. The question is: What should that guidance be?

RANDALL PINKSTON: Didn't you write that—I think it was General LeMay who at one point said, "There are no innocent civilians"? Was it General LeMay who said that?

SCOTT SAGAN: Correct.

RANDALL PINKSTON: That is the mindset of a military person, of someone who is a warrior.

SCOTT SAGAN: I think that's the mindset of a military person from 1945. It's not the mindset of a military officer today.

RANDALL PINKSTON: Okay. That's my question, whether—

SCOTT SAGAN: I think American military officers get it now. They want to be warriors and not to be murderers. They know that there are innocent people, and they know that their job is not just to protect American innocent people but to protect innocent people in foreign countries, even civilians who may support their government that is fighting against us.

RANDALL PINKSTON: I recently saw a movie—this is sort of getting off the topic of nuclear weapons, but it deals in part with one of the topics discussed in Daedalus about the evolution of warfare—cyber warfare, drones, robots, stand-off weapons systems where the human is sitting in a control room in Arizona dropping a bomb in Kandahar.

SCOTT SAGAN: Right.

RANDALL PINKSTON: This movie posited that the decision about whether to bomb or not bomb was not in the hand of the remote pilot but in the hands of a CIA manager at Langley somewhere who, based on their human intelligence or however they determined it, said, "Okay, this building has to be hit." Meanwhile, the pilot in the control room sees kids riding by on bicycles and all of that.

How does that get sorted out? What are the ethical considerations? And this is going on today?

SCOTT SAGAN: It goes on every day. I've been to the base in Nevada where people are making these decisions.

A couple of ethical problems come up. One is that some people say: "Well, you don't have any skin in the game. You're sitting back in Nevada killing people in Afghanistan. You're not being a warrior." My view is that "No, that's not right." I think somebody who is doing this for the United States is being a warrior and that the problem with drones is not that they're controlled remotely.

Drones are actually very discriminate if used properly. The problem with drones was that sometimes our targeting was what was called "signature attacks," that is, attacking not somebody that we knew who they were but because we saw something going on that looked suspicious. That was a recipe, I think, to kill innocent people by accident. And there were a number of incidents that you may have read about about, attacks at wedding parties or farmers who were out there and were seen from above as being combatants, insurgents putting improvised explosive devices (IEDs) on the side of a road.

The Air Force's drone policy, however, is not to attack based on a signature, it's to attack based on positive ID of an individual. I've talked to many of these people who operate the drones and they say: "Actually I know the people I attack; I've been following them. And yes, there sometimes is collateral damage. Sometimes we get somebody else by mistake, or sometimes we hit a car and there are some innocent people in it. We will do that if the number of civilians killed is not disproportionate to the importance of that target. If it's too many—the person is in a crowded marketplace—we will restrain."

RANDALL PINKSTON: That's the military, that's the Air Force. What happens when CIA operatives are in charge of fire and don't fire?

SCOTT SAGAN: The CIA was doing quite a bit of signature attacks, and that's why the Obama administration decided to pull back on those kinds of drone attacks while increasing the numbers of positive-ID attacks.

RANDALL PINKSTON: That was the Obama administration.

SCOTT SAGAN: Yes.

RANDALL PINKSTON: Of course now we have a new administration. We have a president who after election and before taking office in his early briefings—and I don't know if this; I wasn't in the room—allegedly was heard to ask, "When is it appropriate to use nuclear weapons?" Do you recall that report?

SCOTT SAGAN: I have read many reports. We don't know which ones are real and which ones are not. We don't know whether the statements that President Trump made when he was in the campaign were off-the-cuff remarks or whether they reflect his deeper instincts. But they were alarming, I agree.

RANDALL PINKSTON: How then do people like you and your colleagues who are trying to get our elected officials and our citizens to think about these issues, to reveal them? For example, who knew that the president periodically issues directives on the employment of nuclear weapons? It's not something that is headline news every day, but it has been happening since the Truman administration.

SCOTT SAGAN: The Trump administration has released an executive order that says that the Pentagon should have a nuclear posture review to review all of our nuclear doctrine and our explanatory declaratory policy to go over our deterrence and how much is necessary, etc. So they are going to be doing that.

I hope that these principles of distinction, proportionality, and necessity are front and center in that debate internally about what should our new policy be, because that's where our principles belong, at the center, at the heart of this kind of debate.

RANDALL PINKSTON: But it doesn't have to be public because the president and the Pentagon planners can do this behind closed doors because it's a matter of national security.

SCOTT SAGAN: They can.

RANDALL PINKSTON: So we wouldn't really know if there's been a change.

SCOTT SAGAN: George H. W. Bush had a classified nuclear posture review and then just a public statement, and Obama had a long public document that they did of their posture review. It's not clear what the Trump administration will do.

But I'm worried about the American public and their responses in this area. Every August there is a poll that takes place that tries to gauge the American public's support, or lack thereof, for what happened in August 1945: Were we correct? Did Truman do the right thing?

That poll shows that since its height in 1945 at 85 percent public support for using the bomb at Hiroshima and Nagasaki, it is now down to 45 percent. If you use a demonstration strike option, then support for dropping the bomb goes down to 30 percent.

RANDALL PINKSTON: What do you mean by "use a demonstration strike option"?

SCOTT SAGAN: One option that President Truman had but rejected was to drop the bomb in an uninhabited place. But he knew that they didn't have very many. He also knew that it might not go off, and that might not be good. So they decided to reject that. Given that option today, a large number of Americans would support that because it would demonstrate the power without killing a lot of innocent people.

When I see that data, many scholars and pundits have said that this means that the public has a nuclear taboo, that we're not willing to use nuclear weapons. Nina Tannenwald, a professor at Brown, has written a book about the nuclear taboo. Steven Pinker, the Harvard psychologist, has written that the humanitarian revolution is so strong that there is a noncombatant immunity, that we can't imagine killing that many people as we did, the 100,000-200,000.

Looking at that, my co-author Ben Valentino, a professor at Dartmouth, and I noticed that you can't really tell from that data whether the public has changed their views about nuclear weapons or whether they've just changed their views about Japan. Because maybe it was just hard to imagine using nuclear weapons against our friends the Japanese.

Moreover, none of those questions have ever put the audience—the participants in the study—into the situation that we were at in 1945, when we felt we had at least to choose between an invasion killing lots of Americans versus dropping the bomb to try to shock the Japanese government into surrender.

So what Professor Valentino and I did was to create a survey experiment, taking a representative sample of the American public, having them read a story that replicates in a reasonable fashion or to a reasonable degree the conditions of 1945 today. We call this paper, which is coming out in International Security, "Revisiting Hiroshima in Iran."

RANDALL PINKSTON: In Iran.

SCOTT SAGAN: The story that everyone reads, saying "this is a fictional story but we want to get you to read it and contemplate what you think we should do," is that Iran is accused by the United States of violating the Iran agreement, the nuclear deal, and we put sanctions back on. The Iranian Revolutionary Guard Corps, in retaliation for the sanctions, attacks sinks a U.S. ship, killing 2,406 people. The president and Congress declare war. The president calls for unconditional surrender.

After the initial fighting begins, he's told that, "You have two options: You can either drop a nuclear bomb on the second-largest city of Iran, and we think that that will end the war; or you could invade and march up to Tehran, and we will lose 20,000 troops if we march to Tehran. We will still win."

Then the respondents are asked, "Which of these options would you prefer; and, regardless of what you preferred, if the president chose the nuclear option, would you support it, would you approve of it?"

These are survey experiments. We use the same exact situation, except we vary one thing. So, like a real experiment, you can actually see how that one additional thing changes people's views. We have one condition where 20,000 Americans die or 100,000—like Hiroshima—Iranians; it is deliberately designed to drop on a city; we have a second in which 2 million Iranians are killed, still 20,000 Americans versus 2 million; and one where we say we're going to drop conventional weapons on the city and 100,000 will die.

What we find is that rather than having the 45 percent who say, "It was the right thing to do against Japan," we now get 60 percent of the public saying, "We would approve of dropping the bomb on Iran." If you increase it to 2 million, you still get 60 percent of the public saying we should drop the bomb on Iran. If you say, "Well, we're not going to drop the nuclear bomb, but we'll kill 100,000 people deliberately with conventional weapons," you get 67 percent of the public supporting that.

We find this to be really shocking and disappointing, because these are not collateral damage attacks, these are deliberate attacks, these are violations of the laws of war, it is a violation of our own principles, and yet the public, when it comes to revenge—

RANDALL PINKSTON: Proportionality is out the window.

SCOTT SAGAN: Absolutely.

RANDALL PINKSTON: Distinction is out the window.

SCOTT SAGAN: And that is wrong. But that is what we're finding. So this is a really disturbing piece of research that suggests that we have a lot more to do to try to inculcate these principles.

I used to teach just war doctrine and Laws of Armed Conflict by saying, "We should look at these laws, and do they comport with your moral compass, with your moral intuition?" That is how we would try to discuss them.

I now feel, however, that we need the Laws of Armed Conflict because they counter some of our moral intuitions, because our moral intuitions aren't very moral sometimes; they are filled with notions of retribution and revenge.

RANDALL PINKSTON: They are emotional responses.

SCOTT SAGAN: That's right.

RANDALL PINKSTON: They are responses to fear.

SCOTT SAGAN: Yes. In one of our experiments we asked people to describe why they chose what they did, and we found a slight increase in the number of people who made comments about Iranians being animalistic, not being human. The more people you killed in an experiment, the more they made comments like, "The Iranians deserved it. They deserve to die."

So what we suggest is that what people sometimes do is not assign the culpability of a target based on what the target has done, but how much they need to justify it because of what evil we have done to them.

Tolstoy in Hadji Murat, his novel about an insurrection, said that Tsar Nicholas was cruel to the Poles. In order to justify his cruelty he had to think the Poles were evil in proportion to the cruelty that he had done to the Poles. I think that is a very common feeling. We rationalize bad things that we do by dehumanizing the victims of what we've done.

RANDALL PINKSTON: To sum up our discussion today, you say that when it comes to the employment of nuclear weapons, the rules as they now stand—the principles of distinction and proportionality—are not sufficient to deal with the legal and ethical challenges posed by nuclear weapons, that America now, and our nuclear allies, need to change their thinking about nuclear weapons, to reduce reliance on them, and to only use them when it's absolutely necessary, because nothing else will work.

SCOTT SAGAN: That is what I believe. I think that we need to have a much more open, transparent debate about these weapons in a world in which we have adversaries who have nuclear weapons and where the proliferation of weapons to other states is a very serious problem. I don't see an easy way to get rid of nuclear weapons.

I do think, however, that the United States could have fewer of them and needs to think really long and hard about what are the purposes that these weapons might still hold. I think those purposes should not be the deliberate killing of civilians under any circumstances.

I also think that we could reduce the numbers that we need for legitimate deterrence purposes by having a much more conventional weapon orientation to our war plans. Most of our war plans are like that. I think the Strategic Command's war plans should be increased in terms of their conventional quality rather than their nuclear quality.

RANDALL PINKSTON: Dr. Scott Sagan, thank you for joining us on Ethics Matter.

SCOTT SAGAN: Thank you. It has been a real pleasure.