Stephen Pinker made the same basic argument in his book The Better Angels Of Our Nature a few years ago. The problem with the general form of the argument is that benevolent past trends are not a reliable indicator of the future when major social trends are all in the process of altering radically. In 300 CE a writer could with perfect validity point out that although the Roman Empire had lots of problems, in general most people in the Empire were better off than their distant forebears of 800 years previously. But by 500 CE the Empire had split, Rome was a shadow of its former self, and Europe was about to enter into 1,000 years of ignorance and superstition. Likewise in 1913 all the indicators were positive: global trade and prosperity were at heights never before seen in human history, basic germ theory was slowly leading doctors to wash their hands and thus stop killing their patients by acting as a perfect vector for infection, and trade between European nations made war unthinkable.

Today we're at the end of our largely accidental experiment with representative democracy and its complete non-fitness for purpose is ushering in (as Plato correctly explained it must, more than 2,500 years ago) a new age of populist-generated tyranny. Past trends are thus very misleading when it comes to discerning the shape of the medium-term future.