In his many years of counting the dead — in accounting for the horrific loss of life from war and natural disasters around the world — Gilbert Burnham seldom, if ever, found survivors who slammed the door, who were unwilling to answer questions about the nightmares they had lived through.
In fact, his experience was just the opposite: Hundreds of people who had lost relatives to either acts of man or acts of God — airstrikes or hurricanes — were eager to tell Burnham’s researchers what had happened.
The cooperation of survivors is essential to accurate estimates. A husband, after all, will be able to say what killed his wife; a mother knows better than anyone why her daughter died.
“When we’re doing house-to-house surveys,” says Burnham, an expert in estimating death tolls, “people will say, ‘Are you sure these people want to talk to you? They’ve had all this emotional trauma.’ ... The interesting fact is, everybody wants to talk.”
A few months ago, Harvard researchers got a high response rate when they knocked on what remained of doors in Puerto Rico in the aftermath of Hurricane Maria. The Harvard study estimated deaths related to the hurricane at more than 4,000.
A more recent study, accepted by the Puerto Rican government, estimated 3,000 deaths from Maria.
The latter estimate was not cooked up by Democrats, as President Donald Trump alleged the other day, but resulted from a George Washington University study that looked at excess deaths over the six months following Maria’s devastation.
Trump might not like to hear this, but there’s real science to making these estimates. Gilbert Burnham has been at it a long time.
A professor of international health at Johns Hopkins University, he founded the Center for Humanitarian Health in 1998 to study disasters and train public health professionals to prepare and respond to them. Part of the center’s mission is to account for deaths in war and disaster zones around the world.
Last year in Mosul, Iraq’s second largest city, Burnham led a survey of 1,200 households that had been through the 29-month ISIS occupation and the nine months of war it took to liberate the city. The objective was to measure deaths, injuries and kidnappings in Mosul households from June 2014, the start of occupation, to June 2017, the end of the military operations to drive ISIS out.
The doctors and nurses who went door-to-door found the people they interviewed eager to talk. “In fact, the problem the interviewers had was, how do we get the people to stop talking?” Burnham says. “In Mosul, not a single household refused to talk. This is cathartic, this is a therapeutic opportunity to tell us what happened.”
And what happened, Burnham’s team concluded, was this: Far more people — 15 to 18 times as many people — were killed or injured during the battle to liberate Mosul than during the ISIS occupation. Air strikes and artillery might be more accurate than ever, Burnham says, but using them in urban areas increases the risk of death to civilians, especially, he adds, when there’s no “intelligence on the ground” to increase precision in attacks.
Every time I speak with Burnham about his work and that of others in this highly specialized field, I ask about the value of their findings. Civilian deaths in war, for instance, seem almost abstract to Americans, or something we’d rather not contemplate. Political and military leaders avoid the subject, and a prime example of that was the official reaction, in 2006, to Burnham’s disturbing findings about civilian deaths in the war in Iraq.
His team surveyed hundreds of Iraqi households and, after comparing Iraq's prewar mortality rate with its post-invasion rate, estimated that there had been 655,000 excess deaths in the first 40 months of the war. That result shocked a lot of people. Some in the media derided it, and the Bush administration dismissed it. (The White House claimed there had been 35,000 civilian deaths after three years of war.)
“In conflict or disasters, these deaths are really hard to measure,” Burnham says. But the estimates are grounded in shoe-leather investigating, knocking on doors, and deep data dives.
That’s how it was done in Puerto Rico, Burnham says. The most recent estimate, the one Trump dismissed, is based on a hard look at initial deaths from the storm as well as what Burnham calls “subterranean data,” or the hidden causes of death related to the hurricane. Those causes would not be apparent right away. But a household survey might show that someone with diabetes or asthma died because he could not get medicine, or could not get to the hospital, or maybe the hospital he managed to get to did not have electricity.