In November, President Trump granted executive clemency to three U.S. military officers accused or convicted of war crimes: deliberately killing foreign civilians or prisoners, who posed no immediate threat, in Iraq and Afghanistan. Those pardons have been widely criticized by former U.S. military lawyers, senior military commanders and Pentagon civilian authorities, including then-Navy Secretary Richard V. Spencer, who says he was fired for his views.

But what does the U.S. public think of these pardons? Are Americans opposed, agreeing with retired Gen. Charles C. Krulak, a former commandant of the Marine Corps, who told reporters that Trump’s intervention “betrays [American] ideals”? Or does the public side with Trump, who told a campaign rally that “I stuck up for three great warriors against the deep state”?

How we did our research

To find out, we contracted with YouGov to survey a representative sample of 1,000 U.S. citizens older than 18. We used the most clear-cut case in the recent pardons — that of Army Lt. Clint Lorance, whom a military court convicted of second-degree murder for actions in Afghanistan in 2012.

Respondents were asked whether they approved or disapproved of “President Trump’s decision to pardon Clint Lorance, the U.S. Army Lieutenant who was found guilty of second-degree murder by a U.S. military court for ordering his soldiers to kill a group of unarmed Afghan civilians who posed no risk to U.S. troops.”

Of our respondents, 41 percent approved of the pardon and 59 percent did not. That proportion seems to have shifted significantly since the war in Vietnam. In 1971, Lt. William L. Calley Jr. was court-martialed and convicted of murdering 22 civilians in the 1968 My Lai Massacre. He was sentenced to life in prison. A 1971 Gallup/Newsweek poll found that 11 percent of Americans approved of the verdict.

Who supports pardons?

Like almost all other issues in U.S. politics, support for Trump’s pardons is intensely partisan. Twelve percent of Democrats and 45 percent of independents approved of Trump’s pardon, while 79 percent of Republicans approved. Although the partisan breakdown of Vietnam War-era polls is not available, the overwhelming disapproval of punishing Calley suggests that attitudes did not divide cleanly along partisan lines.

Partisanship is especially clear when respondents described the reasons for their opinion of Trump’s pardon. For example, one Republican respondent who approved of the pardon wrote, “Thankfully we now have a president that defends the military.” A Democrat argued that “Trump doesn’t understand how the military works, and his meddling in military affairs is doing extreme harm to our ideals.” Trump’s statements have also primed his supporters to favor pardons. One respondent, echoing a Trump tweet, wrote, “Because like Trump says, ‘we train these young men to fight and kill and when they do just that, they get punished.’ ”

Do Americans believe in supporting “our soldiers right or wrong”? To gauge this, we asked respondents whether they agreed with the statement that “the United States should never punish American soldiers for their actions in times of war, even if they deliberately kill foreign civilians.” Overall, 28 percent agreed. As one respondent put it, “Military men have to be supported. No matter what.” This view was also highly partisan; 49 percent of Republicans agreed but 13 percent of Democrats did.

That sentiment, too, is less prevalent today than during the war in Vietnam. A 1972 Gallup/Newsweek poll asked Americans whether they favored “amnesty for Lieutenant William Calley or any other Americans who have been convicted of war crimes in Vietnam.” At the time, 26 percent opposed blanket amnesty, while 49 percent favored it and 25 percent said they didn’t know.

Many Americans appear to believe that if troops are fighting a just war, they should be excused from responsibility for violent acts, even war crimes. Our survey finds that respondents who agree that the “United States was morally justified in going to war against Afghanistan when the war began in 2001” are significantly more likely than those who disagree to support Trump’s pardons, by 52 to 22 percent. As one pardon supporter explained, the “Lieutenant was fighting for our freedom.” Another simply wrote, “soldier is protecting our country.” Those who said the war was morally justified were 14 percent more likely to support the pardons, even when controlling for party identification as well as age, race, gender and education.

This finding is consistent with our research on American attitudes about soldiers’ moral accountability. In a recently published article (with commentary from just-war theorist Michael Walzer, philosopher Jeff McMahan and political scientist Robert Keohane), we report on a survey experiment in which a representative sample of Americans read about two make-believe countries, one that unjustly invades its neighbor and the other that justly retaliates against an unjust invasion. That study found that half opposed imprisoning soldiers who shot 48 women and children during a just war, compared with 73 percent when the same crimes were committed by soldiers fighting on the unjust side. These findings suggest that Americans’ views are influenced by their sense of retribution and revenge and subject to a phenomenon psychologists call “moral self-licensing,” in which beliefs about past good behavior allow people to justify or overlook morally dubious behavior in the present.

Trump’s pardons have sparked debates among scholars about civil-military relations, politicization of the military and the rule of law. But Trump’s pardons can also affect the battlefield if U.S. troops feel less need to follow military discipline. The empirical evidence from Afghanistan and Iraq demonstrates that when Americans kill foreign civilians in counterinsurgency campaigns, it helps the enemy recruit and discourages local populations from helping the United States. As one of the soldiers who testified against Lorance bluntly said: “If you kill local citizens, [other locals are] no longer willing to help you.” If this happens, other U.S. soldiers will pay the price for the misdeeds of those Trump excused.

Scott D. Sagan is the Caroline S.G. Munro professor of political science, the Mimi and Peter Haas university fellow in undergraduate education, and senior fellow at the Center for International Security and Cooperation and the Freeman Spogli Institute at Stanford University. He is the co-editor, most recently, with Matthew Bunn, of “Insider Threats” (Cornell University Press, 2017).

Benjamin A. Valentino is associate professor of government at Dartmouth College, coordinator of the War and Peace Studies Program at Dartmouth’s Dickey Center for International Understanding. Most recently, he is co-editor, with Jeremi Suri, of “Sustainable Security: Rethinking American National Security Strategy” (Oxford University Press, 2016).