There's an epistemological dilemma/equivocation inherent in the phrase "elite overproduction". Obviously, the number of true elites is limited by the actual amo

There's an epistemological dilemma/equivocation inherent in the phrase "elite overproduction". Obviously, the number of true elites is limited by the actual amount of elite input taken by society's institutions (in the same way the number of rocks on top of a pile is limited by the surface area). When you measure an "elite overproduction", what must be happening is that one noisy signal (metric) of eliteness (e.g. Harvard degrees) is showing a much higher false positive rate than true positive rate, as determined by comparison with a more sensitive (but often useless) metric, in this case "actually being elite" (working for USG or The Atlantic). When you have this problem in any other field, you switch to a different metric! In other words, when Turchin says "elite overproduction", he is implicitly trusting the very metric he is criticizing. Of course trusting academia as a kingmaker is not a surprising stance for... an academic, but we are not and plow on. If you reframe the question, you instead see "inability to identify elites", that is, a decrease in the signal-to-noise ratio of the metrics used to identify people who are trustworthy. In this framing, it is less surprising that this "elite overproduction" viz. "qualification debasement" is happening at the same time as a general decline in the public trust of institutions! After all, isn't it rational to stop trusting a noisy metric? We could of course spend many paragraphs picking over the nitty-gritty mechanisms by which these increasingly noisy signals eventually produce mistrust, but what we want is the crow's perspective [1]. There is also the question of whether "trustworthy" should have anything to do with "elite", but it does usually have some connection to social achievements. What bothers me is that I personally can't really recommend my own method of trust-but-verify to most people: I have a pretty good command of mathematics, and I determine who to trust, in many cases, by checking their math. This is an odd privilege: for example, a brief glance at Trump supporters sharing graphical claims of voter fraud is enough for me to immediately detect lying-with-statistics, but for someone without a graduate degree in a mathematical science it may not be so easy. 1: http://chaosbook.org/FieldTheory/01-Intro.pdf#%5B%7B%22num%22%3A33%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C0%2C0%2C0%5D http://chaosbook.org/FieldTheory/01-Intro.pdf#%5B%7B%22num%2...