3 ms·
Hey.. Eric Schmidt , and reportedly DJT exude eudaemonia. [cites available] I dont know about autarky and honour in those cases; might be deformed to work. Thos
by gradschoolfail 2y ago
Hey.. Eric Schmidt , and reportedly DJT exude eudaemonia. [cites available] I dont know about autarky and honour in those cases; might be deformed to work. Those are examples you might be familiar with, plenty anecdata of sociopathy back-inferred from distant effects being incommensurate with local affect.
Fwiw i still find it hard to dichotomize banter of Sir H & Lord H, instrumental locally, but not overly excessive, and often returns to the status quo antebellum. Formulaic?
server error on those links coukd be bw on my side..
- 082349872349872 2y agotry these links, https://archive.org/details/in.ernet.dli.2015.183333/page/216/mode/2up https://archive.org/details/in.ernet.dli.2015.183333/page/21... to https://archive.org/details/in.ernet.dli.2015.183333/page/224/mode/2up https://archive.org/details/in.ernet.dli.2015.183333/page/22... ? As far as EES and DJT go, neither seem to have the proper mean concern with regard to riches, for both seem to have been excessive in getting and have been deficient in giving (NE II,vii.4). That said, based on personal communication with the former, he'd (in my limited sample) score well on Aristotle's checklist (NE IV,iii.14-34) for proper mean concern with regard to honour, while judging from the social media output* of the latter, he'd score poorly. * this is probably an upper bound: with the exception of (the fictional) Dolphus Raymond, I'm having trouble thinking of anyone who deliberately projects a dishonourable public persona to cover an honourable private one?
- gradschoolfail 2y agohttps://timkellogg.me/blog/2024/06/26/politics-analysis https://timkellogg.me/blog/2024/06/26/politics-analysis Any luck with the Schuetzenberger thm? Will get back to aristotle when more bw thanks!
- 082349872349872 2y ago> to speak in shrill tones ... denotes an excitable and nervous temperament... — Ἀ thanks for asking — I've a decent handle on the original thm, but the categorical generalisation still has me stumped (when I first complained about categorical boilerplate in CS publications to a maths friend, he pointed out that when one's only familiar with a single concrete example, of course its generalisation is never going to seem helpful).