Is terrorism the next format for war?
Death statistics in Iraq suggest conflict is turning into a 'war without end'.
As bombings in London attract international attention, one study claims that terrorist patterns of attack might be the natural endpoint for all modern armed conflicts.
Ongoing wars in Iraq and Colombia, which had quite different causes and began as very different kinds of conflict, are developing a characteristic signature of long-term terrorist activity, say economist Mike Spagat of Royal Holloway, University of London, and his co-workers1.
They have found that the death statistics in both of these conflicts are converging on a particular mathematical pattern. This pattern is shared by fatality counts from terrorist attacks in countries that are not major industrialized nations.
Kaldor asserts that US military action in Iraq has been predicated on the view that it is a war of the sort that was fought until the middle of the twentieth century, where two military states battle for control of a territory. But this, she says, is the wrong approach.
"The US failure to understand the reality in Iraq and the tendency to impose its own view of what war should be like is immensely dangerous," she says. Instead of approaching it as a conflict that can be conclusively won by military force, they should see it as an ongoing effort, Kaldor argues.
Neil Johnson, a physicist at the University of Oxford, UK, who collaborated with Spagat on his study, thinks that the apparent convergence between the Iraq and Colombia conflicts could have even more chilling implications. Given that they share the same statistical description, he says, "are they separate conflicts, or are they a part of one big ongoing global war, a mother of all wars?"
Pattern of war
All wars and conflicts seem to generate a common and distinctive pattern of death statistics. Fifty years ago, the British mathematician Lewis Fry Richardson found that graphs of the number of fatalities in a war plotted against the number of wars of that size follow a relationship called a power law, where all the data points fall on a straight line if plotted logarithmically.
This power law encodes the way in which large battles with large numbers of deaths happen very infrequently, and smaller battles happen more often.
Recently, the same kind of power laws were found to hold for terrorist attacks over the past four decades or so. But the precise form of the power law depends on the type of country to which it relates. Terrorist attacks in Western industrialized nations are rare but tend to be large when they happen. Terrorist attacks in the less-industrialized world tend to be smaller, more frequent events2.
Johnson says that the bomb attacks on London's public transport system on 7 July, in which more than 50 people were killed, fit this statistical picture. "They absolutely fall into line," he says.
In Iraq, the battle began as a conventional confrontation between large armies, says Spagat. But the presence of coalition forces "has fragmented the insurgency into a structure in which smaller attack units now predominate", he says. Since 2003, the 'casualties per attack event' for Iraq, measured over 30-day windows, have followed a gradually changing power law. The slope was initially equal to that found by Richardson for traditional warfare, but it is now approaching the value found for non-Western terrorism.
Spagat's team finds that the Colombia conflict, which has been fought between the government and various left- and right-wing guerrilla groups for many decades, is also approaching this value.
The researchers conclude that armies in Iraq and Colombia should be using different tactics. "If you believe that you need to fight like with like, a conventional army is the complete opposite of what you need. You have to do away with centralization," says Johnson. In many ways, he says, it is like fighting an illness that continually evolves, adapts and changes.
The broader worry is that terrorist-style conflicts seem to be sustainable indefinitely. "These wars," Kaldor agrees, "are so much harder to end than to begin."
- Johnson N., et al. Preprint, (2005).
- Clauset A., Young M., et al. Preprint, (2005).