Home

One doomer sentiment I’ve seen floating around my Bluesky feeds in the last couple days is that the current moves by the Trump regime are part of some 4D chess to trick people into protesting against him so he can unleash a violent crackdown that places him as god-emperor for life. And while that is certainly a future possibility there’s a difference between that being some savvy plan pulled off ahead of time and what is far more likely is that the administration does not know what it’s doing.

Of course there’s an obvious rejoinder to that statement which is that he won the presidency. But you have to remember that intelligence is frequently domain-dependent, you can have a hyperoptimized process that works really well in a specific context and is utterly hopeless outside it. This is my explanation for Trump’s victory, the fact that American institutions were set up in such a way as to allow him to walk in a second time despite running what was probably the worst campaign in history. It’s pretty easy to imagine an alternate timeline in which the Democratic establishment made slightly different decisions and a candidate ended up moving enough votes in the swing states to beat Trump.

But my point is not just that he's dumb. It’s also that him being dumb increases our odds of getting to 2029 without Trump, or some other Republican, in power.

And I’m bringing data to back this up.

Democracy By Mistake by Daniel Treisman is a paper which looks at the various ways in which autocracies have transformed into democracies since 1800. They find that in roughly 66% of cases of a transformation of a state from autocracy to democracy the cause is leaders making mistakes.

So in what ways have autocrats made mistakes?

Well, the most relevant for people depressed on social media is hubris – ignoring warnings of popular discontent or bypassing potential policies that could tamp down dissent until it reaches a point where the regime is overthrown which explains 13-17% of cases. For example Louis Philippe, the last French king (although not “monarch”, it’s complicated), downplayed the threat of popular backlash in response to economic depression that began in 1846 only for the Revolutions of 1848 to break out and end his regime.

There’s plenty of examples one could point to of Trump ignoring sources of popular backlash. There’s the obvious shit like tariffs and deportations raising food prices, increasing cost of living across the board despite Trump largely running on bringing those prices down. But there’s also more targeted forms of hubris against people with more capacity to resist him and who are already organized to some degree: see him firing all transgender soldiers an estimated 9,000 to 12,000 people, threatening the livelihood of scientists who rely on federal funding (a group that was overwhelmingly antifascist in the 30s and 40s) and potentially almost half of the FBI.

Initiating failed military conflicts is responsible for 6-9% of transitions. Examples here include something like Argentine General Galtiteri who invaded the Falkland Islands to try and bolster internal support and that England would not try to defend the territory. The invasion only lasted three months before he retreated and resigned shortly after. His successor, General Bignone abdicated office shortly after in the face of mass demonstrations, strikes and financial crisis by the people of the country.

Certainly the Trump regime is being handed a far more powerful force in the form of the United States military. But his calls for annexing Greenland or an intervention in Mexico will have him either declare war against a former ally that is part of a defensive pact with nuclear powers or wage a counterinsurgency against some extremely sophisticated non-state actors that are skilled at entering the United States undetected and have a lot of experience with assassination. Neither will look anything like the first Gulf War, a popular, bloodless (for the United States), short war that saw George H.W. Bush’s approval ratings hit something like 91%. Nor do I think Americans will be overjoyed for months by Trump annexing Panama, a state with a population in the millions that doesn’t have a standing army.

Electoral loss is another factor that occurs in 24-29% of cases. Authoritarian regimes hold elections that are heavily favored to their end so as to reinforce public legitimacy or coordinate elites.

Yet this can backfire. For example, the constitutional referendum held by Augusto Pinochet in 1988 which saw him lose 55% to 45%. Pinochet was shocked by the result and tried to organize an auto-coup in response, but his fellow junta members admitted to the press that he had lost and did not grant him the emergency powers necessary for him to overturn the result. He did not prepare in advance.

Moreover elections in the United States are largely a process run by states or even more locally and so are not controlled by the federal government. And so while it’s easy to imagine further attempts at gerrymandering done in red states in the next few years, they can’t just turn the United States into Syria under Assad overnight. A recent example is something like the Democrat Mike Zimmer winning a special election in Iowa in a seat that went for Trump by 21 points in the federal election in 2024 and that was won by a Republican in 2022 by 10 points.

There’s also counterproductive repression which is a key factor in 12-15% of cases. For example the 1990 uprising in Bangladesh which came out of a student protest being violently repressed, which in turn motivated a wave of rallies, strikes, blockades and street fights that overthrew the dictatorship within just two and a half months.

It is pretty easy to see this one being a factor given that we’re almost five years out from the George Floyd protests. Particularly since how Trump's policies are playing out, I expect any movement sparked by repression to be far more ambitious and far more wide-ranging. The people taking to the street won’t be motivated by a vague sense of racial justice and frustration/anxiety over COVID lockdowns but instead a visceral sense of hurt from the Trump regime hitting them or people they care about in some way.

Succession is a factor in 7-10% of cases, with the most well-known being something like the election of Gorbachev who enacted reforms which ended up leading to the collapse of the Soviet Union. And while J.D. Vance is clearly a believer in some sort of authoritarian project and is also far more intelligent than Trump, he does not have nearly the same charisma or standing. So if Trump dies, it’s easy to imagine a MAGA civil war breaking out that severely constrains what Vance can do and he loses handily in 2028 because nobody serious wants anything to deal with him, no matter how much Curtis Yarvin he’s read.

It should be clear that in the next four years Trump could make egregious errors in all of these categories. Easy to see him ignoring growing discontent, starting a disastrous war or counterinsurgency, appeal to fair elections for legitimacy that expose his unpopularity, enact repression that backfires on him and, finally, die.

That said, you shouldn’t take comfort in this data. Treisman is only looking at what ends autocracies, plenty of regimes hobble along for years making some of these mistakes. Moreover there’s a specific history to each example which makes naively extrapolating from it to the current United States an error.

Nor do I want to downplay the suffering that will take place in the next four years even in the timelines where we kick these bastards out.

All I want to claim is that the United States is not currently a fully authoritarian state, that leaders in authoritarian states make errors all the time which result in their downfall and so it is completely predictable that Trump would do the same despite not having secured power.

And so the future was not decided on the 6th of November 2024. Rather it is still something to be contested.