This weekend I was playing one of these wonderful Panzer Campaigns war games when I was struck (yet again) by the sheer scope of WWII in the Eastern front. Granted I was playing a tiny scenario of the Minsk 44 war game, not the greatest showdown of men and material compared to other battles in the Soviet Union. Still I was there in command of the 3rd and 48th Armies (1st Belorussian Front), and using air support from the 16th Air Army. At my command: 32590 men 1256 guns 737 vehicles and 330 planes. I say it again, this is not the biggest battle of military history. But I couldn't bring myself to think about what would be in real life to have all these resources at my disposal.
Nowadays, they speak about the so-called three block wars, low intensity conflicts and small wars. These types of conflicts appear to have been particularly frequent in the recent past ten years.
Are big wars a thing of the past?
How big is the average war, anyway?
Professional staticians prefer to speak about the frequency distribution of the values observed rather than about averages only. Everybody has heard of the bell-shaped curve of the archetypal Gaussian normal distribution (see the image at right). Is a curve where the x-axis indicates the value measured and the y-axis indicates how frequent that value was found in a population. The average is the value in the x-axis where the curve has a peak and is the most typical value found in a population.
What the frequency distribution of the sizes of wars looks like?
Back in 1960, a very smart fellow named Richardson took casualties data from wars on record and plotted them in a way similar than the one shown above. Number of casualties in the x-axis and frequency of the wars having x-casualties in the y-axis.
[
Please note the word "similar". There is some mathematical geeky tricks behind the type of plot Richardson made, unfortunately these tricks are beyond the scope of this blog.]
Richardson didn't find a normal distribution for the size of wars (see the graphic at the right, taken from
"Modeling the Size of Wars" by Lars-Erik Cederman). The data showed lots of small wars, fewer large ones and just two of cataclysmic size (WWII and WWI). In other words, he found that
the size of wars was inversely proportional to their frequency. Also, his plot shows no peak, which means that
there is not a typical size for wars.The type of distribution shown above are so-called "power-law distributions" because the frequency decays following a power law. Power law distributions are
typical of systems that self-organize into a critical state
where a tiny fluctuation can trigger either an event of tremendous magnitude or just a smaller, almost non detectable one. The friction between tectonic plates is a good example of these "self-organized at a critical state" systems. The tectonic plates collide with each other, releasing energy in the form of earthquakes. The intensity of earthquakes show exactly a power law distribution.
Scientists like
Cederman, have been arguing that
warfare is one of those self-organized at a critical state systems. The analogy of the tectonic plates and the release of energy in the form of earthquakes is more than appealing. It also leads to the unsettling conclusion that, once the pieces are all set, there may be nothing to stop a world war.
It was 11 A.M. on a fine summer morning in Sarajevo, June 28 1914, when the driver of an automobile carrying two passengers made a wrong turn. The car was not supposed to leave the main street, and yet it did, pulling up into a narrow passageway with no escape. It was an unremarkable mistake, easy enough to make in the crowded, dusty streets. But this mistake, made on this day and by this driver, would disrupt hundreds of millions of lives, and alter the course of world history.
So, to answer the question in the title of this entry: there is no way to tell if humankind will endure another world war. We should expect frequent small wars, that's for sure. But we should never underestimate the power of a chauffeur's mistake ...
Cheers,
