You have good and reputable sources. Pinker's book is largely a collection of data and review of previous thought on civilization as a process, in addition to analysis of crime data from recent time. It is quite convincing and not really a departure from existing thought, but more an approachable overview of it.

He does a little suggestion of his own ideas at the end, but it is properly humble and is not really proclamation. If you are interested enough to ask, it is worth a read. My favorite nonfiction book.

I am aware of one peer-reviewed critique of this argument, by Taleb, acting more as a statistician than as a historian.

http://www.fooledbyrandomness.com/violence.pdf

He argues that violence in a period is dominated by deaths from armed conflict, so you must focus on the frequency of those events to understand any long-term trend of violence that might exist. He catalogues deaths from armed conflict going back two thousand years and calculates the average "waiting time" between two WW2-size conflicts, The conclusion is that the average waiting time is more than twice the period since WW2. Hence, our current period of peace is not statistically significant.