The Pinker Problem

Michael Harris
3 min readAug 16, 2020


Image by Steven Pinker

Steven Pinker insisted in Twitter yesterday that

The two world wars were spikes, not harbingers of a trend. Since end of WWII, death rates have roller-coastered down.

Here is the tweet:

Nassim Nicholas Taleb offered a swift reply:

N. Taleb writes in Statistical Consequences of Fat Tails:

For fat-tailed variables, the mean is almost entirely determined by extremes. If you are uncertain about the tails, then you are uncertain about the mean.

Many traders and investors know the above but apparently a large population of social scientists who are otherwise good and enjoyable people fail to understand it.

One fundamental reason for the lack of understanding is that fat tails are not covered in undergraduate curriculum and even in the first part of graduate courses.

I will try to offer a slightly different - but not fundamentally -account of why the assertion that the world is “more peaceful” is false. It goes like this:

  1. Before nuclear arms technology, there was a stochastic process that generated wars between large powers and deaths, as a consequence. There were many reasons for these wars but main one was forceful redistribution and control of resources.
  2. After the advent of nuclear weapons and deterrence, this process hit an absorbing barrier. By absorbing barrier we mean that this stochastic process of conventional wars between large powers went bust and cannot continue any longer because of fear of nuclear retaliation. Note that USA, Russia, China, UK, France, India, Pakistan all have nuclear weapons and several other countries are rumored to have developed them. Therefore, wars remained regional with small death counts.
  3. The death generation process is now fat-tailed under a possible new absorbing barrier at the next superpower conflict, meaning that this may equate to the end of humanity, at least as we know it. This is again from N. Taleb, the Thanksgiving Turkey problem.

Steven Pinker is enjoying a peaceful era only because the dynamics of the stochastic process have changed and the next war if it occurs will be the last one.

The mistake is that Steven Pinker thinks this is the same stochastic process that generated wars before that it is still active nowadays: it is not in my opinion. A new fat-tailed process has taken over due to nuclear deterrence but that is even worse than the previous one. In the past wars killed a smaller fraction of populations that were involved in their immediate vicinity. A nuclear war can nearly kill everyone on the planet by making it uninhabitable.

Peaceful may imply much higher risk. In my opinion this is what Steven Pinker misses. When you get peace because of much higher risks, it is paradoxical to say this is a more peaceful world. One way to put it is as follows:

This is a more peaceful but extremely dangerous world because of the price for peace.

It is a paradox, but that is what it is.

About the author



Michael Harris

Ex-fixed income and ex-hedge fund quant, blogger, book author, and developer of DLPAL machine learning software. No investment advice.