He's not trying to deal with the question of potential violence, though. As the parent comment said, his books are mostly descriptive and just lay out data pert

He's not trying to deal with the question of potential violence, though. As the parent comment said, his books are mostly descriptive and just lay out data pertaining to extant trends. It's not a book about possible existential risks or tail risks. These are important topics to address separately, and I don't think that Pinker would dispute that (although he is a little too dismissive of AI risks for my taste, but that's separate from those books). You may wish to read his reply to Taleb's critique of these books, where he addresses the criticism that he ignores tail risks.