Tolerance for a War's Death Toll Depends on How You Look at It

Political scientists say people are willing to tolerate higher casualties when it seems their side is winning.
Political scientists say people are willing to tolerate higher casualties when it seems their side is winning. (By Mark Wilson -- Getty Images)
By Shankar Vedantam
Monday, December 18, 2006

William Boettcher and Michael Cobb have a question for you: What is the exact number of U.S. troops you are willing to see die in Iraq?

It's a question that makes most people very uncomfortable. When political scientists recently posed a similar question to volunteers, one in four refused to answer. This was even though the scientists did not mention Iraq explicitly; volunteers were asked only how many U.S. troops they would be willing to see die in an unnamed military conflict. The most common answer was zero. The median answer -- half the people were below this number and half were above -- was 500.

What's the purpose of such a question? The scientists want to understand how people come to the conclusion that the casualties in a given war have reached an unacceptable number. Casualties have long been known to play an important role in swaying public opinion against a war, but recent experiments have found that the relationship is far from straightforward.

For one thing, most people don't keep close track of the death toll in an ongoing war. When Boettcher and Cobb conducted a national poll in September and asked people how many troops had died to that point in Iraq, nearly two-thirds of Americans were off by more than 20 percent. The current U.S. death toll in Iraq is a little more than 2,900.

Many political scientists think people decide whether the number of casualties in a war is unacceptable based on factors unrelated to the actual numbers. The real factors include whether a person thinks the war is worth fighting and whether the country is united behind it.

Political scientists Christopher Gelpi and Peter Feaver have argued that what matters is whether the United States is winning or losing. People will tolerate many more casualties when it looks as if the country is winning.

Boettcher thinks partisan loyalties play a powerful role in shaping how people think about casualties.

"Casualty unacceptability is only somewhat related to the number of actual casualties," Boettcher and Cobb said in a joint e-mail. "If you oppose the war, dislike President Bush, are a Democrat, would like to increase troops, would like to decrease troops etc... you may find casualties unacceptable without having any knowledge of the actual number."

The way casualties are framed can also influence how we think of them. It is sobering, for example, to say that the number of U.S. troops who have died in Iraq is now close to the number of Americans who died in the Sept. 11, 2001, attacks. But the losses in Iraq are less than one-tenth the size of U.S. losses in Vietnam and less than one-hundredth the total of U.S. losses during World War II.

Boettcher and Cobb, who work at North Carolina State University, recently came up with mock newspaper stories to show how framing the same information differently can influence whether people think the war is going well.

The fake news articles described a battle in Karbala. Two-thirds of the volunteers in the experiment described the battle as a success when the article they were given said 25 Americans and 125 insurgents had been killed. But a little more than a third of the volunteers called the battle a success when the article said 25 Americans were killed without any mention of insurgent casualties.

You can see why it would be tempting for the Bush administration to mention U.S. deaths in the same breath as insurgent losses. President Bush did just that last week, when he reminded the country that "the enemy has also suffered." In just the past three months, he said, "we have killed or captured nearly 5,900 of the enemy."

The comment was not an isolated example. Bush has embraced Gelpi and Feaver's theory that perceptions of victory are central to shoring up public support for the war -- and tolerance for casualties. Feaver, in fact, now works in the Bush administration.

Gelpi, a Democrat who was opposed to the war from the start, said he and Feaver disagree about whether it was a good idea but agree that the course of the conflict has borne out their argument. When the U.S. military stormed into Baghdad, public opinion was solidly behind the war, despite the casualties. As civil unrest in Iraq has grown and the public outlook on the war has become pessimistic, tolerance for casualties has fallen.

Bush has continued to voice optimism, of course. But there are good reasons why framing U.S. losses in the context of insurgent losses can be ineffective in the long run. For one thing, this approach was widely discredited during the Vietnam War, after it became clear Pentagon projections of enemy casualties were inflated, said Boettcher, who recently published a paper in the Journal of Conflict Resolution titled, "Echoes of Vietnam: Casualty Framing and Public Perceptions of Success and Failure in Iraq."

Besides, the political scientist added, the Iraqi insurgency seems to have a fairly robust capacity to replace those killed or captured with fresh recruits, so it isn't clear anymore what insurgent losses mean in terms of progress toward victory.

"You say 5,000 insurgents died and last week you said there were 15,000 insurgents, so that means there are 10,000 left," he said in an interview. "But three months later, there are still 15,000 insurgents."

Or as the satirical newspaper the Onion put it earlier this year in a fake headline: "Eighty Percent of Al-Qaeda No. 2s Now Dead."

© 2006 The Washington Post Company