Can there be a science of history? Isaac Asimov’s Foundation series proposed an elite cadre of “psychohistorians” who could predict the future; in Peter Turchin’s End Times (Allen Lane), the proposed science of “cliodynamics” (after Clio, the Greek muse of history) is less confident of precise prediction but nonetheless makes much of the author’s own forecast, in 2010, that by the 2020s there would be a surge in political instability and violence in the US. He got that right. Surveying the rise and fall of empires and parliaments through global history, Turchin argues that civic upheavals become due when “popular immiseration” (eg, the stagnancy of real wages) is combined with “elite overproduction”: the number of people qualified (via PhD or otherwise) to high social rank exceeds the number of spaces available, leading to dangerous disgruntlement in both camps. Pumping wealth upwards, from low to high in the social hierarchy, eventually leads to “state collapse and social breakdown”.