The Saving Game

How do you make a natural disaster in a developing nation even worse? Send in the relief workers. Now Harvard's Michael VanRooyen wants to change all that with an innovative program that uses satellite technology and tactics borrowed from the military to turn out a new breed of super-humanitarians.

By examining highly detailed satellite images of the East African landscape and corroborating what he sees with reports from sources on the ground, Baker is helping pioneer the field of human rights surveillance. He’s learned, for example, that military or police vehicles parked in the streets of a fallen city may mean the government is carrying out house-to-house killings. Checkpoints might not be set up to keep people out, but rather, to keep people in and cull the population. A lot of cratering on the images indicates buildings were destroyed by artillery fire. But burned homes without cratering or scorched earth in between often indicate that someone destroyed each house, one by one.

Baker is particularly interested in the town of Kurmuk. In late September, Sudan’s president began telegraphing his intentions to attack the rebel stronghold, a town of 10,000 located where Sudan, South Sudan, and Ethiopia meet. A few weeks earlier, Baker and the SSP team had witnessed, via satellite photos, that tanks and heavy weaponry were indeed heading toward Kurmuk. When they saw that, they issued a report five days ahead of the president’s announcement, warning of an impending attack that could “result in the use of indiscriminate and disproportionate force” against civilians. But the Sudan Armed Forces (SAF) cut right, attacking the village of Sali instead.

Last night, before Baker left the office to grab drinks at Grendel’s with his coworkers, he saw a Reuters report that the Sudanese army had finally invaded Kurmuk. Though both the SAF and the rebels are now saying that the town has fallen, Baker wants to see what happened for himself. It will be several hours before the high resolution images are fully processed.

Two years ago, Baker was new to town and crashing on friends’ couches while working at the Cellar Wine and Spirits in Cambridge. One night, Nathaniel Raymond, then the director of the Campaign Against Torture at Boston-based Physicians for Human Rights (PHR), came in to buy beer and found Baker reading a British spy novel. The two talked for a while, then Raymond asked, “How would you like to do that for real?” A few weeks later, after they’d hung out a few times, Raymond offered him an internship. The pair ended up working on an investigation into abuses committed at CIA “black sites.”

“I saw in him the type of analytic capacity and investigative intuition you can’t teach,” Raymond says of Baker. “You either have it or you don’t. It was like seeing a young Ted Williams, but in an investigative sense. You see that sweet swing and you just know he gets it.”

Raymond, 34, signed up with Physicians for Human Rights straight out of college, and within two years was deployed to Pakistan to assist with investigations into Afghanistan’s human rights abuses. Around the same time, HHI cofounder Jennifer Leaning’s team uncovered evidence of the massacre of 1,000 Afghan fighters by an Afghan warlord. Raymond was assigned to work on an investigation of the massacre, and much of the information he gathered was published in Newsweek in 2002. Later, he learned that several of the sources for the story had been murdered in reprisal. It was a stark lesson about the unintended consequences of humanitarianism.

Now Raymond and Baker work together at SSP. While Raymond runs the program, providing strategy and guidance, Baker focuses on evaluating the lunarlike images that come in from the satellite each day. His colleagues say he has superhuman vision because he can see on satellite photos what they cannot: curlicue tracks in the tank-trampled grass, trucks and other menacing vehicles painted mud red to match the color of the road they’re traveling on, or a line of troops marching on a village.

“Traditional human rights reporting is: We parachute in if we can get access — for three to four days, three to four weeks after the thing happens — and then we tell you sad stories about what [a random villager] saw when the people on the truck came to his town,” Raymond says. “Now, we’re messing with them as they’re moving.” It’s about stealing the element of surprise and offsetting their rhythm. “Our greatest, greatest goal is to get out in front by a couple hours and be able to say to the civilian and humanitarian community: ‘Get the eff out, this is happening,’” he says.

The SSP accomplished a slow-motion version of this last year, when they successfully predicted the Sudan Armed Forces’ invasion of a disputed territory called Abyei. They did so by piecing together disparate scraps of intelligence: the building of an elevated road (for moving heavy armor, they inferred); the construction of a fueling facility; and the buildup of forces and materials, including tanks, artillery, MI-24 attack helicopters, and infantry. As a result, Rebecca Hamilton, the Washington Post special correspondent in Sudan, said the invasion was “perhaps the most clearly forecast crisis in history.”

The SSP takes pains to conceal the precise locations of troops and weapons. Otherwise, says Benjamin Davies, its operations manager, “We would basically be putting [the war] on steroids. But if we say, ‘This city is going to get hit and we’re worried they’re going to use disproportionate force,’ we are giving an all-points warning: ‘An ax murderer is on the loose in Cambridge. Beware.’ We’re not saying where yet, we’re just saying he’s out there and he’s got an ax.” In this, like in many aspects of their work, the SSP team has to write its own rules; there is no playbook.

As with all HHI programs, the SSP is a collaboration among unusual partners — in this case, publicly traded DigitalGlobe provides the satellite imagery and aids HHI with its analysis, while the Enough Project does the advocacy. And though the seed money comes from Clooney, SSP’s budget is reliant on DigitalGlobe, which gives the project access to satellite imagery worth millions of dollars. DigitalGlobe is funded in large part by the U.S. defense community.

 

The SSP is an example of what technology can do for humanitarianism, says VanRooyen. “It’s non-CIA, non-state-sanctioned surveillance — it’s crazy that this is the future.” And in that way, it’s one of the game-changing tools that he’s looking for.

But as he’s learning, it’s hard to revolutionize something as deep-rooted and complex as “doing good” on a massive scale. Davies notes that his work is “culturally terrifying to humanitarians.” Case in point: Human Rights Watch’s Sudan senior researcher, Jehanne Henry, says that she doesn’t put much faith in the value of the SSP findings. But after the SSP’s report on mass graves in the South Kordofan region made headlines on CNN and in the New York Times, Henry did check out their images of the graves. “I don’t think I could take them as hard-and-fast evidence,” she says. “They looked like blobs on the screen.” She questions whether the International Criminal Court (ICC) would use these images as evidence — a hypothetical question, she says, since the court doesn’t seem to be pursuing a case in that region.

But that’s changed. On December 3, Time magazine reported that the ICC was compiling verification of recent war crimes in southern Sudan, allegedly directed by Sudan Defense Minister Abdelrahim Mohamed Hussein, the same man said to be responsible for the crimes eight years ago in Darfur. According to ICC documents acquired by Time, a “significant portion of this new investigation is based on data from the Satellite Sentinel Project, a network of private spy satellites and analysts organized by George Clooney in partnership with John Prendergast’s Enough Project. The satellites have been snapping pictures of northern Sudan since December of last year.” VanRooyen just may be on the right track. These changes are coming because they need to come, he says. From his perspective, they can’t come soon enough.