Modern warfare is a messy, erratic, complicated business, defined by the unpredictability of the next insurgent or terrorist attack. But apparently, all that chaotic violence is nothing a physicist can't figure out. All it takes is a bit of math – and a little inspiration from Alice in Wonderland.
In a paper published last week in Science, researchers present an equation to describe how fatal attacks escalate – whether it's a suicide bombing in Lebanon or an insurgent attack in Kabul. "The way in which those attacks occur as time goes on has a particular mathematical form," says physicist Neil Johnson of the University of Miami, Florida, who led the study. That form, associated with everything from puzzle solving to ship building, is a progress or learning curve: the more you do something, the better (and quicker) you get. In this case, it's the insurgents who are doing the learning.
But they aren't just learning in isolation – they have to deal with the other side, the coalition forces. To interpret how this underlying arms race evolves over time, Johnson pulled an analogy from evolutionary biology. The Red Queen hypothesis (named for the character in Lewis Carroll's story who says it takes all the running you can do to keep in the same place) refers to competing species constantly adapting just to survive. Johnson simply applied the same principle to human conflict.
"We saw that thinking about the race of the Red Queen would give the types of numbers we were actually seeing for the different provinces," Johnson tells Danger Room.
This is not the first time physicists have tried to pin down the eccentricities of war. A few years ago physicist Sean Gourley came up with a simple equation to predict the size of attacks (it didn't work so well). The U.S. government has spent millions trying to create computer models to predict the conflict-ridden future. But even the best of them, like the Integrated Crisis Early Warning System, has been far from fool-proof.
Maybe that's why Johnson's team chose not to look at human conflict in its entirety. Instead, they focused on just one property: timing.
To start, they gathered data on military fatalities from the recent wars in Afghanistan and Iraq, as well as terrorist attacks and suicide bombings going back to 1968. The data (from several public websites like icasualties.org and The Memorial Institute Prevention of Terrorism) is publicly available, but it's not exactly foolproof. For one, it doesn't include fatalities on the side of insurgents, never mind attacks that didn't kill anyone. But when analyzing war, maybe it's better to look where the light is.
The researchers found a surprisingly consistent relationship between the first two attacks and the subsequent escalation rate – even in provinces separated by hundreds of miles. "Kandahar is in the south, Kunar is in the east, Farah is in the west," says Johnson, "yet if someone was just sitting there telling you the pattern of attacks in time, the data from those three regions would be almost indistinguishable."
From that pattern they built an equation. Say two fatal attacks occur in the same region, 100 days apart. Plugging 100 into the equation suggests that the next attack will happen in 66 days. Following Johnson's formula, you could predict not only the date of the third attack, but the fourth, fifth, or fiftieth attack. All you have to do is follow the curve.
Johnson admits that his equation is – and could only ever be – an estimate. But he thinks that even an estimate could serve a purpose. An attack that was "off schedule," so to speak, might serve as a cue to look more closely at what happened during that time. Was it something obvious like a massive snowfall? Or does it indicate a change of tactic on one side or the other that was particularly successful? Against a background of prediction, outliers stand out.
And what does Johnson say to the skeptics that doubt an equation could ever predict human behavior, let alone the complicated patterns of war?
"They're right," he says, "up to a point."
That's the point he's still working on.
*Photo: Marines, modified by Lena Groeger
*
See Also: - Pentagon’s Prediction Software Didn’t Spot Egypt Unrest
