Counting casualties is fraught with uncertainty; scholars' estimates vary according to how they define war and what sources they accept as reliable, among other factors. Nevertheless, a clear trend emerges from recent studies. Last year, 25,600 combatants and civilians were killed as a direct result of armed conflicts, according to the 2009 Yearbook of SIPRI, the Stockholm International Peace Research Institute, to be released Aug. 17. Two thirds of these deaths took place in just three trouble spots: Sri Lanka (8,400), Afghanistan (4,600), and Iraq (4,000). In contrast, almost 500,000 people are killed each year in violent crimes and well over 1 million die in automobile accidents.
SIPRI's figure excludes deaths from "one-sided conflict," in which combatants deliberately kill unarmed civilians, and "indirect" deaths from war-related disease and famine. If these casualties are included, annual war-related deaths from 2004 to 2007 rise tenfold to 250,000 per year, according to "The Global Burden of Armed Violence," a 2008 report published by an international organization set up in the aftermath of the Geneva Declaration. Even this much higher number, the report states, is "remarkably low in comparison to historical figures."
For example, Milton Leitenberg of the University of Maryland's School for International and Security Studies has estimated that war and state-sponsored genocide in the first half of the 20th century killed as many as 190 million people, both directly and indirectly. That comes to an average of 3.8 million deaths per year. His analysis found that wars killed fewer than one-quarter of that total in the second half of the 20th century—40 million altogether, or 800,000 per year.
Even these staggering figures are low in comparison with prehistoric ones, if considered as a percentage of population. All the horrific wars and genocides of the 20th century accounted for less than 3 percent of all deaths worldwide, according to one estimate. That is much less than the probable rate of violent death among our early ancestors.
This seems to me both contrarian and convincing. Generally, we tend to ascribe violence in society to immutable human nature, and suppose we have some sort of natural tendency toward war and violence. This article explicitly argues the opposite.
Our prehistory seems to have grown more bellicose as time went on, however. According to anthropologist Brian Ferguson, there is little or no clear-cut evidence of lethal group aggression among any societies prior to 12,000 years ago. War emerged and rapidly spread (PDF) over the next few thousand years among hunter-gatherers and other groups, particularly in regions where people abandoned a nomadic lifestyle for a more sedentary one and populations grew. War arose, according to this perspective, because of changing environmental and cultural conditions rather than because of "human nature".Steven Pinker also argues the same case, but expands it to all forms of violence, looking at homicide rates, harsh punishment and other forms of violence as well as war. He points out that cruelty was once considered high entertainment. The burning of cats to death was once considered very amusing for example. Another example is be bear-baiting, a sport involving a fight to the death between bears and trained dogs. This sport was a major form of entertainment in England until the 19th Century.
Since Pinker is known for, among other things, his blistering attack on those who deny the importance of genes in shaping human behavior, it is interesting that his argument is explicitly not that we have a genetic disposition toward violence, but rather that violence is rational. The logic of a the preemptive strike makes violence tempting option to both sides. If two armed groups exist because each side know that the other side could raid first, and therefore it seems wise to launch a first-strike. This dynamic can also be seen between nations. In 1967, the Egyptian army mobilized on bad information obtained from the Soviet Union about a looming Israeli attack on Syria. Israeli sources picked up the mobilization and the leader-ship decided, knowing the country to be vulnerable to a first-strike, and itself struck first. The result was the 6 Day War. Because they attacked, Israel was able to destroy the Egyptian air-force while it was still on the ground (it should be pointed out that, despite the fact that the war was a stunning success from an Israeli perspective, most of current the problems of the area grew out of the conflict).
John Quiggin of Broken Timber thinks thinks Pinker is confused and inconsistent as to the matter of violence.
I’ve seen this kind of confusion before. Rational egoist models like homo economicus, ’selfish gene’ models like evolutionary psychology, and ‘realist’ models of international relations (in which nation-states are viewed as unitary actors) use similar styles of argument and therefore appeal to the same sort of person, but they radically inconsistent with each other, because they each posit a single level at which everything can be explained, different in each case.
I don't see the inconsistency in the position. The logic Pinker is using is similar to logic used by those who espouse homo economicus and "realist" theories of international relations. "Homo economicus" and "realist" international theory are both suspect for different reasons, the former because it assumes the human being to be rational, the latter because it makes similar assumptions about states, when in reality foreign policy is generally the result of special interest infighting and politicians trying to maintain their power, not "great powers" attempting to maximize their power and influence. However, the argument Pinker makes about preemptive attacks doesn't rely on either of these (or even, in this case, on selfish genes). The level that this takes place in dispute, this effect clearly happens at the level of human organization, whether they are tribes, clans, mafia families or nation-states, as the 6 Day War illustrates.