How Americans Feel About Going to (Nuclear) War

by Dr. Christopher Preble

Two years ago, long before a U.S. president threatened to rain “fire and fury like the world has never seen” on North Korea (channeling, perhaps unwittingly, Harry Truman), Scott D. Sagan and Benjamin A. Valentino conducted a clever survey experiment to test Americans’ attitudes on the use of nuclear weapons. They summarize their findings in the current issue of International Security.

Those expecting the public to constrain President Donald Trump’s bellicose impulses cannot be reassured by these findings. They are merely the latest reminder that the median American voter can be quite hawkish when it comes to the use of force. This includes an inclination to support military action that is likely to result in large numbers of civilian casualties.

The main object of Sagan and Valentino’s research was to test whether public attitudes on the use of nuclear weapons had shifted markedly since 1945. Some evidence suggested that it had, including changing responses to polling on the narrow question of whether the atomic bombings of Hiroshima and Nagasaki were justified. Americans asked in November 1945 about their feelings toward the bombings were strongly supportive of Truman’s decision. 53.5 percent supported dropping both bombs, and another 22.7 percent wished Truman had “quickly used many more of them before Japan had a chance to surrender.” A mere 4.5 percent were opposed, and another 13.8 percent favored a demonstration of the bomb over an unpopulated area.

By 2015, Americans were three times more likely to have opposed the use of any atomic bombs against Japan (14.4 percent), and just 2.9 percent agreed with the retributive “use many more” bombs language. Meanwhile, 31.6 percent favored a demonstration, and 28.5 percent supported Truman’s decision to drop both bombs.

Based on this and other research, some scholars have imagined a strong public aversion to the use of nuclear weapons. But the polling about Japan focused on one particular case, in a particular moment in time. Sagan and Valentino created a modern-day scenario to see if the public is as opposed to the use of nuclear weapons as some have suggested.

Their poll presents a situation in which Iran has initiated a war against the United States that has already killed more than 10,000 American service members. The U.S. president is contemplating ending the war through either a ground invasion likely to kill an additional 20,000 U.S. troops, or three different air strike options against a major Iranian city: nuclear attacks that would kill either 100,000 or 2 million civilians, or purely conventional strikes that would kill 100,000.

When asked if they would prefer each of these three air strike alternatives over the ground offensive, only the 2-million-dead-civilians option was seen as worse. Americans, based on this survey, would willingly trade 100,000 Iranians killed (either via nuclear or conventional attacks) to save the lives of 20,000 American soldiers. Moreover, solid majorities (59 percent or more) said they would support a president’s decision to attack — with either conventional or nuclear weapons — even if it wasn’t their preferred course of action.

What explains Americans’ willingness to kill large numbers of civilians? One explanation is the basic human instinct of retribution. Such feelings are especially acute when we perceive our actions as legitimate responses to aggression by others. Sagan and Valentino quote Truman on this score: “Nobody is more disturbed over the use of Atomic bombs than I am,” Truman wrote in a letter dated August 11, 1945, “but I was greatly disturbed by the unwarranted attack by the Japanese on Pearl Harbor and the murder of our prisoners of war.” Inherent in this approach to the use of force is the belief that the victims of the retributive violence are in some ways culpable in the targeted state’s actions — even though the vast majority of people killed or maimed at Hiroshima and Nagasaki played no role in the Pearl Harbor attacks or the mistreatment of American prisoners of war.

In the Sagan and Valentino experiment, a deep hatred for Iran might explain strong public support for attacks that would result in 2 million civilian deaths. These sentiments are revealed in the open-ended answers that respondents gave as explanations for why they supported killing on a massive scale. Representative responses include “They were dancing in the streets and partying in favor of what happened to the twin towers;” “Islam is a religion of fanatics [and] they need to be taught a lesson;” and “Kill the cockroaches.”

As a final test of the public’s willingness to kill 100,000 or more Iranians in order to avoid a ground war that might kill 20,000 American soldiers, Sagan and Valentino created an analog to the conditional Japanese surrender that Truman ultimately accepted. Immediately after the Hiroshima and Nagasaki attacks, and at the urging of Secretary of War Henry Stimson, Truman signaled to the Japanese government that the United States would not demand the resignation of the emperor or hold him accountable for war crimes. These concessions proved critical. As Sagan and Valentino note, Emperor Hirohito later said that without such assurances, he would have continued the war — even after nuclear bombs were dropped on his country.

Truman concealed his willingness to accept a conditional surrender from Japan, and we don’t know if Americans in 1945 would have supported such a deal in lieu of attacks that killed hundreds of thousands of Japanese civilians. Nevertheless, Sagan and Valentino created a third option for their contemporary scenario: a deal that would allow Ayatollah Khamenei to remain as a spiritual leader in Iran, but without political power, and that would grant him immunity from war crimes prosecution. The deal was offered as an alternative to the ground invasion that risked 20,000 U.S. combat fatalities and the nuclear strike that would kill 100,000 Iranian civilians. Presented with this third option, 41.1 percent were willing to accept the deal, while the percentage of respondents preferring a nuclear strike dropped to 40.3 percent.

“The finding that more than 40 percent of Americans are willing to kill 100,000 Iranians to avoid permitting Ayatollah Khamenei to stay in a position of spiritual leader,” write Sagan and Valentino, “suggests that something other than pure utilitarian assessments of the value of American versus Iranian lives influenced the assessment of options in this scenario.”

Indeed. That a near-plurality of Americans would favor the use of nuclear weapons on Iranian civilians, even if such use was not required to end the war, suggests a hunger for vengeance, not a purely rational calculation of costs versus benefits.

It is not obvious that Americans would be equally enthusiastic about a nuclear strike that killed 100,000 North Koreans — but they might be. A 2013 Gallup survey found that 9 out of 10 Americans harbored unfavorable views toward Iran, making it the least-liked country among a list that also included China, Syria, Pakistan — and North Korea. Earlier this year, however, the Pew Research Center published research showing that Americans possess “overwhelmingly negative” views of North Korea (78 percent favorable versus 12 percent favorable). But the same poll found a strong desire to continue economic pressure on the regime, and Trump’s “fire and fury” comments obviously suggest something more than mere pressure.

It isn’t clear that there would be strong public support for a preventive U.S. attack — especially a preventive nuclear attack — on North Korea or any other country. Americans may be willing to support massive uses of force that indiscriminately kill hundreds of thousands of innocent civilians in retaliation for an attack that claimed American lives, but less willing to do so if the United States initiated the use of force. The fact that the Japanese struck first in 1941 — and, in the Sagan and Valentino scenario, the Iranians did — seems to factor heavily in Americans’ willingness to visit “fire and fury” on non-combatants. But future research should test different permutations of Sagan and Valentino’s scenario. We simply don’t know if most Americans would favor the use of nuclear weapons — weapons that by their very nature would claim the lives of many non-combatants — in most instances.

For now, however, we are focused on the views of just one American, and there is little reason to believe that Trump is unwilling to use nuclear weapons. He said during the campaign that it might be appropriate to fight back with nuclear weapons after an ISIS attack, and that he wouldn’t take “cards off the table.” Now, he’s holding the cards.

That should terrify many people — not just those living in Pyongyang.


Published by War on the Rocks.

Dr. Christopher Preble

Dr. Christopher Preble

Dr. Christopher Preble is the Senior Fellow and Director of the Reimagining US Grand Strategy program at the Stimson Center and formerly was the Vice President for Defense and Foreign Policy Studies at the Cato Institute. His 30-year career also includes time spent as a Commissioned Officer in the U.S. Navy.
Download PDF