Pinker’s The Better Angels of Our Nature feels similar to Piketty’s Capital, in that they both deal in large, noticeable changes and trends in broad statistical aggregates.

It’s the kind of stuff that’s very useful for putting bounds on your thinking – you don’t have to agree with the authors about every little detail, but whatever theories you come up do have to explain the big picture.  Things like “this number is 1/5 the size of this one” or “the worst cases of X are much better than the best cases of Y” are useful guidelines in the otherwise confusing territory of social science.

(This is all supposing that Pinker’s statistics are not so bad that they can make up gaps like this out of thin air, which is not a perfectly safe supposition.  Piketty was subject to a lot of scrutiny and, from what I’ve read, has come through in the clear; my impression is that Better Angels has taken more palpable hits.)