Patrick Ball is used to being the odd guy out. When he walks into the headquarters of most truth commissions, he finds himself surrounded by lawyers. As a statistician and a self-professed data geek, Ball is there to crunch numbers and “hack code,” as he puts it.
Ball heads the Human Rights Data Analysis Group (HRDAG) at Benetech, a Palo Alto, Calif.-based non-profit technology consulting firm with a social justice focus. His team brings quantitative analysis to truth commissions, which otherwise rely on anecdotes from victims, survivors and alleged perpetrators of war crimes and genocide. It's an unusual application of technology — few college students solving problem sets in statistics labs imagine they're learning skills useful in defending human rights. But when the challenge is to find patterns of violence amid conflicting claims and denials, a statistician like Ball is an invaluable ally.
“Human rights violations don't occur one at a time,” says Ball. His job is to help clarify whether violent incidents number in the thousands or the tens of thousands.
The essence of Ball's job is the ability to see ordinary material as statistical data. “Everything is data to us,” he says. “A pile of scrungy paper from border guards — 690 pages — that's data.” His team finds it, codes it, analyzes it, interprets it. But statistics is a world of careful hypotheses, not bold proclamations. Data, he says, “is what we're able to observe. That's not the same as what is true.”
Statistician Patrick Ball examines some of the 80 million pages of the Guatemalan National Police archives — discovered by accident — that show the police participated in kidnapping, torture and murder during the country's 36-year civil war. (Ann Harrison)
Still, data that Ball and his team observe — and quantify — has changed our ideas about what's true in places around the globe. In Peru, for example, Ball's team estimated that the number of dead or “disappeared” in that country's 1980s war against terrorists was twice as high as the estimate by a human rights commission in Lima. In Guatemala, the group helped prove that genocide had been committed against the indigenous Mayans. And in Kosovo, the data they collected and analyzed unraveled Slobodan Milosevic's defense during his trial at an international tribunal.
In that case, Ball used hundreds of pages of border-crossing reports, analyzing who moved across the Kosovo-Albanian border and when. The pattern that emerged from that data, combined with 11 other sources on civilian deaths in Kosovo, cast doubt on Milosevic's claim that Kosovars were fleeing NATO bombings, not Serb violence. Ball's work helped prosecutors make a case that the deaths amounted to “ethnic cleansing.”
HRDAG grew out of Ball's independent consultations with truth commissions in South Africa, Haiti, Guatemala, Timor-Leste and Peru. The group also has worked with commissions in Ghana and Sierra Leone, as well as nongovernmental human rights groups in Cambodia, Sri Lanka and Burma (also called Myanmar) and with the International Criminal Tribunal for the Former Yugoslavia.
Much of the work can be tedious, and some of the victories minor. At times, though, Ball's work surpasses even his most ambitious expectations. For a truth commission in El Salvador, for instance, he wrote software to aggregate and analyze the human rights records of army officers; the results forced a quarter of the military leadership to retire. The issue was so politically sensitive, Ball remembers, “We figured they were going to blow our office up.” Instead, the officers sued the commission — an unexpected recourse to the very rule-of-law principles that truth commissions try to enshrine.
Today, Ball's team is working on its most sensitive project yet: analyzing the 80-million-page archive of the Guatemalan National Police. Human rights advocates have long blamed them for much of the rampant kidnapping, torture and murder during the country's 36-year civil war, but there was no proof until four years ago. Then tons of police records — stacks upon stacks of musty, molding paper — were found in an abandoned ammunitions depot.
Although the team has gone through only about 10 percent of the archive, a new picture already is emerging. There is “no doubt that the police participated in the disappearances and assassinations,” says Carla Villagran, former adviser to the Project to Recover the Historic Archives of the National Police.
Changing the picture of a nation's horrific experiences can help give atrocity victims' suffering “meaning in some bigger story,” Ball says. The science of numbers can help victims separate painful histories from destructive mythologies of violence.
Still, for all the good that analyzing patterns can do, Ball knows that numbers and graphs don't say everything about the bigger story. “Statistics define the limits of what's plausible and what's not plausible,” he says. “Statistics do not tell us how it felt to be there.”