I think I’m sufficiently interested in the subject of Stephen Pinker’s The Better Angels of Our Nature: Why Violence Has Declined that I should actually read it instead of simply reading the coverage. But this putative table of wars and atrocities raises in my mind the question of how we would even know if violence has declined.
Suppose instead that the apparent decline in violence is like the apparent decline in economic volatility formerly known as the Great Moderation. The first half of the 20th century, after all, seemed to indicate an increase in violence including the episodes Pinker splices up as the First World War (16th worst), the Russian Civil War (20th worst), Joseph Stalin (16th worst), Mao Zedong (11th worst), and the Second World War (9th worst). Given that those rankings are scaled to population size and include events going as far back as the 3rd century, that’s rather a lot of violence for 50 years. The cycle has switched back into decline since then. But how do we know that’s a real decline in violence rather than a shunting of violence into “tail risk” through nuclear war? A model in which nuclear weapons make great power conflict much less frequent and much more deadly seems like an obvious suggestion. Looking backwards from the aftermath of the multi-faceted nuclear exchange that devastates the entire Northern Hemisphere in 2043 we’re not going to say the world got “less violent” and then suddenly became “more violent” again. We’ll say nuclear weapons made the world seem less violent while masking the fact that the technologies of stabilization were actually creating new vulnerabilities.