On the Brink of WWIII? Estimating the Probability of a new Large Conflict

A plot of the data reported by Peter Brecke on the number of casualties in conflicts from 1400 to now, scaled to the world’s population. The future is always difficult to predict, but these data can be analyzed to show that wars are a statistical phenomenon generated by interactions among human beings. There is no evidence that wars are masterminded from outside: no lizard people, little green men, the Illuminati, a cabal of the WEF, or anything like that. It is we, humans, who are creating this monster, and we have to live with it — if we can.

One of the things you learn when you use models to predict the future is that the future always surprises you. Yet, we still try, and, sometimes, we find that the future follows at least some patterns. So, a few years ago, with my colleagues Martelloni and Di Patti, we analyzed the most extensive database on wars available, the one reported by Brecke in 2011. You see Brecke’s data plotted in the image, above.

We started with the idea that there might have been periodicities in the number or intensities of wars. If that were the case, then predictions would be possible. The Babylonian astronomers had already explored long ago the idea that maybe the motions of planetary bodies drive the world’s events, and that was the origin of our modern horoscopes. There are ways to determine whether there are periodicities in a set of data. We tried, and we found that there is no such thing in Brecke’s data. Too bad, but so is life: no horoscopes are possible about future wars.

The opposite case would be that wars occur randomly, “at the roll of a die.” In that case, no prediction would be possible. The only thing we could say is that no war could involve more than 8 billion victims, but, apart from that, wars could occur anytime, anywhere, without any regularity. We would be like soldiers staying in a trench, expecting to be hit by a shell at any moment.

But the data show a different pattern. Wars are neither periodic nor random. We found that wars follow a “power law.” (see the figure, below). It doesn't mean you can predict when a new war will start, but at least you can estimate its probability. As you would expect, large wars are less probable than small ones, but a power law is also called a “fat tail” distribution because the “tail” events (the very large ones) are more probable than in other kinds of common kinds of distribution. Unfortunately, it means that large wars are not as unlikely as we would want them to be.

You may figure out the mechanism of power laws by considering the paradigmatic example: the “sandpile” model proposed by Bak, Tang, and Wiesenfeld.

The idea is that in a sandpile, one grain of sand starts rolling down. Then, it hits other grains of sand, which also start rolling down. And, soon, a large number of grains start sliding down: it is a landslide. It is an example of the “Seneca Effect” that says that “growth is slow, but ruin is rapid.”

The same mechanism operates for all those systems called “Complex Adaptive Systems,” where the elements of the system are linked to each other by “feedback” effects that tend to amplify or dampen external perturbations (called “forcings”). It happens for physical phenomena, such as earthquakes, landslides, avalanches, and the like, where the elements of the system dissipate accumulated gravitational energy. It also occurs in biological, social, and economic systems. In wars, the system dissipates the previously accumulated economic energy.

So, one quarrel starts a battle. A battle starts a war. A war starts a bigger one, and so on, until you have a full-fledged world war, where everyone is fully convinced to fight on the side of good against evil. An example is how things grew out of control at the start of WWI, as told by Vernon Lee in her “The Ballet of Nations.” As Shakespeare said, “When sorrows come, they come not as single spies, but in battalions.”

Such being the situation, the future does not look good. The probability of a new, large (or even very large) conflict is low, but it is not zero. It is perfectly possible that the same mechanisms of energy dissipation that led to the world wars of the 20th century will lead us to a new world war in the coming years (or even in the coming days or weeks). It is the curse of the fat tail distribution.

Can we do something against this dire destiny? Unfortunately, it is not easy. One problem is that we cannot identify a single cause for wars. There holds the statement valid for all complex systems that, “there are no causes and effects, only forcings and feedbacks.” In other words, we have no evidence for the existence of dark or supernatural entities (*) driving wars. There is no Sauron who stands in his Dark Tower of Barad-dûr and, from there, sends his hordes of orcs to invade the lands of men. And there is no Minas Tirith, either, the citadel where the forces fighting for good bravely resist the wave of evil crashing on them. That doesn’t mean no evil forces are acting to create wars; it means that there is no such a thing as a smoke-filled room, somewhere, where a single group of dark figures are masterminding the affairs of the world. Just a small group of fanatics may be enough to create a major war. A financial lobby of people determined to make money on conflicts can do it even more effectively, but they may also fail.

What we can say is that, if wars are an effect of accumulated energy in the form of economic capital, they will end when it runs out. And that is bound to happen, sooner or later. In the meantime, the best we can do is to resist the onrush of madness. If we are just pebbles in a big landslide, we can still refuse to roll down, and we can try to stop other pebbles nearby from rolling.

We should also try to avoid the mistake that Steven Pinker made when he examined just a few decades of history to arrive at the conclusion that wars were a thing of the past. The data for 600 years may not be enough to conclude that wars will last forever, either. If the past century has been a period of wealth accumulation, what we are facing now is a much leaner period, where there may not be enough surplus to build up the momentum for a new major world war. We may be approaching exactly that.

There have been periods of peace in human history that followed moments of great turmoil. The Antonine Age in Rome, the Heian period in Japan, the Han Dynasty in China, and many others. The only thing we can say is that the future is never exactly like the past, and hope remains alive for all of us.

(*) And even though we found no evidence of supernatural entities driving the events we examined, it may not be a bad idea to pray to God for peace. I do.