On Facebook, my frined Timothy Scriven pointed to an opinion piece by classics professor Ian Morris headlined In the long run, wars make us safer and richer It’s pushing a book with the clickbaity title War! What is it Good For? Conflict and the Progress of Civilization from Primates to Robots.”. Timothy correctly guessed that I wouldn’t like it.

Based on the headline, I was expecting a claim along the lines “wars stimulate technological progress” which I refuted (to my own satisfaction at any rate) in Economics in Two Lessons”. But the argument is much stranger than this. The claim is that war, despite its brutality created big states, like the Roman empire, which then delivered peace and prosperity.

For the classical world at 100 CE or so, the era on which Morris is an expert, that argument seemed pretty convincing. As the famous Life of Brian sketch suggests, Roman rule delivered a lot of benefits to its conquered provinces.

The next 1900 years or so present a bit of a problem, though. There have been countless wars in that time, and no trend towards bigger states. On the contrary two or three dozen states (depending on how you count them) now occupy the territory of the former Roman Empire.

You could cut the number down a bit by treating the European Union as a new empire, but then you have an even bigger problem. The EU was not formed through war, but through a determination to avoid it. Whatever you think about the EU in other respects, this goal has been achieved.

Morris avoids the problem by a “no true Scotsman” argument. He admits in passing that the 1000 years of war following the high point of Rome had the effect of breaking down larger, safer societies into smaller, more dangerous ones, but returns with relief to the era of true wars, in which big states always win. That story works, roughly, until 1914, when the empires he admires destroyed themselves, killing millions in the process.

After that, the argument descends into Pinker-style nonsense. While repeating the usual stats about the decline in violent deaths, Morris mentions in passing that a nuclear war could cause billions of deaths. He doesn’t consider the obvious anthropic fallacy problem – if such a war had happened, there would not be any op-eds in the Washington Post discussing the implications for life expectancy.

I haven’t read the book, and don’t intend to. If someone can’t present a 700 word summary of their argument without looking silly, they shouldn’t write opinion pieces. But, for what its worth, FB friends who have read it agree that it’s not very good.