So, what makes "ruler by popular right as acknowledged by election" greater than "ruler by divine right as acknowledged by coronation ceremony"?
Coronation ceremonies are too easy to fake. Or, rather, too easy to duplicate. Any pretender can put one together, and probably it’ll have most of the compelling power of the true article.
…that’s the glib answer. The slightly-less-glib answer is that, while in an abstract sense (I posit) it doesn’t matter what the legitimacy criterion is so long as it’s unambiguous, in practice it does actually have to get some cultural traction in order to serve its purpose. Things like “being crowned by the Pope” and “being the eldest son of the king” have reliably failed in this regard; they don’t protect you from coups and intrigue, because no one really takes them seriously in the right way. It’s not unacceptable to depose a king-chosen-by-primogeniture via force of arms, no matter how hard various kings have tried to make it so, because in the end it’s hard to spin your particular lineage as mattering in any very profound way.
(Interestingly, the one sort-of-counterexample I can think of is the imperial line of Japan, which – despite some turbulent periods and nasty intrigue – is in fact treated as an unbroken divine bloodline that really does matter, or at least has been treated that way for a very long time. I suspect it is not an accident that for much of that time “Emperor” has meant “sacred figurehead” rather than “wielder of supreme political power.”)
“Legitimate authority arises from the will of the people as expressed via elections” is stupid, in the sense that basically no one actually gets to see his will expressed that way, but it does make for a sticky meme and people believe in it.
I assert that people are also willing to believe in credentialing systems.
While elected governments have a long history of stability in a handful of countries (including the US), aren’t there also plenty of examples in other places of (eg) military coups overthrowing elected governments?
I think that’s less a discredit to elections than a credit to military rule, in the sense that both are very few steps removed (relative, to, say, the Emperor School proposal) from “asserting a social order” in a less decorated sense. You want to have an organization with overwhelming means of violence in some territory, you want to have enough people agree (coerced or not, whatever) to some particular way of doing things. Even in a world where nobody believes ideologically in popular or military sovereignty, a ruler can get thrown out if they don’t have popular or military support, whereas an emperor in a world where nobody ideological believes in one is Emperor Norton.
If the Emperor school conveyed the kind of amazing organizational skill such that any political side with a graduate was overwhelmingly stronger than the side without it, then you’d see similar dynamics for that.
I don’t really buy this. Really overwhelming undeniable will-of-the-nation mandate elections, like the ANC in immediately-post-apartheid South Africa, probably do manage to express “anyone who stands against this order is going to lose hard in any contest of force”…but those are the exception and not the rule. It’s a lot more normal, in a stable democracy, for an election result to reflect the “will” of a sub-55-45 majority of the electorate – or, in many places (such as America), a sub-55-45 majority of a smallish slice of the electorate – with every possible relevant factor in a civil war (urban vs. rural, old vs. young, rich vs. poor, etc.) skewed. There is no real information there in terms of the regime’s stability in the face of an actual legitimacy crisis. And I think everyone understands this very well. Who really thinks that the outcome of Clinton vs. Trump tells us anything about the lay of the land when it comes to violence? That, if the outcome had been different because Comey decided to keep quiet or whatever, that it would have told a different story in that regard?