Cellphone Traffic Is the Crystal Ball for Violence Here …
WHY YOU SHOULD CARE
In this West Africa nation, cellphone traffic picks up just before the violence.
When violent clashes between rogue soldiers of the Ivory Coast army and civilians in the western town of Vavoua killed six people in December 2011, the U.N.’s top representative in the nation was sharp in his condemnation.
But the U.N.’s peacekeeping operations in Ivory Coast (the UNOCI) could have turned to an unusual tool for predicting — and perhaps even preventing — the tragedy: the town’s cellphone traffic in the days leading up to the clashes.
Globally, phone networks are clogged with traffic immediately after religious riots, civil unrest or terrorist attacks as people either check up on loved ones or reassure them. But in western Ivory Coast, researchers have found something else:
Local mobile phone traffic increases an average of 10 percent in the four days before small-scale acts of violence, and the number of active cellphone users goes up by 6 percent.
What’s more, during that four-day window, cellphone traffic patterns are different from those in other periods when people speak more on the phone — before soccer matches and festivals.
The findings suggest that, unlike with surprise terrorist attacks, local communities in Ivory Coast often have an inkling of rising tensions that may spill over into violence, or recognize that something unusual is afoot, according to the researchers. But the results also point to an unexpected predictive weapon that international peacekeeping missions can use in countries rife with this kind of rampant localized violence.
When the violence is meant to intimidate, it is often preceded by subtle hints and threats, meant to add to the sense of fear.
Sera Linardi, behavioral economist, University of Pittsburgh
“U.N. peacekeepers can go in before an event instead of after it,” says Sera Linardi, a behavioral economist at the University of Pittsburgh and one of the authors of the research, which was published in December 2015 by Elsevier’s Social Science Research Network and subsequently presented at multiple conferences, including, in 2017, at Harvard. “Without tapping phones, with just the traffic volumes, you can predict something unusual is coming up.”
Most other research on links between cellphone usage and violence has focused on whether access to telecommunication gadgets helps specific groups gain an advantage over rivals in war-torn nations like Iraq or Syria. Some research has found that increased cellphone access, especially in Africa, allows greater coordination of violent attacks. “We are able to show that the availability of cellphone coverage significantly and substantially increases the probability of violent conflict,” write researchers Jan Pierskalla of the German Institute of Global and Area Studies and Florian Hollenbach of Duke University in their 2013 paper “Technology and Collective Action: The Effect of Cell Phone Coverage on Political Violence in Africa,” published in the American Political Science Review.
But Ivory Coast findings, published by Linardi and researchers at New York University and the University of Essex, break new ground by effectively pointing to a crystal ball on upcoming violence, especially in countries with high mobile phone penetration — 92 percent in Ivory Coast.
Because it involves large-scale meta data and not actual snippets of conversation, this approach can’t match the efficacy of intelligence surveillance on isolated individuals in predicting major terrorist attacks, says Linardi.
But when it comes to the smaller-scale violence, a preferred weapon to intimidate rival communities in many developing nations, cellphone traffic volumes can help identify where clashes may break out next. “When the violence is meant to intimidate, it is often preceded by subtle hints and threats, meant to add to the sense of fear,” says Linardi. “That’s what we think local communities are picking up.”
The researchers examined 400 million calls between 500,000 registered cellphone users over five months, using records from Orange, the country’s second-largest telecom provider. Orange reached all sections of society across the country, making the data set reliable for analysis. The researchers mapped the data to regions and key dates, including violent clashes recorded by the UNOCI. The study used phone data from December 2011 to April 2012, a period just after the country’s second civil war.
Cellphone traffic, the researchers found, also increased days before major soccer matches and festivals. But in those cases, the evidence indicated that many of the additional calls were being made across service providers to people some distance away. With the increased cellphone traffic before acts of violence, the additional calls were mostly local.
That all adds up, says Linardi. Behavioral scientists have long known that people, when excited or happy, tend to speak about it to a greater range of people. When they’re scared and worried, on the other hand, they only turn to their closest friends and family.
That these clashes occur even though communities sense that something is wrong points to the deep distrust between government agencies and ordinary people, say the researchers. Communities either don’t warn local law enforcement agencies, or the agencies don’t act despite alerts. In many countries, authorities are seen as biased against specific ethnic or racial groups.
Either way, the Ivory Coast research suggests a way around that lack of trust, say the researchers — at least where law enforcement agencies or international peacekeepers have the will. Linardi cautions that there will be “false positives” too when people fear something unusual is happening, and cellphone traffic rises, but things go back to normal on their own.
But in a continent where more than 11 million people have died in civil wars since the end of World War II, extra caution can’t hurt. Ivory Coast, witness to some of that bloodshed, may just have the answer.