“good” might be overstating things — i think our axioms here are extremely pessimistic — but i might agree that 90% good / 10% might be defensible. 1/

but i don’t think that’s empirically defensible. “democratic revulsion” here means overthrowing a system, not merely a party losing an election. 2/

i can’t speak for Matt, but i think under the worldview he’s expressing, it was a feature not a bug that the US pre-Trump was governed by coalitions with highly overlapping technocracies and technocratic ideas (ideas i think Matt still supports). 3/

pre-Trump, democracy’s capacity for mischief was largely neutralized by this overlap. the two-party technocracy would win either way. 4/

it was better if Democrats won — they were the more competent, less corrupt set of technocrats — but the system gave voters an outlet to confer legitimacy for a pretty consistent underlying technocracy. 5/

“democratic revulsion” — democracy actually taking a substantive role in shaping government rather that merely rubberstamping legitimacy — are events like Trump or Brexit, or Hitler wrt Weinar, maybe Erdogan in Turkey (overthrowing enforced secularism) or Modi in India has in my view… 6/

a pretty awful empirical history. In my view, democratic outcomes are best when they provoke gradual shifts over time rather than radical shifts in an instant. 7/

But the “democracy as failsafe if a manipulative technocracy is too ridiculously bad” justification basically leaves democracy’s only substantive role as imposing radical shifts. /fin