Inside Harvard’s spy lab
How a team of Harvard geeks is using a satellite—plus a little help from George Clooney—to rewrite the rules of humanitarianism.
Satellite images by DigitalGlobe; group portrait by Bob Packert; and Clooney photograph by Jonathan Hutson
LATE-AFTERNOON LIGHT slants outside the windows of a Harvard Square conference room where half a dozen twenty- and thirtysomethings huddle around a table covered with laptops, several cups of coffee, and one falafel sandwich. It could be a grad student study session, at least until a young woman named Brittany Card stands up in front of a white board covered in drawings of soldiers and tanks in Sudan.
“I’m just going to go through the sitrep from memory, so everyone’s on the same page,” Card begins, sounding more like a general in uniform than a 23-year-old in pearls and a plum-colored dress. Her situation report on the afternoon of March 27 goes on to cite massive troop movements, aerial bombardments, and a flurry of acronyms. As she talks, the group looks at satellite images of scrub desert and buildings projected on a pull-down screen at the far end of the room.
Last year, South Sudan split from Sudan, and the North African countries have teetered on the verge of war ever since. A day earlier, Card continues, the fragile cease-fire seemed to snap. It appeared that Sudan Armed Forces (SAF) had bombed an oil field in South Sudan; meanwhile, southern militias from the Sudan People’s Liberation Army (SPLA) had apparently attacked an oil field in the north. The question was, who started the fighting — and what would happen next?
Card is the data analysis coordinator for The Satellite Sentinel Project, which has been asking questions like this since late 2010, when a foundation cofounded by actor George Clooney put up the money for an audacious project to use satellites to spy on combatants in an active conflict zone. Operating out of the Harvard Humanitarian Initiative, the project’s goal is to protect civilians, but to do that the team has had to learn to think like military commanders.
“Is this a SAF play or a southern play?” asks Satellite Sentinel Project’s director of operations, Nathaniel Raymond, a 34-year-old with mussed-up hair and tortoiseshell glasses.
“Whoever’s play it is,” replies Benjamin Davies, the 34-year-old fast-talking deputy director, “we had rapid events take place” on the border.
And while they weren’t watching. Like everyone else, they had been focused on Sudan’s Kauda Valley, where the Sudanese government has hemmed in rebels and civilians alike, blocking food shipments and conducting bombing raids that drive them into the surrounding Nuba Mountains.
The team begins throwing out ideas for what could be happening. Davies theorizes the Sudan military could finally be preparing for an assault on the Kauda Valley. No, Raymond says; they would have seen more activity from all the troops in that area. “It’s like The Two Towers. You look out, and there are a lot of orcs and torches,” he says. If you’re stuck in the Nuba Mountains, he adds, “you are saying, ‘Where is Gandalf right about now? Can you text him again?’ ”
The group is fond of movie analogies. Before the night is through, they’ll reference Harry Potter, The Matrix, WarGames, and The Hunt for Red October as part of their unusual mix of war-room bravado, nonprofit earnestness, and dorm-room antics.
Suddenly a thought occurs to Raymond: What if the SAF troop buildup in Kauda is a trick to draw in southern rebels? He slams the table. “Oh man, it’s obvious. You draw them in and then you hit their flanks.” He points to an area in Sudan by the cities of Muglad and Babanusa, where tanks dropped off by train would have an uncontested route to the border. In minutes, Card finds that the nomads usually seen in the region are much farther south — that could mean they’ve been driven out. The team’s manager of imagery analysis, Isaac Baker, 32, calls up satellite shots that show roads being built from Muglad and tanks stationed in Babanusa. The evidence is mounting.
This theory would be a change from the one the Satellite Sentinel Project has been building. Just a week earlier, on March 16, Clooney and other activists led a protest in front of the Sudanese Embassy in Washington, D.C., to draw attention to the plight of the refugees in the Nuba Mountains. “It’s about to start raining, and once it starts raining there, thousands of people are going to die,” the actor said. He then crossed a police line and was arrested, and his message was broadcast everywhere from CNN to TMZ.
Now, however, the Harvard group suspects everyone was looking at the wrong spot. If their theory proves true, they will have predicted an invasion before it happened. If it’s not, they will have wasted thousands of dollars in free imagery from a satellite company that has already given them millions’ worth, and taken their eyes off the real conflict. “How confident do we feel about this?” Raymond asks.
They decide to go for it. Since the images won’t come in until tomorrow, all they can do now is wait. It’s close to 9 p.m. when the group breaks up. “It’s addictive, isn’t it?” says Jody Heck, a Harvard sophomore. “I have to study for a 10 o’clock exam tomorrow.”
USING SATELLITES TO SEARCH for war crimes in the Sudan was George Clooney’s idea. He had started making trips to the country six years ago. In October 2010, just months before South Sudan voted to declare independence, he returned with the Enough Project, a Washington-based nongovernmental organization working to end genocide and other crimes against humanity. “If entertainment is going to trump news,” Clooney says by phone from Los Angeles, “then entertainment should go where the news is.”
Whenever violence had occurred in Sudan in the past, the government had always been able to deny it. Sitting in the desert one night with Enough’s cofounder, John Prendergast, Clooney asked, “Why is it that you can Google Earth my house, but you can’t do the same thing to war criminals?” There had to be a way, he continued, they could turn satellites into the humanitarian equivalent of paparazzi.
They could certainly try, figured Jonathan Hutson, Enough’s communications director. He had previously worked with Nathaniel Raymond at Physicians for Human Rights in Cambridge, where they had used them to investigate mass graves in Afghanistan. A few nights after Clooney’s trip, Hutson found himself in the actor’s suite at D.C.’s Willard InterContinental, eating pizza and setting up a conference call with Google and the United Nations. Their goal, Hutson says, was to figure out a way to “stop a war before it starts.”
It took less than three months for the Enough team to launch the Satellite Sentinel Project, with $750,000 in seed money from Not on Our Watch, the humanitarian foundation Clooney started with actors Don Cheadle, Brad Pitt, Matt Damon and others. Hutson got Raymond to direct the operations of the project, and Raymond got Harvard to host it. Finally, commercial satellite company DigitalGlobe agreed to donate images — which can cost thousands of dollars apiece — and helped train Isaac Baker and student interns to analyze the footage.
With four staff members and a half-dozen interns, the new team quickly learned to search for clues of impending attacks. Nine months after launching the project, they detected SAF troops gathering within 40 miles of the Sudanese village of Kurmuk. After the Satellite Sentinel Project issued a report about it over the Internet, more than 1,500 villagers fled across the border to Ethiopia. By the time the invasion took place, there were few people left to kill. “We saw that coming and went all Paul Revere up in that,” Raymond says.
The project’s ability to warn civilians of impending violence “is unique in my experience,” says Stephen Wood, an ex-CIA analyst who is vice president of DigitalGlobe’s Analysis Center. Just as important, though, is how they document past abuses. “We’ve watched villages being absolutely destroyed, and being able to help explain how dire that is has been very significant.”
Last summer, for instance, the Satellite Sentinel Project alleged Sudan was killing civilians and burying them in mass graves in the town of Kadugli. Yet in a Washington Post article, the United States’ special envoy to Sudan said US intelligence reports showed no evidence such graves actually existed.
The team kept looking. “We had multiple people speaking to us saying bodies were being buried near a [particular] water tower,” recalls Benjamin Davies. One day, Ben Wang, an 18-year-old intern from Tufts, was looking at satellite images when he noticed the tower had moved. He pointed to a hole in the ground where it had been. “The grave is there,” he said.
Over the next month, the team watched the tower move back to its original place, covering up the grave. In August, they released a report and, by year’s end, Time magazine reported that the International Criminal Court was investigating war crimes based largely on information gathered by the group.
As Satellite Sentinel Project reports were increasingly being cited by Congress members and UN officials, the team began to realize something new was happening. Rather than remaining passive observers, they were affecting the actions of the combatants. The Sudan Armed Forces started hiding their tanks inside tarps and bunkers, camouflaging them not from the enemy on the ground but from a satellite 300 miles above it. Major offensives began starting on American holidays — Thanksgiving, Presidents’ Day weekend — as if the fighters hoped the people watching in Cambridge would be away from their computers. Then this past January, days after the project issued a report mentioning road construction, 29 Chinese workers helping build the road for the SAF — innocents in the conflict — were kidnapped by Sudanese rebels. They weren’t released for 11 days. “It was the sum of all fears,” Raymond says. “It’s what we work every day to avoid.”
Raymond and the others at Harvard may toil on laptops half a world away from any violence, but their work isn’t virtual. Every member of the team has had “Sudan dreams,” as they call their nightmares of shooting and being shot. “There is an immense intimacy to the violence,” Raymond says. “We are not sifting through reports to create a static archive of events. We are actually affecting the ways in which perpetrators make decisions.”
It’s a heady responsibility for a team whose eldest member is 34. But while members hasten to add that Harvard professors and DigitalGlobe analysts are advising their moves, there is something about the digital generation of activists that uniquely suits them to the task. “We could not reproduce this with people who have been trained in other [nongovernmental organizations],” says Davies. The qualities said to characterize Generation Y — the ability to multi-task on multiple technologies, a facility for social networking and teamwork, and even the individual sense of entitlement over deference to hierarchy — all help this group analyze data and make decisions quickly. “People overvalue expertise,” says Raymond, perhaps the first time those words have been spoken at Harvard. “Critical thinking and the ability to learn complex systems is more important than someone walking in with six PhDs.”
The amateur satellite sleuths were put to the test in March when Clooney was set to meet with President Obama and testify before the Senate Foreign Relations Committee. On a recent trip to the Nuba Mountains, he’d witnessed a rocket attack and interviewed children who had lost limbs from bombings — but his testimony would not be proof. And while the Satellite Sentinel Project had satellite images from the time showing smoke from alleged bombings, they couldn’t find the “shooter” — artillery or aircraft — that caused it. “We are presenting a report with the biggest movie star in the world meeting with the president of the United States,” recalls Raymond. “You want to find the shooter.”
Baker kept at it. After days of triangulating from nearby airstrips and squinting at his computer screen, he finally noticed a speck he hadn’t seen before — it was an Antonov AN-26, a Soviet-era cargo plane Sudan uses as a makeshift bomber. “This was literally the smoking gun,” says Davies.
Two days later, Clooney showed the satellite image to the Senate committee, with the Antonov outlined in blue. He credited the Harvard team with the evidence.
“Their level of expertise is incredible, but more than that is their level of commitment,” Clooney says now. “Sometimes they are up all night trying to figure this stuff out, for no other reason than they are trying to save lives. And they don’t get enough acknowledgment for that.”
IT LOOKS AS if the project staff has been up all night. It’s March 29, two days after they’d decided to search for the flanking attack, and they’re again gathered around their conference table. Eyes are bleary, and the group is quiet. They’ve got the satellite images they asked for, but not the proof of an imminent invasion. “We wanted a softball,” Davies says. “Sometimes you don’t get that.”
Baker, as he tends to do, is still searching his laptop screen for overlooked clues. Davies starts projecting Beyonce videos from YouTube on the screen. “We see a lot of dead bodies in here,” he says. “Beyonce is dead-body kryptonite.”
After more than an hour, Baker finds some new checkpoints and signs of tank movements, but still no smoking gun. Raymond prepares to call it a night.
In the days to come, Baker will find two tanks that could signal an invasion. By late April, news reports would indicate the region edging ever closer to war. After South Sudan seized the oil-rich Sudanese town of Heglig, Sudan bombed a bridge in South Sudan, killing several civilians. Despite pleas from the United Nations and African Union, the violence would continue. The team may have been incorrect about the location of attacks, but they had accurately predicted SAF was mobilizing for a fight.
That realization is still weeks away, though, and tonight the mood is somber. “Should we move [the satellite’s focus] south or stay tight on the border?” Raymond muses aloud.
The question is a fraught one. After providing some $16 million in pro bono imagery and analysis, DigitalGlobe has been negotiating new pay rates going forward. While Clooney has helped raise an additional million dollars for the project over the past year, much of that money has already been spent. That means even as violence escalates in Africa, the future of the Satellite Sentinel Project is in doubt. “I am doing the best I can to raise money through speaking engagements,” Clooney says. “I believe we are going to be able to keep this up. The question is can we find ways to sustain it.”
For now, though, it’s time for the team to regroup. “This is not a bad day; this is a good day,” says Raymond, rallying the troops with a kind of battlefield speech. Redirecting the satellite “was the right thing to do,” he says. “We needed to make sure we were seeing the whole picture.”