June 5, 2018 Reading Time: 3 minutes
nuclear bomb
The probability of immense wars, as measured by casualties, has gone up significantly, even though we have not witnessed any such events since World War II. (Wikimedia)

Is mankind becoming more “peaceful” and is the world a safer place? Harvard psychologist Steven Pinker certainly seems to think so. Since the end of World War II, war casualties and even homicides are on the decline; the “Great Peace” is upon us.

Pinker attributes this improvement to the progress of mankind. Empathy, self-control, and reason are making strides against revenge and hate: the “angels” are beating the “demons.” As Pinker himself wrote: “The most promising explanation … is that the components of the human mind that inhibit violence—what Abraham Lincoln called ‘the better angels of our nature’—have become increasingly engaged.”

Yet, Pinker is wrong. He mistakes a superficial observation (fewer casualties) for a fundamental change in the probability distribution of wars and armed conflict.

Pinker’s Shortcoming

We must first establish whether current times are indeed more peaceful (less violent). What if we are simply fooled by randomness, as the title of Nassim Taleb’s book goes? What if we naively observe a pattern when no such pattern truly exists? Then, we have no reason to attribute, as Pinker does, a general enlightenment to modern mankind.

Pinker argues that he simply used “descriptive” statistics. The problem with descriptive statistics in a world of fat tails, however, is the following: the mean is very unstable. Say the United States decides to drop several nuclear bombs on North Korea and two million people die. These deaths, caused by one single act of war on one single day, severely impact the mean. Suddenly, our 10-year, 30-year, and 50-year means no longer look as favorable.

According to Nassim Taleb, Pinker confuses the recent declines in wars and casualties with a change in the statistical properties of warfare. Therefore, trying to look for possible explanations of such an alleged decline is foolish.

World War II Changed Warfare Forever

I recalled the discussion between Pinker and Taleb while watching a recent documentary series: World War II In HD Colour. A major turning point during this deadly conflict was the advent of the nuclear bomb. Suddenly, a country could be blasted into submission through its use.

Before nuclear weapons, tanks would simply barge in and try to gain a foothold in enemy territory: casualties in these military strides were inevitable. Yet, traditional warfare scales difficultly. It seems that modern-day politicians have become more cautious: stepping on the wrong toes can force a country to take the desperate measure of nuclear retaliation; (the Korean war is one among many examples). This led to an unprecedented shift in the probability distribution of warfare (war casualties). Battles have become less frequent, but potentially much deadlier.

In Taleb’s terms: warfare volatility seems to have gone down (medium-sized wars with, say, one thousand to ten thousand victims are less probable), but at the expense of fatter tails.

This is why Taleb called the “Long Peace” a statistical illusion. The probability of immense wars, as measured by the number of casualties, has gone up significantly, even though we haven’t witnessed any such events since World War II. Or, as Taleb himself has put it: “Pinker conflates nonscalable Mediocristan (death from encounters with simple weapons) with scalable Extremistan (death from heavy shells and nuclear weapons.) The two have markedly distinct statistical properties.”

Remember that the badly planned and rather ad hoc Hiroshima and Nagasaki bombings led to hundreds of thousands of victims in a single strike: something that no ground army can reasonably achieve within a similar time span.

steven pinker
Pinker implicitly assumes that warfare still follows a normal distribution (blue curve), whereas in fact, war has a fat-tailed probability distribution (red curve).

Fewer Angels, More Hayek

Fortunately, our social institutions–markets, for starters–do not depend on angels. Rather, social institutions emerged to withstand and thrive despite the existence of demons. Otherwise, they would have ceased to exist a long time ago.

We do not need human beings to be “angels” to achieve progress. We need institutions that are robust to “demons.” And in fact, that is precisely what a Hayekian process of discovery and competition is about. Competition makes the world better through evolution and by strengthening institutions, rather than by creating angels.

Olav Dirkmaat


Olav Dirkmaat is a Professor of Economics at Universidad Francisco Marroquín and Director of the university’s public choice center: the Center for Analysis of Public Decisions (CADEP). His research focus is on applied public choice, particularly questions of public finance (taxes, public debt, fiscal limits) as well as governance decentralization.

Get notified of new articles from Olav Dirkmaat and AIER.