This gets discussed occasionally on the Manifold Discord and I wanted to share some skeptical points that one of the top forecasters (Semiotic Rivalry) made there:
“for me to go >5% on this [authoritarian takeover/coup] i’d have to see them openly disobeying orders from the supreme court or like, the very least should be killing the filibuster”
“I feel like the lightest first step of the fascist takeover would be to have the VP overrule the senate parliamentarian on what can be permitted to go into a reconciliation bill, which is totally legal and tons of Dems wanted Biden to do, and they failed to even do this”
Supreme Court is still constraining him, e.g. Trump wasn’t allowed to fire Federal Reserve commissioner Lisa Cook
Revealed preferences suggest people don’t actually believe dictatorship/catastrophe to be very likely: they aren’t moving abroad or stocking up on guns. (To which people replied that it’s not easy to find a good place to live abroad due to economics, language)
This was largely in response to me saying that I find it hard to think through Trump/MAGA military (self-)coup possibilities. Because although military self-coups appear to be rare in consolidated or backsliding democracies, they’re not entirely unheard of, and it sure seems like Hegseth, Trump, etc. are working towards this. They are systematically dismantling military guardrails:
They keep pushing the envelope on deploying military domestically (in conflict with the Posse Comitatus Act), which blurs lines among the population and the military for this
They’ve fired a lot of military leadership as well as all Judges Advocate General (JAGs) which generally serve as a constraint on executive overreach.
The pardoning of war criminals, the broader pardoning of J6′ers and Trump allies, and the commands to engage in war crimes like shooting down a boat of non-combatants (allegedly drug-traffickers)
The general pattern of purging appears to be that Trump/the administration gives an illegal/norm-breaking order which functions as a loyalty test: it forces everyone involved to either comply, step down, or refuse to obey (which tends to get you fired—something that the Supreme Court hasn’t been adequately protecting besides the Fed).
The coup form I expect, if it happens, would not be a direct command to military generals, but to order his most loyal militarized groups (e.g. red states National Guard, ICE) to take control of the democratic/election process. Opposing military would then have to coordinate on action, which would be very difficult. The general population could resist en masse (South Korea 2024-style), but so far protests have been small, and in the US there’s a vocal and dangerous base supporting Trump. That said, base rates suggest a coup is still very unlikely, and coups are difficult. I don’t know what probability I would give it, I’m mainly trying to understand the mechanisms here.
Other thoughts:
Trump attempted to overthrow an election before
Orbán is often mentioned as comparison (rightly so), but he was able to amend the Constitution in the first year due to Hungarian law, which is a major difference
An economic crisis would be a major cause of discontent and the AI boom is really unfortunate
I haven’t even talked about AI but it’s a wild card, probably would favor Trump. Executives are largely very appeasing (OpenAI, xAI), appeasing (Google, Meta), or softly defiant (Anthropic)
People think Trump is too old and a unique figure, but I’m not confident that a successor wouldn’t be as bad. At some point, they either put a successor on the ballot or Trump himself. A successor could pull away power from Trump and then lose. There’s generally a lot of possibility around this to dis-unite any coup-interested faction. However, I find the sentiment that “Trump is uniquely bad and his successor will not have the same power and therefore it’s not a concerning scenario” overconfident, and there’s plenty of systemic reasons to expect a successor to be pretty bad
Your comment above is the most informative thing I’ve read so far on the likelihood of the end of democracy in America. I especially appreciate the mix of key evidence pointing in both directions.
Couple of thoughts, kinda long and rambly
This gets discussed occasionally on the Manifold Discord and I wanted to share some skeptical points that one of the top forecasters (Semiotic Rivalry) made there:
“for me to go >5% on this [authoritarian takeover/coup] i’d have to see them openly disobeying orders from the supreme court or like, the very least should be killing the filibuster”
“I feel like the lightest first step of the fascist takeover would be to have the VP overrule the senate parliamentarian on what can be permitted to go into a reconciliation bill, which is totally legal and tons of Dems wanted Biden to do, and they failed to even do this”
Supreme Court is still constraining him, e.g. Trump wasn’t allowed to fire Federal Reserve commissioner Lisa Cook
Revealed preferences suggest people don’t actually believe dictatorship/catastrophe to be very likely: they aren’t moving abroad or stocking up on guns. (To which people replied that it’s not easy to find a good place to live abroad due to economics, language)
This was largely in response to me saying that I find it hard to think through Trump/MAGA military (self-)coup possibilities. Because although military self-coups appear to be rare in consolidated or backsliding democracies, they’re not entirely unheard of, and it sure seems like Hegseth, Trump, etc. are working towards this. They are systematically dismantling military guardrails:
They keep pushing the envelope on deploying military domestically (in conflict with the Posse Comitatus Act), which blurs lines among the population and the military for this
They’ve fired a lot of military leadership as well as all Judges Advocate General (JAGs) which generally serve as a constraint on executive overreach.
The pardoning of war criminals, the broader pardoning of J6′ers and Trump allies, and the commands to engage in war crimes like shooting down a boat of non-combatants (allegedly drug-traffickers)
The general pattern of purging appears to be that Trump/the administration gives an illegal/norm-breaking order which functions as a loyalty test: it forces everyone involved to either comply, step down, or refuse to obey (which tends to get you fired—something that the Supreme Court hasn’t been adequately protecting besides the Fed).
The coup form I expect, if it happens, would not be a direct command to military generals, but to order his most loyal militarized groups (e.g. red states National Guard, ICE) to take control of the democratic/election process. Opposing military would then have to coordinate on action, which would be very difficult. The general population could resist en masse (South Korea 2024-style), but so far protests have been small, and in the US there’s a vocal and dangerous base supporting Trump. That said, base rates suggest a coup is still very unlikely, and coups are difficult. I don’t know what probability I would give it, I’m mainly trying to understand the mechanisms here.
Other thoughts:
Trump attempted to overthrow an election before
Orbán is often mentioned as comparison (rightly so), but he was able to amend the Constitution in the first year due to Hungarian law, which is a major difference
An economic crisis would be a major cause of discontent and the AI boom is really unfortunate
Protests so far have been small, No Kings Protests of “5 million” was [greatly exaggerated](https://bsky.app/profile/siebepersists.bsky.social/post/3lruu445wgk27) (posted on BlueSky but I’m not at all active there otherwise). I think bigger protests will be necessary (but not sufficient)
I haven’t even talked about AI but it’s a wild card, probably would favor Trump. Executives are largely very appeasing (OpenAI, xAI), appeasing (Google, Meta), or softly defiant (Anthropic)
People think Trump is too old and a unique figure, but I’m not confident that a successor wouldn’t be as bad. At some point, they either put a successor on the ballot or Trump himself. A successor could pull away power from Trump and then lose. There’s generally a lot of possibility around this to dis-unite any coup-interested faction. However, I find the sentiment that “Trump is uniquely bad and his successor will not have the same power and therefore it’s not a concerning scenario” overconfident, and there’s plenty of systemic reasons to expect a successor to be pretty bad
Your comment above is the most informative thing I’ve read so far on the likelihood of the end of democracy in America. I especially appreciate the mix of key evidence pointing in both directions.