It’s harder to travel light as a man, I think. Have you looked at Ex Officio? They make comfortable clothes that are designed to be washed in a sink and dried overnight on a hanger.
Are you going to be on the road the whole time or will you have a base of operations?
You seem to have the right attitude. Unshakeable self-confidence is a good founder trait, at least when combined with a small amount of doubt (most startups need to pivot, and the pivot can't happen if the founders are hell-bent on a rigid goal). Good luck!
Make your first game, sell it. Make your second game, sell it. Make your third game, sell it. Etc. The moment your income from games is comparable to your income from the job, quit your job and enjoy life full of making video games.
If you try it the other way round, chances are it will result in you being homeless.
I don't know how much time you got until the end of college, but it's October, so let's assume at least 8 months. If by the end of June 2025 you have at least *three* video games completed, generating a monthly income of at least $100, then I think you have a chance. Otherwise, you don't.
I mean, monthly $100 is peanuts, that by itself wouldn't save you from homelessness. But it's at least something, a proof of concept. It shows that you can complete the product, and you can find a customer willing to pay. Now you "only" need to get 10x better to be able to survive, and that is not impossible; practice should make you better, plus you can keep selling the old games while adding the new ones.
On the other hand, if you can't get a monthly income of $100 by the end of June 2025, it means you are not ready. You need to get a few years of experience first, and that means getting an IT job now. Don't quit the job until you have produced and started selling your third game. (You think it is difficult to make games while having a job? Wait until you have kids...)
If you're going to self promote, you might want to start using capital letters. You don't exactly create confidence that you have a quality product without them.
Scott Cawthon is arguably the second most successful indie game creator ever (after Notch), but it took him 11 years of making games before he made FNAF, his first real success. Over that time period he made 73 different indie games. All that to say, unless you have the funds to support yourself for a decade (or even just 5 years, assuming you're twice as talented and lucky as Scott Cawthon), you shouldn't quit your day job. Scott was working at Dollar Tree when FNAF took off.
Self promotion only becomes relevant *after* you actually made something. You are not there yet. (And my advice was to figure out whether you can even get there.)
Are you a CS major? I spent much of my youth/teens tinkering around with video game programming and then majored in CS. Got great grades and was always working on video game/graphics-related side projects in my spare time. After college I got a job at a AAA video game studio. Even the first year at that studio made me a FAR more competent software engineer than all my past experience combined. Problems that would take me a week became trivial. I'd highly recommend trying to get some real industry experience before taking on such a big project on your own.
I actually agree with your point on AAA games which is why I left the industry... just didn't inspire me like a lot of smaller indie games or older games do. IMO limitation is one of the greatest drivers of creativity. I'd actually love to have my own small indie studio someday.
If you haven't come across it already, I'd check out an engine like Godot.
I spent a bunch of time making a game about 5 years ago. It barely got any wishlists on steam, and while some of the early stages of development were quite fun, the processing of finishing the last 10% of playing through the same section over and over again combing out bugs was completely demoralising and took forever. I kinda wish I'd spent that time doing almost anything else and, from the research I was doing on the indie game market at the time, that's not an uncommon outcome for a lot of indies.
i'm doing that right now except also having a job which unfortunately gives me much less time to work on the game but on the plus side, food and electricity. mine is a fully text-based game so i'm really aiming for the .01% of the .01% of the .01% of audiences.
I don't know the specifics, but you probably want books on project management. In the absence of a structure large enough to include project managers, you will have to wear a bit of that hat, which includes negotiation, estimation, and presenting a case to a decision-maker.
I think you're right to ask (and right that there are absolutely movie experiences that can traumatize young kids, who sometimes don't know or can't fully process that they aren't witnessing real events), and I don't think it's a question that can be settled on the basis of some principle. It's not actually "should we shelter kids from difficult feelings or toughen them up to real life?", it's "is THIS child ready to have a salutary, if sad, experience with THIS movie?" You know your kid and his sensitivity level--does he remind you of you at that age? Does he tend to take things in stride or does he have intense feelings sometimes that don't make sense to you as an adult, does he perseverate or worry to the extreme about things related to loss?
Mine used to have very intense feelings about lost objects, which he tended to personify. It wasn't "I'm sad I don't have this thing anymore" so much as "this thing will not be OK without me to take care of it." (He also once in awhile had a panic meltdown for incomprehensible reasons, eg that a toy was lying at the bottom of a wading pool. Well that one was somewhat comprehensible, it clearly held symbolism for him.) He was an incredible packrat because getting rid of possessions felt to him like, maybe, dumping a pet by the side of the road--you don't do that just b/c it's old and not fun anymore, and in that same spirit of care we had to keep old toys, papers he had scribbled on... and oh my Lord, we left behind a rotting stick at the creek once whose tip was shaped a little like a horse's head and he brought it up for 2 years whenever he couldn't sleep. He's outgrown this completely now at 11, thankfully. All that to say, these things seemed to be proxies for him for a deep aversion to the idea that irreversible loss and sorrow exist. I literally went back and tried to find that damn horse-head stick because he could not. stop. thinking about it. (And believe me I was trying to ease him along into accepting that sometimes things are just gone. And eventually he did.)
So... I probably wouldn't have shown him the kangaroo movie at that age. (I screwed up on a few movies. I wanted him to love The Iron Giant but showed it to him too early and he thought it was sad & scary.) But when I saw him shift over to being less sensitive, which might have been around 7, then I probably would. It's not an either-or question, it's a question of when.
But if none of this rings a bell at all, if it all sounds so unlike your kid that mine just might be a space alien, maybe you should just go ahead! I do think a lot of kids could handle themes like this at 5--or be sad but in a way they can feel is helping them, as maybe you were.
Your story about your son is a good example of why you don't have to deliberately introduce most kids to tragedy and loss. The little losses of their lives feel huge to them. They are very emotionally alive.
3 year old and a boiling pot, really? I was going to fully agree with your opening sentence but I don't think you have to go *that* hard. Great if it worked out for you and all, but wow, that's really young.
If you remember it fondly then it probably does have value. But had you seen or heard similar stories at that age before you saw it? The context matters. If the first time someone is confronted with death and cruelty is watching the 1942 Bambi, it will seem a monstrous horror. But most kids who see it have already seen birds killed by cats, heard parents explaining where bacon comes from, or headshotted opponents in a tactical FPS, so Bambi is just a little sad, not a searing tragedy.
I agree that context matters. Protecting children by taking away the opportunity to watch a sad movie makes it even harder to watch the next sad movie. Before you know it you have university students asking for trigger warnings.
I saw Bambi when I was five and I cried. I still cry at movies but I know that emotions over fiction pass quickly. I can learn from them and enjoy them even while experiencing strong emotions. Better to learn early, I think.
There's definitely something weird about memories of movies watched as a child. I find (and others I've spoken to agree) that when you rewatch a movie you'll be amazed to find whole sections, whole themes, whole subplots, or whole really obvious stylistic aspects that you have no memory of. This holds even for movies I watched *dozens of times* as a child!
It's really quite bizarre. It makes me wonder if I'd watched an edited version. But no, I didn't, I just tuned things out apparently, and most children do. Just one example: a lot of 90s kids movies are full of slapstick, throughout the movie. I hate slapstick, and yet I remember liking these movies and had no memory of the sheer amount of it. I must have just tuned out the parts I didn't like.
The same applies to bad dialogue and bad acting, though part of that's lack of knowledge of what's bad and good, but not all of it.
As another parent, I don't know anything else about the movie, but I'd strongly consider showing it to him. It's important for your kids to get in contact with the full range of human emotions. In general imo most parents are on the far end of safetyism and children can handle much more much earlier than they think if you actually teach and help them along the way.
Funnily enough just yesterday I've had a related discussion with my wife; Our daughter (3) watched a children's tv show about a family of dinosaurs we haven't seen before, and then we notice that several episodes go like this: 1) the protagonists meet other dinosaurs which are just as intelligent as them 2) they get into conflict 3) the protagonists win the conflict and ... eat the opposition.
At first my wife was shocked, but after a short talk we both agreed that it might be a good counterweight. We had been complaining about some other tv shows she had seen before which often portrayed even dangerous predators as actually nice deep down and that all animals can get along great through the power of friendship. We know even quite a few college-educated adults who have completely unrealistic ideas about animals (along the lines of "bears would never attack humans unless unnecessarily provoked, anything else is animal hating propaganda") and it often strongly biases their politics in a bad way.
Obviously you still need to talk with the kid and contextualize what they see, but if you do they can handle it just fine.
I agree about kids getting in touch with the full range of emotion, but most are quite able to do that without seeing, when quite small, a movie with a tragic death in it. Think about real kids. They cry far harder and oftener than adults do, and while some crying is set off by anger or frustration, lots is set off by grief. Kids grieve when they lose a favorite toy, when they suddenly start missing their parents when at school or with a babysitter, when their feelings are hurt during play -- and also when they see sad things happen. I can remember a few times when I was small and cried in sympathy when some other kid fell down. I saw my daughter do that too when she was small. And while the things kids cry with grief over seem small to adults, they seem huge to the kids. Don't you remember being a kid, how it felt?
My introduction to death was when I was about six, being brought by my mother with her while she went to visit the deadhouse (as the hospital mortuary was referred to), where the body of a neighbour was laid out before being coffined . Everybody knelt and said a prayer, then afterwards there was the funeral and burial, and I saw the coffin being put into the grave and the grave filled in.
I did have a couple of dreams about death and being buried myself after that, but I put it together that hey, when that happens, I'll be dead so I won't know or care.
So I find it hard to think that a movie could be very traumatic as an introduction to loss, when the Kangaroo is hopping away in good health and still alive, just not going to be around Dot any more 😃
Absolutely, though, I wouldn't let six year olds watch horror movies or anything with explicit violence.
I think precisely because it's a smaller problem, it's more likely to haunt a child. Kids often have separation anxiety, and don't commonly have a self-preservation instinct (or any other sense of their imminent mortality). We had a dog die this summer and then got a new one; the conversation with my five-year-old about dog cremation was morbidly entertaining, but having the new dog run away from me on a walk and go missing for fifteen minutes clearly made more of an impression.
This should not be misconstrued as an argument against watching the movie; if nothing else, practice separation is presumably actively helpful in handling real separation anxiety.
My intuition is that certain kinds of media and themes will shock and/or distress kids the first time they see them, regardless of the age they are when they're first exposed to those themes. I've known people who were exposed to horror movies when they were six and people whose parents sheltered them until they were sixteen, and it seems like they had roughly the same "trauma" response to their first horror movies. While it might be harder for a six year old to dismiss the "trauma" of their first horror movie...I dunno. Teenagers are often extremely good at leaning into and even enhancing their own "trauma" with rationalizations for why scary stuff might be real. I know I managed to be as scared about alien abduction at 14 years old as I was of monsters in the closet at six.
You weren't asking about horror movies, but probably exposure to the concept of loss is similar. It's going to hurt regardless of the age your kid is when they first experience it as a theme. Five seems like a reasonable age to be exposed to the concept of loss and why it makes people sad, especially if it's handled in a beautiful way. And while I'm not a parent, my intuition is that it's better to have initial exposure to the themes of loss via media rather than a sudden shock of it in real life (the sudden death of a pet or grandparent, etc).
Edit to add: Be prepared for your kid to be heartlessly disinterested in your beloved Dot and the Kangaroo. He might be unforgiving of the rough animation, slow pace, etc after being trained on 2020s modern media (presuming you've allowed them to see any).
He may not cry, so go ahead. It's a way of introducing children to the idea of parting and ending of things, and at least Kangaroo isn't dead, she's just leaving to let Dot return to her human life (and it's open ended to the possibility that they might meet again later).
Sometimes people leave (they move away, they die) but while that's sad, it's not a bad thing and you go on with your life.
I think it's important to introduce this stuff to kids early. Tragedy sticks with you, and I look back on the tragic - even borderline traumatic - stories of my childhood as the most enriching.
I was not ready for Bambi when I was 5. I was ready to learn about death, but the movie presented it in the most traumatic form imaginable: the death by violence of a mommy. After seeing Bambi I understood the reality of death better, but it really left a huge dent in my sense of wellbeing. I tormented for years by stories that formed in my head about little animals left in the nest grieving, terrified and starving to death because their mothers did not come back. I I think you should err on the side of caution with kids about matters like that.
When I was college age I taught nursery school for a while part time, and when the school guinea pig died we showed kids her body the next day, and answered their questions, and let them examine her body or pet her (and then wash their hands really well). We also told parents about the guinea pig's death and how we'd talked with the kids about it. I think that was a decent introduction to death for the kids.
Movies hit kids differently I think. My son helped us bury his grandparents dog, and I think that was a positive experience. I would be reluctant to show him Bambi though.
I remember being very captivated by Bambi's father, though, and how he sort of revealed himself to Bambi (and to the child viewer) - which I don't think would have happened without the death of his mother. I remember also that frightful word - "Man!" Which implicit lesson re nature and loss has only grown more true as time passes.
Once I was sitting next to one of the local springs with a den of Cub Scouts among others, I think it was, listening to a park staffer give a little talk about the "spirit" of the springs, a variety of salamander, and she asked the assembled group if they knew what the salamander's chief predator or threat was.
We all sat awkwardly unable to answer for a few moments.
Then a kid piped up bravely, and with something of that Bambi drama: "Man?" And I think we all, adults and children, thought to ourselves, yeah, that tracks.
"Uh, good guess? Actually, it's crawfish", she said.
Dumbo seems like a good precursor to Bambi. It's wrenching but Mother doesn't die.
It was an old favorite from when you were five, so you know for a fact that at least one kid can handle it. The question to ask is whether you think your son is meaningfully different from you in his ability to handle sad stories. In general, I agree with the majority here that it's good for kids to encounter difficult emotions in fiction. And if it becomes too much, you can always pause the movie to talk about the movie and give him a chance to decide whether he continues or not.
It was an old favorite from when you were five, so you know for a fact that at least one kid can handle it. The question to ask is whether you think your son is meaningfully different from you in his ability to handle sad stories. In general, I agree with the majority here that it's good for kids to encounter difficult emotions in fiction. And if it becomes too much, you can always pause the movie to talk about the movie and give him a chance to decide whether he continues or not.
Yeah I think they need the opportunity to experience and rehearse different kinds of emotions in a safe manner while they're developing. If they have a strong reaction then have a discussion with them afterwards to help them process and contextualize the feelings, but I don't think that shielding them from children's movies is going to help much in the long run.
One thing I think is pretty true is that not only *can* most humans have all of the standard suite of human emotions, but that we *will* have them with some regularity, because the brain doesn't like to lets parts of itself just atrophy inactive forever. if you don't have an y appropriate targets for an emotion in your experiences, you will attach that emotion to *something* going on in your life, in a way that may be less appropriate and more damaging than just having an actual correct target.
For negative emotions, movies are probably a good target because they provide accurate contexts to attach those emotions to, while having those events not be something in your own life that you have to constantly fear or obsess over.
I've watched several videos of the SpaceX Super Heavy Booster going straight back to the launchpad, which is one of the coolest things I've seen currently happening in the space program. A question I've never seen answered: Why could they never recover the space shuttle fuel tank like they could with the rocket boosters? It seems like a huge piece of equipment to throw away and replace every single time.
The SRBs (solid rocket boosters) were jettisoned at a speed of roughly 4,800 km/h, while the ET (external tank) was jettisoned at over 28,000 km/h – close to orbit – so reentry was much more violent.
(numbers from Claude, so double-check them before building your own reusable launch system)
If you are going to post LLM output to make factual claims, please do us the courtesy of performing the verification yourself, or otherwise leave out the supposed details. We can all type a prompt into a chatbot. We also don't need more imaginary numbers floating about for search engines to find and become the foundation for future myths.
> If you are going to post LLM output to make factual claims, please do us the courtesy of performing the verification yourself, or otherwise leave out the supposed details.
Would you have felt better if I had posted numbers from a superficial Google search? Or from Wikipedia? How thorough and well-sourced would my verification have to be according to your standards?
> We can all type a prompt into a chatbot.
Then why doesn't everyone? State of the art LLM chat bots are perfectly capable of answering simple questions such as the above, and in great detail – enough details to enable further, independent research and verification, if desired.
I don't think I've ever gotten a truly useful answer out of it, though I haven't tried in a while. The AI worship around here is really annoying and dare I say may blind some people to its limitations.
Ironically, I find the art generators vastly more impressive than the LLMs, despite the former getting far more hate. Of course that may be why.
I think that ChatGPT is useful in finding specialized nomenclature. E.g., if one is looking for a named law or model or theorem, and one can describe what the law/model/theorem is about in layman's language, the LLM can be useful in finding the name of the thing.
On the other end, if one wants to survey possibilities and select them according to some measure, e.g. 20 lowest-boiling inorganic gases, good luck, unless some human has already compiled such a list - even if every candidate is already documented in Wikipedia, in the LLM's training set.
( And I've been steering clear of politically controversial questions, where the RLHF Woke indoctrination is likely to obscure what the _capabilities_ of the technology really are. )
Whatever Google embedded in its search is pretty awful. I just tried
> What is an example of a molecule with an S4 rotation reflection axis but no mirror planes and no center of inversion?
It replied with
>A classic example of a molecule with an S4 rotation reflection axis but no mirror planes or center of inversion is methane (CH4); its tetrahedral geometry allows for three S4 axes, making it a prime example of this symmetry element without additional symmetry features like mirror planes or a center of inversion.
which is just wrong. Methane has 6 mirror planes. In fact, _this_ LLM "knows" this. If I ask it
>How many mirror planes does methane have?
I get:
>Methane has 6 mirror planes.
>Explanation: Since methane has a tetrahedral geometry, you can create a mirror plane by selecting any pair of hydrogen atoms and passing a plane through them and the central carbon atom. This gives you 6 possible mirror planes.
"In 1962 the United States conducted its final atmospheric nuclear test series, Operation Dominic. The devices tested were designed and built by the Los Alamos Scientific Laboratory (LASL) and the Lawrence Radiation Laboratory (LRL). During the test series, LRL conducted four tests of a radically new design called the Ripple concept. Tests of the Ripple concept demonstrated performance characteristics that eclipse those of all nuclear weapons designed before or since. For numerous reasons discussed in the article, the Ripple concept was not pursued, but the technology it pioneered has been in continual development—for peaceful purposes—to this day. Until now, very little has been known about these tests and the concept behind them. This article, the result of a multiyear investigation, sheds light on the Ripple program for the first time, allowing for a largely complete account. Included are the origins of the concept and its designer, the technical characteristics, the significant role played by the geopolitical context, the test series in detail, and the cancellation and legacy of the program."
So I'm going with Chastity here since she knew what she meant in the first place and the ChatGPT did not suggest it as one possible answer, and did get the LRC and LRL confused when replying to her.
I just noticed - LRL does not correspond to Langley Research Center (LRC). So yeah, the AI is too stupid to work out that "L" and "C" are different, it's just regurgitating something from its training data.
People are asking questions here and hoping that someone has the motivation to research a real answer (or has expertise to share). It used to be that such questions were accompanied by "and a cursory search came up with these links which leave me confused" or a Fermi estimate, and it would be nice to return to such standards. Adding unverified LLM numbers as answers doesn't help, nor would "my random friend said". True/necessary/kind (2/3) are the tests we are supposed to be applying, right? https://slatestarcodex.com/2014/03/02/the-comment-policy-is-victorian-sufi-buddha-lite/
You might be right but you haven’t proven the LLM figures posted by Adrian wrong yet, and regardless of source that would be necessary to this argument, an argument that I otherwise don’t care about.
Which makes none of your comment true (or rather not yet proven), necessary or kind.
>You might be right but you haven’t proven the LLM figures posted by Adrian wrong yet, and regardless of source that would be necessary to this argument, an argument that I otherwise don’t care about.
We shouldn't have to prove a negative here. For the time being LLMs are simply not accurate enough.
Where LLMs are wrong, which is not infrequently at the moment, they're usually wrong in ways which are not easily apparent to people unfamiliar with the subject running a cursory search.
"Why was the Space Shuttle's external fuel tank not recovered" isn't a complex question which requires some unique insight only shared by five experts worldwide, two of which frequent ACX, nor does it require a Fermi estimate by the Bayesian gurus that upheld the standards in days long gone by.
Looks like "sometimes it's cheaper to throw something away rather than re-use it" is the answer, so far as I can find one.
Good discussion on a Reddit site about this question, we all got side-lined by "is ChatGPT answer good enough?" from the original question, which is "WHY did they not re-use the external tank?"
A combination of "they wanted to shave every pound of weight off" and "plans were there to use them to build a space station but never went anywhere as the adjustments would mean too expensive, too heavy, too much new equipment to make this possible", so it ended up "as light as we can manage and make it throw away to that end":
This is the very first time I haven't been disgusted by the idea of having upvotes on act, because I would upvote this comment and downvote its parent.
...no, they are perfectly capable of autocompleting a piece of text that begins with some combination of the words you typed in and whatever else the vendor chooses to prepend in a manner that results in a statistical match for text found on the internet.
This is not the same thing, because the internet is full of rubbish, and also because there is nothing in the process to distinguish between "here's the answer" and "here's a piece of text in the style that an answer would be written in, if you were given an answer". Your "numbers from Claude, so double-check them" disclaimer implies you are at least somewhat aware of this, and it would be disingenuous to now claim otherwise.
Hence people specifically wanting a response from a human: yes, humans can also be wrong, make things up and/or lie, but our well trained intuitions for how to detect that stuff at least have some small hope of matching the territory in this case; when an entirely alien mechanism is generating the text and also our mental model is demonstrably mistaken about what it is even doing in the first place, there is essentially none.
> > perfectly capable of answering simple questions
> ...no, they are perfectly capable of autocompleting a piece of text that begins with some combination of the words you typed in and whatever else the vendor chooses to prepend in a manner that results in a statistical match for text found on the internet.
Potayto, potahto. I used to think like you, until I started using LLMs in earnest. Sure, I'm still encountering hallucinations on a regular basis, but the "statistical parrot" mental model falls far, far short of their real capabilities.
> Sure, I'm still encountering hallucinations on a regular basis
Potato, potahto. Outside tech demos, when people ask questions they want actual answers and not hallucinations. It's amazing that the dog can sing, but it's not going to replace my CD collection.
I'm not convinced the "statistical parrot" model is wrong, rather I think that that's a good description for a lot of what people do. It's not a complete model of people and it's not what we mean by understanding, which is why LLMs are such a mixed bag.
If you google household ingredients for washing a floor, you get a bunch of hits for vinegar, vinegar & dishsoap, and vinegar and baking soda combined. The last of these is nonsense, because the 2 active ingredients cancel each other out. I asked GPT4 for ingredients a few months ago and it gave me vinegar and dishsoap. I asked whether adding baking soda would help, and it agreed heartily: "Adding baking soda to your cleaning mixture can enhance its effectiveness, especially for tackling tough stains and odors on linoleum floors."
Jeffrey Soreff, a chemist who posts here frequently, has posted many wrong answers it has gotten from GPT4 for chemistry questions that are easy to look up the answer to. Recently he posted that it doesn't understand what a tetrahedron is -- can't make an image even when he explains that it's a pyramid with a triangular base.
GPT4 often does no more than compile the most frequent google hits, but then it packages them so that they sound authoritative. I don't think either a superficial google search or a chatbot query is adequate for questions like OP's. You have to google for answers and then you poke around and check the one you think is probably accurate. If you don't know how to poke around and check that particular question then you just don't know for sure what the answer is.
It's also kind of rude to chatbot an answer to somebody's question. With the same amount of typing the person could have asked a chatbot this question instead of you. Obviously they are looking for a different source of information.
"rude"? What is "rude" about it? Did I insult anyone? Some people do seem to be offended, though…
I openly stated my source. Feel free to ignore such comments.
Edit: I am actually quite surprised about the general reaction to my lighthearted comment. Admit it – "double-check them before building your own reusable launch system" is at least worthy of a smirk, no?
I notice you don't respond to my main point, examples of inaccuracy. Anyhow, about the rudeness: It's sort of like answering somebody's question by sending them to this: https://letmegooglethat.com
Using LLM output in online discussions or forums can come across as impolite for a few reasons, especially if it’s clear that the response isn’t a personal one:
Lack of Authentic Engagement: Posting a generated response might make it seem like the person didn’t genuinely engage with the question or community. People generally appreciate thoughtful replies that show understanding and connection with the original question or topic.
Unfiltered or Imprecise Information: Sometimes, LLMs might generate responses that are too generic, overly detailed, or miss subtle context cues that a real person would catch. This can make the response feel like an awkward fit for the conversation and might even be misleading if not carefully reviewed.
Lack of Personal Touch or Effort: Communities often value responses that show effort, nuance, or personal insight. Posting LLM responses can seem dismissive, as though the question wasn’t worth the time to answer individually.
Potential for Misinformation: If people recognize a response as AI-generated, they may also distrust its accuracy. Unless the response is verified, it might not meet the standards of a community that values reliable, accurate information.
Risk of Redundancy or Dullness: LLM responses may sound “robotic” or repeat information already available in standard sources, lacking the freshness or original thinking that people often look for in online discussions.
When using AI-generated answers, giving credit or adding a personal summary can help avoid these pitfalls and maintain the quality of engagement.
Actually, it's not nonsense, I've heard of baking soda + vinegar. You apply baking soda to the grease on the floor, then add vinegar and mop it up. I don't know how well it works, because I've never tried it, but it's not implausible. I think the idea is that this makes the grease lift off, but I'm not sure. Or maybe it's just something that someone tried, and it worked for them.
> Would you have felt better if I had posted numbers from a superficial Google search? Or from Wikipedia?
Yes. Because those would have contained context and metadata and citations which could be further checked and traced back, and terminate in a NASA PDF or something. Even if they had contained literally the same ex cathedra statement word for word as ChatGPT, you would be no worse off in trying to factcheck it, and the *lack* of all that would have told you something useful: that it is a low-quality source of dubious veracity that may well be wrong. Meanwhile, some LLM obiter dicta kills all curiosity and is the junk food of writing: fattening webpages while providing no nutrition.
Also, "results from a Google search" should, ideally, not be credited to "I Googled and found this", but "[this site] says...", because the fact that you found it from Google doesn't tell you a whole lot about its reliability. (Google would like the fact that they brought it up to mean something, though.)
Agreed. I tend to include urls with information I find, so that people reading the comment can see exactly where I found the information (and, usually, what organization it is associated with).
On a related note - even a very superficial Google search is often improved by including the name of a plausibly authoritative organization in the search terms. ( Bluntly, I got a bit sick of the back-and-forth on the shuttle H2/O2 tank meta level questions above, so I did a cursory Google search - but including _NASA_ in the search, and then commented, quoting from the NASA site about the shuttle and citing the URL. )
For anyone that did want to know the numbers, it looks like the above is broadly right. I've done a brief Google but haven't dug especially deeply (though I see a bunch of sites that seem to agree).
The fuel tank was jettisoned after main engine cutoff (MECO) but prior to orbit (https://en.wikipedia.org/wiki/Space_Shuttle_external_tank) - the shuttle then used its online maneuvering system (OMS) engines to get thr rest of the way to orbit.
Speed at MECO was 17,000 mph (https://pages.cs.wisc.edu/~yat/space/facts.htm) which is 27.4 km/h. 17m seems pretty rounded but about the right number, and I can't find any other numbers out there. It's pretty close to the 28k that Claude gave.
This is broadly right; a massive, ongoing complaint in the 80s and 90s from the folks who would broadly be SpaceX employees today was that we constantly threw away the ET when it was almost at orbital velocity and that we should have found a way to push it to orbit and use it for space stations.
If you are going to post the output of some system of norms and values, please provide evidence that this set of norms and values has been backtested across millennia of human culture and indeed promotes human thriving. We can all judge others easily based upon our own standards as the absolute correct stance. We also don’t need more imaginary moral systems floating around for impressionable algorithms to find and become the foundation of future moral myths.
I think this is unreasonably strict. From my experience I'd estimate that LLM provided figures are no less accurate than cursory Google searches. I'd be very surprised if Claude gave figures outside reasonable confidence intervals for questions like this more than 5% of the time.
I think either people heavily exaggerate hallucination rates on cutting edge models because of bias, or otherwise I'm very very curious to see what kind of tortured queries they're giving to get such inaccurate results.
LLMs are unlikely to give figures less accurate than a cursory google search, the step where the error is more likely to enter in here is when you ask the LLM to explain something, and it gives an answer which it justifies based on the presumed relevance of those figures. The presumption in accepting the LLM's answer is that it's more likely than an uninformed person to be generating a correct answer, for which those figures are an appropriate explanation. In areas where the average person doesn't have enough domain knowledge to generate the right answer with some cursory googling, LLMs are wrong quite a lot, but this also makes their inaccuracy hard for the average person to check. It's easiest to check on straightforward factual matters which you're familiar with, but you know the average person is not.
>It's easiest to check on straightforward factual matters which you're familiar with, but you know the average person is not.
For example, I recently asked ChatGPT 4o "Which Valar took part in the War of Wrath?"
The actual answer is that the published Silmarillion doesn't explicitly name any Valar as doing more than agreeing to the expedition, uses language that's ambiguous but can be (and often is) read to imply that some or all of the Valar are directly involved (referring to "the Host of the Valar" and "the Might of the Valar" doing various things in the war), and includes details that are usually read as implying that the Valar didn't accompany the expedition (namely, the Maia Eönwë commanding the army rather than Manwë, Oromë, or Tulkas, and after the final battle Eönwë ordered Sauron to return to Valinor for judgement by Manwë as he felt he lacked the authority to judge a fellow Maia).
ChatGPT answers this question okay. It glosses over the ambiguity of the text, but the overall framework isn't badly wrong, and it offers up some mostly-plausible speculation on how four of the Valar might have been involved. It does mention some stuff that Tulkas did in the Book of Lost Tales (the earliest version of the story), but doesn't seem to notice that that was a BoLT-only part of the story.
My follow-up question, "Did the involvement of the Valar differ in different versions of the story?", intended to tease out the problems in the bit of the answer about Tulkas, resulted in some pretty bad hallucinations. For example, it says that in the Book of Lost Tales, "Tulkas, Manwë, and others were imagined as physically fighting in the War of Wrath." Tulkas did explicitly take part in the War of Wrath in BoLT, but Manwë and the rest of the Valar emphatically did not. Manwë actively opposed the expedition in BoLT and Tulkas, most of the Elves of Valinor, and many of the "Children of the Valar" (i.e. Maiar) defied him and went anyway. ChatGPT also badly overstates Tulkas's involvement in later versions of the story (where he actually isn't mentioned at all) and brings up some stuff that I'm pretty sure is hallucination about Ulmo being explicitly involved in some versions.
I recently posted an example of such an encounter with an LLM (I always ask Google Gemini, because I don't want to sign up for an account, but I already have a Google account).
I asked a question taking the form "here is a couplet from a broadsheet ballad - what does the singer mean by these lines?", and noted in an earlier thread that the answer I received was abysmally bad.
But, of note, I got a response in that other thread saying that I shouldn't be calling that a bad answer because it looks like a good answer if you're unfamiliar with the facts.
It's still not clear to me why that should make the answer better.
I'd say it makes the answer worse! Because if it's *obviously* wrong, you're going to catch that and not propagate it, but if it looks plausibly right, you might be fooled into thinking that it's trustworthy unless proven otherwise.
Along with other types of nerdery more commonly represented on this blog, I'm also a martial arts nerd, and I've spent a fair amount of time asking ChatGPT questions about martial arts. My takeaway is that ChatGPT is quite familiar with the sorts of names people tend to mention in association with martial arts, the sorts of adjectives people use and which styles are most frequently mentioned, but its accuracy in actually answering even basic and straightforward questions related to the martial arts is much worse than even cursory googling. But to someone who doesn't actually know anything about the subject in question, it sounds perfectly credible,
tldr - in the first case ChatGPT o1 wound up getting the explanation for the color of CuCl4 2- badly wrong, and I had to lead it by the nose to force it to finally cough up the right answer ( detailed transcript of the session at https://chatgpt.com/share/671f016f-3d64-8006-8bf5-3c2bba4ecedc )
in the second case whatever Google is embedding in its searches (Gemini???) falsely claimed that methane has no planes of mirror symmetry (in the course of giving methane as an incorrect answer to my original query)
Just to be clear: I _WANT_ AI to succeed. I would very much like to have a nice quiet chat with a real-life HAL9000 equivalent before I die. It is probably the last transformational technology that I have a shot at living to see. But it is _not_ reliable (nor at AGI) yet.
If I ask an LLM something that is easy to Google, it is likely to give a response that is close to that answer. I seldom ask an LLM for such things, because I usually try searching first based on likely keywords (I often want to go deeper so I need useful further links, not a tepid summary, and this saves time). I probably have a higher prior on incorrect hallucinations than someone who goes to ChatGPT first.
LLMs are currently more likely to be misleading...but perhaps not by a huge margin. Most of the answers I get to web searches are quite wrong, and usually obviously so. (Most of them are so wrong they're irrelevant.) But I ignore the (blatantly) wrong search responses. LLMs tend to give one answer, and when it's wrong, it often isn't obviously wrong.
He reported that they *were* from a Chatbot (and which one) which is the important part. He gave his source. Most web searches don't yield a verifiable source either, and some of them return invented answers. (Not being invented by an LLM doesn't mean they weren't just invented.)
>After the solid rockets are jettisoned, the main engines provide thrust which accelerates the Shuttle from 4,828 kilometers per hour (3,000 mph) to over 27,358 kilometers per hour (17,000 mph) in just six minutes to reach orbit. They create a combined maximum thrust of more than 1.2 million pounds.
The Space Shuttle was a terribly suboptimal design hobbled by political compromises. It is a wonder it looked and worked as well as it did. NASA was already ossifying into a terrible bureaucracy, slowly losing its skills and spirit from the glory days of Apollo. Adding fuel tank recovery and refurbishment would have added years and billions of dollars to the schedule and budget, it was not even seriously considered. Even for the modern SpaceX catching the booster is pretty audacious, and it was maybe one second from a failure, according to the Musk's accidental Diablo sound overlay.
Alas, Starship can't get past Low Earth Orbit, nobody has an in-space maneuvering stage that can fit inside a Starship and take a Clipper to Europa, and neither of those things is going to change in two years even if you tell the engineers to get started today.
It depends what the downside of the Shuttle not having the specific military mission capabilities would have been, had the occasion to use them come up.
There are two parts of this:
1. Could a non-shuttle launch vehicle perform the mission?
2. How much of a luxury was the mission, i.e. what happens if we can't do it at all?
For 1, I understand the answer was mostly yes. Launching large spy satellites (which I understand to be the main driver for payload size and polar orbit capabilities) wound up mostly being done by disposable boosters (Delta and Titan, IIRC) anyway.
The main leftover mission I'm aware of that other launch systems couldn't do was to snatch a Soviet satellite out of orbit and return to Vandenberg. I am not familiar with the thinking that this would be a message or desirable thing to be able to do, so I will tentatively classify it as a luxury mission.
Right, that is a good point. A better (and more expensive during the design stage) approach would have been having a configurable setup where, like with the SpaceX Falcon Heavy, the boosters could be recoverable unless the mission profile forces them being expended.
How many enormous projects are ruined by requirements which are decided in advance, which turn out to be unachievable, but which then can't be changed later on once we learn more? It seems like the answer is "most of them".
If Starship had stuck with its original specs, it wouldn't have worked -- they needed to try a few things and figure out what was practical and what was not. On the other hand, Elon doesn't have a flawless record here either, and the Cybertruck suffers from similar problems where it's a worse vehicle than it would have been if they hadn't made certain dumb commitments at the planning stage.
In addition to what others are saying, the big orange fuel tank was actually the cheapest part of the Shuttle, by far. It had no engines, so it was basically just expensive pipes, tanks, and insulation.
In order to make that reusable not only would significant weight be added, it would also make it more expensive. The savings in reusability would have been more than cancelled out by the lost payload and refurbishment costs.
As I recall (from reading a fantastic book on the history of the Challenger disaster which I recommend here without any reservations https://www.amazon.com/Challenger-Story-Heroism-Disaster-Space/dp/198217661X); the original plan was to have two part launch system, where the shuttle is first flown on a carrier to a suitably high altitude and then launched from there for whatever it's mission was.
Both parts were envisioned to be re-useable but the cost was well, astronomical.
In this case, the chatbot got it right - the external fuel tank carries all the propellant the shuttle's main engines will use taking the shuttle all the way to orbit (well, except for a small circularization burn with the maneuvering thrusters). So the tank can't be discarded until the Shuttle is at orbital velocity, roughly 8 km/s. At that point, there's no question of it coming back to the launch site or parachuting into the ocean anywhere near the launch site; it's going to come down halfway around the planet.
And it's going to be subject to the same sort of reentry heating environment as the Space Shuttle itself. A simple aluminum tank with just some spray-on insulation to keep the propellants chill before launch, is not going to survive that. A tank which could survive that, would probably weigh enough that the already-marginal Shuttle couldn't carry any actual payload (and certainly not the big military spysats that were part of the requirement).
The only remotely sensible proposal for reusing the Shuttle external tanks was to take them *all* the way to orbit, and then use them as pressurized habitat or propellant-storage elements on a large space station. A single external tank would have more interior volume than all the pressurized elements of the current ISS combined. But nobody had the budget to build a space station that big even if they got the pressure vessels delivered to orbit for free, and their orbits would have decayed long before NASA got around to using them, so they just ditched them in the ocean instead.
The space shuttle fuel tank was just a big tank, it didn't contain anything capital-intensive or fancy like advanced rocket engines. Those were on the Shuttle itself, and those were recovered.
Even recovering the SRBs didn't make sense, because they were only "reusable" in a marketing sense: the cost of fishing them out of the ocean and refurbishing them was greater than the cost of just manufacturing additional SRBs, but reusability was one of the justifications for the expense of the shuttle program so reusable that was deemed.
I mean, the simple answer as to why they couldn't recover the fuel tank was because the entire launch stack was designed around a set of premises, and one of those premises is that that tank was going to be jettisoned and break up on reentry instead of being recovered, and if they'd wanted to recover it that would have required a fundamentally different spacecraft than the one they designed. The Super Heavy Booster is an entire rocket, with engines and electronics and computers and cameras and radios and miles of wiring and sensors. The external tank was just a tank.
It never really got past power point engineering as far as I can tell, but ULA had a proposal for Vulcan that involved basically detaching the engine section and recovering only that - for basically that reason, the majority of the cost of the rocket is the engines and avionics, while the tanks (basically just big empty aluminum cans) are bulky, kind of delicate, and therefore hard to recover.
What would happen to our society if a large-scale, long-term blackout occurred? There is a high chance that it would get quite bad very quickly. Transportation and health services would likely cease to function within a few days, and many people would face food and water insecurity almost immediately. This highlights the urgent need for greater investment in preparedness, as there aren't even exercises to train those responsible for managing such crises. If you're interested in more details, I have written a new post in my living literature review that offers a deep dive into the consequences of blackouts: https://existentialcrunch.substack.com/p/the-consequences-of-blackouts
> What would happen to our society if a large-scale, long-term blackout occurred? There is a high chance that it would get quite bad very quickly.
I think you've independently reinvented the best argument for "prepping."
It's less for the Big One or Zombie Apocalypse and more for longer stretches without power and services brought on by severe weather and / or state inadequacy.
Also the strongest argument for your own solar + battery setup, re Scott's last post.
I like that you're optimistic and think something can be done collectively - I personally believe the only "real" solution is personal and independent, because you can actually prep and get your electricity off the grid through your own efforts without having to persuade a lot of other 'general public' people who don't believe in thinking ahead.
Yes having a backup for smaller catastrophes makes sense, but I generally think that for everything taking longer than ~ 2 weeks, is is important to be able to rely on the state. Otherwise, everything will turn quite bad.
Generally, I don't think long term prepping really works. I've been looking into global catastrophes and societal collapse a lot and my conclusion is that prepping for a duration longer than two weeks mainly buys you the privilege of dying a bit later than the rest.
I think that if the society you live is generally unprepared, most people would die. You might live longer if you are prepped, but I doubt it would be be a great life to have. But, we could plausibly make it through large catastrophes, if we take the time to prepare. And generally, I think we are going in this direction, albeit slowly. For example, more and more countries are starting to consider large scale catastrophes in their risk assessments.
This is a common fantasy but just seems like a way to make your eventual death more cinematic. I've seen Die Hard and Home Alone too, but I don't like my chances against an armed gang, because the defining aspect of gangs is that they outnumber me. I mean, *maybe* I can use my superior planning and knowledge of the terrain to my advantage and fight off dozens of dudes on my own, but that sounds like a fantasy rather than a plan.
If it comes down to a world where we're having gunfights over the world's remaining food resources then I'm probably dead. But there's a bunch of far more likely scenarios where having a bunch of sensible preps will turn a horrible situation into a much more comfortable one, even if the people without preps aren't dying, they're just spending their days standing in line waiting for supplies.
I'm not a real prepper, but I think part of prepping is also training. No matter how much food you have stored, you'll run out eventually. Do you have the skills to obtain more, whether by growing, finding, hunting, or something?
Same goes for everything else: heating, cooling, washing, repair work, etc.
In Gaza, the optimal approach to prepping would probably involve finding out a lot of intelligence about Hamas and then offering it to Israel in exchange for passage to elsewhere.
At some point, you're prepping for the collapse of civilization and that's beyond your resources. But a couple weeks without power is not the end of civilization, it could happen, and having some notion of how you'll heat your home/charge your phone/cook your meals is probably smart.
A comment in support: a few years ago, we lost power for 11 days after an ice storm. Now, the greater area we were in didn't lose power for nearly that long, but because we live in a rural area with a lot of trees, there were hundreds of line breaks and it took a long time to fix them.
Depends. I can see solar panels and batteries, and not doing much at night for a couple of weeks. It wouldn't even be that bad. You should have a bit of food storage, but that should be easily doable in a rural environment (i.e. space wouldn't be a problem).
What do preppers do about water? Seems like that's the biggest barrier if there's a long term catastrophe, unless you live but a river, and it's hard to store a month's worth of water like you can with food
Storing water treatment supplies is easy. Getting access to actual cleanable water is harder (most preppers, I'm guessing, live in rural areas where that's a lot easier)
FWIW, best practice is to fill them with standard tap water, maybe add a tiny bit of plain chlorine just to be super extra safe, and then dump and refill them every 2-3 years. Because I'm lazy and they weigh 40 pounds, I didn't refill them for 4 years. Then I had to empty them due to water damage from a broken sprinkler and decided to taste the water just to see how it was. It was perfectly "fresh," tasted exactly identical to running tap water.
Many preppers live near streams or springs (or have wells) for this reason as well; small-scale water filtration is remarkably cheap and extremely effective these days, so you don’t really need to worry about the centralized distribution aspect as long as you can access any water at all. The exception would be in the case of significant contamination from pesticide runoff, or industrial chemical dumping, or viral load… none of these are common concerns in the USA.
On any given day I have a 16-20 day supply of water *and* a reasonably stocked bug-out-bag, Plus a gun and ammo to defend them/myself in my condo, which is 1/2 mile outside of literal downtown Seattle.
I am not at all interested in living in the aftermath of a civilization-ending collapse, so if it looks like it's going that way, that gun will have a different purpose. But I don't see any reason to needlessly suffer or potentially die during local or moderate disasters if I can either literally shelter in place until the lights and water are turned back on, or hike out to an area where I can then be evacuated to my family in california, or friends elsewhere in the country.
If you ration it to a gallon per person per day (plenty for drinking and cooking, but very little left over for washing), a standard tank-based home water heater holds 1-2 person-months of water. It relies on the main water supply's pressure to push it to the taps, but even without that you can still draw water from the drain valve at the bottom of the tank.
If you have a little bit of notice of an impending natural disaster that might interrupt water supply, you can also fill up your bathtub and store additional water there. Although if you use tub water for drinking, the standard recommendation is to add bleach (1/8 tsp per gallon) or water purification tablets to sanitize it.
Relying on a water heater and/or bathtub for potable water in an emergency is ill-advised! Certain kinds of natural disasters or deliberate attacks are likely to taint the water supply coming into your water heater, and even if it doesn't, you have to have both equipment to drain the heater and more storage containers if you need to evacuate.
Yes, it's not a complete solution. I bring it up because there's a wide range of situations where using your bathtub or water heater will give you a fair amount of additional potable water, or at least water suitable for washing.
My water heater doesn't take any special equipment to drain. There's a standard hose bib on the bottom of the heater. I turn the knob and water comes out. If you're using it for its primary intended purpose (flushing out sediment from the bottom of the tank or emptying the tank for maintenance), you want to attach a garden hose first so you can direct the water so it goes to the storm drain instead of flooding my garage. But if you're tapping it for drinking water you just need a clean container to catch the water.
Read _Dies the Fire_ by S.M. Stirling. Of course, in the scenario of that novel, not only did the power go out permanently, but all high-energy chemical reactions stopped working. So firearms stopped working, too. YMMV if it's just a long-term power outage. ;-)
Agreed, and I think large-scale USA blackouts might happen shortly before or on election day. I can't think of a better time to for foreign actors to sow further discord in the US, if not provoke an actual civil war, and my understanding is that it's relatively easy infrastructure to attack.
Also, I'm a superfan of theprepared.com, which is basically the Consumer Reports of emergency preparedness products, as well as having a lot of general education.
i’m reposting here a question from the old expiring open thread that didn’t get much response, which I’m hoping is because jt went up after the thread went stale.
I know many democrats who are genuinely terrified of Trump winning. The say he will actually do all the worst and wildest stuff he’s just talked about doing so far, because he will quickly install appointees everywhere. He will quickly become a dictator, jailing those prominent people who speak out against him or take legal action against him. I would like to know what impediments there are to this happening.
Last night I googled “impediments to Trump ruining country” and every single hit I got was about how Trump is for sure going to ruin the country. A typical one was a Washington Post article “A Trump Dictatorship is Increasingly Inevitable.” Inevitability doesn’t even *come* in degrees. WTF?
For instance, what impediments are there to Trump’s bringing some charge against Chuck Schumer, winning the case, and getting the guy locked up for a few years? So one thing I wonder about is about legal constraints.
Another is practical constraints. For instance the logistics of rounding up 10 million or so undocumented immigrants seems pretty daunting. Who rounds them up? Where do you put them til they are shipped out? How do you transport them back to where they came from? What do you do if their country of origin won’t take them back? For that and for other proposed Trump plans, it seems to me you need staff who are not only willing to carry out such plans, but are also skilled
at the logistics involved. Seems likely potential Trump appointees are carefully vetted for loyalty, but not for skills (to vet for skills, you need skillful people).
A third is resistance by local governments. What happens if the government of a blue state objects strongly to some Trump plan being carried out? Seems like many things require the cooperation of local officials. Wouldn’t some states refuse to cooperate in carrying out plans they are strongly opposed to? Would Massachuetts, for example, help round up the undocumented immigrants in its state? And then there would be some sort of legal fight about that, but surely it would drag on for a long time.
Just to be extra clear in these contentious times, I am not interested in debating anything having to do with this topic. it’s a request for information.
Later edit: I am horrified by the way people are answering. Dozens of replies have rolled in with no substantive content, just general impression kinds of stuff: People worried last time Trump was president and nothing that awful happened, others in government will restrain him, Trump actually believes in democracy, the Democrats have done all the bad shit they say Trump will do, no they didn't, you are a meanie asshole, you are a poopy head, you eat boogers for breakfast, etc etc.
Finally Daniel actually named 2 things that are impediments. I wish he had had more to say, but his was a genuine answer.
All you smart and usually admirable people are melting down so badly you can't even remember the question. This is the worst I have ever felt about ACX.
It doesn't matter whether you hope there will or will not be lots of impediments to Trump taking the steps he wants to once elected. The question is what impediments will he encounter if he attempts to carry out the policies the Democrats are most strongly opposed to: Legal impediments, practical ones, pushback from local governments, pushback from powerful interest groups.
Still later edit: I looked around some, and this article seems pretty intelligent to me, and honest. Though of course one never knows for sure. It's by a Turkish journalist:
It's worth noting the Ukraine/Russia conflict was merely less intense during the Trump administration, but went on the whole time. I don't think it makes sense to imply that he deserves credit for the fact that Russia didn't increase the tempo until later, particularly when his own actions mostly involved delaying or threatening to cancel support for Ukraine, and since then he's contributed to a collapse of pro-Ukraine sentiment in the Republican party.
Also he's openly advocating for Israel to go all-out against Iran.
2020 was just a continuation. Trump said the election would be stolen in 2016. Previously he said the 2016 Iowa caucus was stolen. He was asked repeatedly if there would be a peaceful transfer of power and he declined to answer.
There is no world in which he would admit 2020 was legitimate.
"Well, they’d just watched the establishment unite against their guy for four years, impeachment based on dubious pretenses, Russiagate based on the Steele dossier which was transparently a DOJ plant, Obama spying on the Trump campaign, etc. And the media refused to admit that any of it was wrong."
As well as all the back-patting stories later published in the same media about "yes, we *did* co-ordinate to make sure the election was won by the right side".
Shadow campaigns are not anti-democracy when *we* do it!
"Their work touched every aspect of the election. They got states to change voting systems and laws and helped secure hundreds of millions in public and private funding. They fended off voter-suppression lawsuits, recruited armies of poll workers and got millions of people to vote by mail for the first time. They successfully pressured social media companies to take a harder line against disinformation and used data-driven strategies to fight viral smears. They executed national public-awareness campaigns that helped Americans understand how the vote count would unfold over days or weeks, preventing Trump’s conspiracy theories and false claims of victory from getting more traction. After Election Day, they monitored every pressure point to ensure that Trump could not overturn the result. “The untold story of the election is the thousands of people of both parties who accomplished the triumph of American democracy at its very foundation,” says Norm Eisen, a prominent lawyer and former Obama Administration official who recruited Republicans and Democrats to the board of the Voter Protection Program."
"They got states to change voting systems and laws". No big deal, not autocratic at all!
"Sometime in the fall of 2019, Mike Podhorzer became convinced the election was headed for disaster–and determined to protect it.
This was not his usual purview. For nearly a quarter-century, Podhorzer, senior adviser to the president of the AFL-CIO, the nation’s largest union federation, has marshaled the latest tactics and data to help its favored candidates win elections. Unassuming and professorial, he isn’t the sort of hair-gelled “political strategist” who shows up on cable news. Among Democratic insiders, he’s known as the wizard behind some of the biggest advances in political technology in recent decades. A group of liberal strategists he brought together in the early 2000s led to the creation of the Analyst Institute, a secretive firm that applies scientific methods to political campaigns. He was also involved in the founding of Catalist, the flagship progressive data company."
Wait wait wait - I thought Cambridge Analytica was the devil's handmaiden for doing the same kind of thing? But of course, they were on the 'wrong' side, while the "secretive firm" was working on the 'right' side this time!
I get the impression he just couldn't stomach admitting he lost, not that he tried to stay in power. Compare what real dictators do when they attempt a coup: the military arrests the opposition.
I specifically didn't use the word coup, because, while it might technically be one (although maybe not, I've seen argued that the definition of coup is fairly stringent), it'd a very mild version of one.
But I don't think we can just give him a pass for a pretty naked power grab, particularly with reasons like "he didn't really mean it, guys".
If Trump actually beleived he won and was cheated, his actions take a different tone. I think it's plausible Trump beleives he won because much of his persona is built around magical positive thinking around his capabilities and what the world owes him. Also, there is a >0% chance he *was* cheated in the 2020 election, but of course we will likely never know for sure.
I remember when January 6th happened, my thoughts were something like, "oh, they're finally doing something about being ignored, but this won't end well."
I haven't done in-depth research to find out what election challenges were placed and what they said was wrong with them, but I heard nothing of substance about it from news media, except that all challenges were refuted. The instances of voter fraud that I HAD heard about, such as finding some military mail-in ballots that wouldn't have been counted, were only in favor of Democrats (I had heard of only a few hundred such ballot misplacements).
I mean personally if he honestly believes it because his ego and magical thinking put him wildly out of touch with reality, I don't really see that that's much better than a cynical power grab.
>Biden didn’t talk much about prosecuting Trump, but his DOJ has certainly done so. If you want to know what politicians believe, watch their feet, not their mouths - Trump has much more respect for Democratic institutions than his opponents.
Biden does not control the DOJ and what the DOJ went after are crimes connected to the fake elector scheme which Trump was actually involved in. The DOJ also went after Biden's son and Biden did not stop it, did he? Trump however has pardoned Manafort, Stone and Kushner. All people he has direct ties to. The fake elector scheme alone which led to a number of convictions is tied to Trump directly and is absolutely as undemocratic as you can get.
I think you are either very misinformed on this topic or are willfully spreading wrong information.
This is incorrect - one of the impeachments of Trump was about him asking the premier of Ukraine to give him dirt on both Joe Biden and Hillary Clinton (the latter based on the nonsensical conspiracy theory that her email server was hidden in Ukraine). He also appointed a special counsel just to investigate the prior investigation of himself. He has also, repeatedly, suggested that several of his political opponents are guilty of treason and gone so far as to suggest that the national guard or military should deal with them. https://www.npr.org/2024/10/21/nx-s1-5134924/trump-election-2024-kamala-harris-elizabeth-cheney-threat-civil-liberties
Oh come on now. What in Trump's history or personality suggests credibly that he would actually _de_ politicize the DoJ? Talk about drinking kool-aid, you're swimming in it there.
He dropped prosecuting HRC simply because he hadn't figured out that he'd need to have put his own loyalists into the top jobs; know he knows better.
No arguments here about Clinton or Nixon. The former should have been removed from office and the latter got away with multiple prosecutable crimes. (I get Ford's logic on the pardon, which was much more specific than simply "everybody needs to move on", but it was the wrong call.)
Just for the record though Trump was not charged or convicted for banging the porn star or for paying her to shut up about it. Also the current DoJ is absolutely correct to be trying to prosecute him and others for things like the fake electoral-college slates and the phone call to the GA officials and related actions which attacked the very heart of our constitutional system.
> If you want to know what politicians believe, watch their feet, not their mouths
This is not good advice, we must watch both feet and mouths. What politicians say they will do is, more often then not, different than what they actually do, but it's not unimportant or uncorrelated. What they say has cultural power: it slides the window of what is considered acceptable (to their followers) and determines what their opponents will be thinking about.
Trump saying "I didn't actually lose, there was election fraud" could have been a testing of the waters, a check of which powerful people and institutions would agree with him, and thus hypothetically give him support should he attempt some sort of coup.
I'm not sure that's what actually happened, but the fact that it's possible already proves that, when politicians say things, it's not just words, that is an act with consequences. A similar reasoning can be applied to him saying he was going to prosecute Hillary during his campaign: even if he never was going to do it, he was (deliberately or not) creating a cultural ambient where a president prosecuting their enemies is something somewhat more plausible.
Agreed. Historians routinely use "what politicians say" as one of several proxies for what they believe. The trick is to triangulate multiple sources when available and to read "what they say" critically in light of who they're saying it to and what they're trying to accomplish by saying it.
I don't think you can avoid a values debate in answering a question like this, because parts of your question make implicit values judgements, or at least seem to. For example, you seem to be putting "retaliation against political opponents through the criminal courts" in the same rough category as "rounding up and deporting every illegal immigrant". Note that I personally disagree with the latter policy (and obviously the former too...) but lumping "standard conservative policies", "extreme conservative policies", and "dictatorial actions that threaten democracy" together is something people on the left did throughout Trump's first term and it made it all but impossible for anyone to discuss anything with any clarity. Even here, there've been comments saying things like "Trump did several radical unpopular things like refusing to concede the election and appointing the justices that overturned Roe". Any statement like that (listing threats to democracy and standard right-wing policies you disagree with) is certain to make anyone not already on the left stop taking you seriously.
Note that "you" is generic, not about you personally. I think answering your question with any clarity would require very clearly separating "policies that a lot of people object to" from "threats to the democratic political structure itself".
<Any statement like that (listing threats to democracy and standard right-wing policies you disagree with) is certain to make anyone not already on the left stop taking you seriously..
You could still answer my questions, though, even if you don't take me or my frightened friends seriously. Surely the question itself is worth taking seriously. I restate it below, as a nudge to all the people who are answering based on vibes and general impression kinds of things, or else just starting to fight about who's the biggest asshole, Trump, people voting for him, or people voting against him. My question is what impediments are there to (1) Trump doing the things lefties are particularly horrified by. (immigrant deportation being one, I think the others are pretty well known) (2) Trump doing wacko things like nuking hurricanes (3) Trump disabling the election process, and becoming able to stay in power until he chooses to leave, appoints a successor, dies or is overthrown. By impediments I mean laws, practical difficulties, actions by other parts of the government such as local officials, & pressure from large interest groups (eg businesses who want to keep undocumented immigrants here because they depend on their labor).
One concern now is that Trump's incipient dementia is both quite obvious and a good deal more advanced than Biden's was in 2020. [It's important to compare apples to apples on this topic, not Trump now to Biden in 2024 but to Biden in 2020.]
That doesn't guarantee that Trump will now quickly slide into the full-on dementia stage -- real life is that the pace of that has a lot of individual variation and isn't predictably linear, and also as POTUS he'd of course have top-notch medical care. But sensibly it does create realistic odds for a mental-capacity slide greater than the one that Biden has had while in office. Maybe only a little greater, maybe a lot -- no way to know in advance.
If that does happen it will _not_ quickly or certainly lead to Vance taking charge. The VPOTUS has exactly zero executive authority in our system until the moment he gains complete authority, either by the POTUS' death or resignation or by the full cabinet invoking the 25th Amendment. Trump will never agree to resign so long as he can still speak sentences. The 25A has never been tested, and also would a bunch of Trump appointees have the nerve to do that while Trump is still able to walk and talk? Color me verrry skeptical about that scenario.
If the POTUS is a dementia patient and the VPOTUS is not being placed in charge, and unlike the first Trump administration this one really is full of Project2025 contributors in all the top non-elected positions....that's a basis for taking seriously what the latter crowd explicitly wants to do. And having read their detailed written plan, I take that very seriously indeed.
Reminder: if the Pres fights the 25th, it then requires 2/3 of each House of Congress to remove him. There are definitely cases where this would lead to some very interesting decisions for Dem officeholders.
How so? No Republicans in Congress would vote against Trump, and there certainly won't be close to 2/3rds Dems in either house. So for Dem officeholders it's a freebie, they can do whatever they want on that vote without changing the outcome. Vote yes if they represent a hard-blue district, or not if they don't, whatever.
Doesn't matter though because there's zero chance that a majority of a Trump cabinet would do that, and also Vance wouldn't have the balls to sign on.
I don't think you bit. Even if the stuff in the leftie press is not an accurate representation of what Trump will do, I am asking what impediments there would be to his doing it. Do you know? You can consider answering it an intellectual exercise. Even if Trump isn't going to try to various crazy things, it surely makes sense to consider what guardrails there are against some president doing crazy things. The impediments I have in mind are legal impediments, impediments built into the structure of government, practical impediments, resistance from local government, pushback from big interest groups.
What are your reasons for thinking "what Trump publicly says he'll do" bears a strong relation to "what Trump actually wants to do"/"what Trump will do"?
I get the feeling you see much of the criticism of Trump as hysterical and misguided, but I'm sure you've noticed that he does not have a strong commitment to truth-telling.
"businesses who want to keep undocumented immigrants here because they depend on their labor"
I think a lot of the problems around immigration could be sorted out if there was honesty on this: let businesses like construction and agriculture be open about "we need cheap labour; the natives won't do it, or if they will, they expect better wages and conditions than we can/are willing to provide, so we need immigrant workers". If the liberals are horrified about exploiting minorities, then they'll have to accept that the workers can't come across the border because they *will* be exploited if they do.
Either way, set a target of "who can legally enter as immigrant labour" and *stick to it*. People who break the rules get sent back. Those who follow the rules get a chance for citizenship or at the least, working their way up into better jobs. Decide if Jose the fruit picker gets the vote after living and working in the US for ten years and both sides *stick to that*, no blurring the rules that technically Jose shouldn't vote but... he has a driver's licence and we don't impose ID rules on voters so if he does vote, naughty Jose but nothing is done about it.
I think mostly people are pissed off about the "illegal" part, not the "immigrant" part, despite the attempts to paint any opposition as racism. Put structures in place, let the people see that those are being implemented, and you'll cut back a lot of the objections. Jose follows the process to come here and pick fruit cheaply for MegaFarm Corp, there's a better chance he can integrate into the local landscape. Felipe comes here illegally and gets booted back over the border, everyone agrees the rules are being followed, nobody has a problem with Jose being here and Felipe not being here.
>By impediments I mean laws, practical difficulties, actions by other parts of the government such as local officials, & pressure from large interest groups (eg businesses who want to keep undocumented immigrants here because they depend on their labor).
Laws are very important, but largely an imponderable. AFAIK, every recent administration has done at least _something_ that is arguably illegal or unconstitutional. And yet, I can't tell at what line Federal employees would refuse to follow illegal orders.
One practical constraint that I'm reasonably sure of: There isn't enough prison space in the USA to imprison all of the voters voting for the losing party. Peak total incarceration in the USA in roughly 2008 ( https://en.wikipedia.org/wiki/Incarceration_in_the_United_States ) was about 2.3 million. This is more than an order of magnitude less than e.g. the roughly 81 million votes for Biden in 2020 (and similar numbers for Trump).
Might Trump dissolve Congress? Might Harris suspend the first and second amendments? I have no idea. Both are clearly illegal and clearly physically possible. Would Federal employees go along? I have no idea.
Thanks. Those things you have no idea of? I have no idea of either. And I'm not sure many people arguing about the election do either. And yet people are turbo-powered and savage in their certainty about what should happen and what will happen.
ES: 2 cents, educated guess, not very familiar with Trump's administration.
We live in times where, frequently, a politician who explicitly and/or implicitly has some crazy ideas gets elected as head of state. The usual pattern seems to be that, once in power, they actually do implement lots of crazy policies and make crazy decisions, but way, way less then both their followers and their opposition thought they would. It seems that was the case with the first Trump administration: he neither became a dictator, like some anti-MAGA people said he would, nor did whatever the Q-Anon people thought he would do. But he did build around 52 miles of new primary border barriers. (This is not to understate the consequences of the Trump administration, just saying, it's less then what he seemed to promise during campaign.)
I think that, aside from legal limitations to the power of any politician, there is a big social-cultural factor that keeps them from doing anything too out of the Overton window: behind the curtains, there must be a set of people that tell them that it's a bad idea. I imagine Trump must have heard the phrase "sir, this wouldn't work/isn't possible/isn't viable/doesn't work like that" multiple times per day when he was president. I have the impression that this might actually accounts for a significant amount of the total impediments there are to politicians going mad with power.
My model for this is something like the following: assume some president would like to become a dictator (you can substitute that with "implement some radical policy they like"). A politician is either:
- type A: believes that, if they just did what they actually wanted to do, their opponents would use a lot of resources to remove them from power, and not enough of their allies would back them up to preserve them in power.
- type B: does not believe the above.
If the president is type A, they don't try to become dictator. If they are type B, they necessarily need to consult some of their allies and underlings in order to attempt some sort of coup. If an ally is type A, they advise against. If they are type B, they consult their own allies and underlings, and the process continues recursively.
> The usual pattern seems to be that, once in power, they actually do implement lots of crazy policies and make crazy decisions, but way, way less then both their followers and their opposition thought they would.
This seems like the kind of heuristic that works until suddenly it does not. Currently in Slovakia we have yet another government of Robert Fico. Previously he was the kind of guy who has big mouth, but luckily after election only does maybe 1/4 of what he promised. But not this time. He reformed the criminal law practically the day he got elected (reduced the status of limitation of many kinds of crimes; mostly economical but also e.g. rape), and fired all prosecutors and policemen who in the past acted against the members of his party. Crimes bosses that it took literally decades to put in prison are now walking free. He also established some new taxes, and other minor stuff I don't remember. Shortly: mostly harmless, until suddenly he was not.
I suspect that the most dangerous moment of a populist politician is when he starts losing popularity. When he can no longer be certain that he will win another election... even if not the next one, then definitely the one after that. Especially when it seems that if he no longer has the immunity and power to stop prosecution, he might serve some prison time. Then it may turn out that until now he was playing on easy mode (just telling the voters what they wanted to hear, and mostly doing nothing), and now he is playing on hard mode (trying to destroy anyone who looks at him the wrong way, and desperate enough to take any kind of allies).
> […] is Increasingly Inevitable.” Inevitability doesn’t even *come* in degrees. WTF?
Come on, it obviously means "approaching inevitability".
When someone says, "my trash can is increasingly full", they mean that it will soon have exhausted its capacity for additional trash, even though "full" is a binary condition, like "inevitable".
Yes, I understood what they meant. But there are adjectives that denote a certain state that does not exist in partial forms -- for instance "infinite," "pregnant," "dead." It irritates me to see words used in ways that blur their meaning. Their architecture is lost, their crystal structure. But my objection to the Post headline is mostly based on something else. It seems like a way to summon maximum horror without actually committing to a prediction. Sort of like saying "one day soon blood may come spurting out of your mouth and ears."
The government was able to deport 100,000s of immigrants in the 1940s, it is interesting that most people don't think the government have the state capacity to do that in the 2020s.
I'm not sure the government doesn't. But organizing it does sound daunting. Covid made me much more aware of things like the difficulty of transporting things to large numbers of people -- for instance there was a whole nasal swab debacle. Reading up on that stuff, I came to see that for large tasks the government does not routinely do, somebody in government has to locate the people here and there, often civilians, who happen to have the required skills and info, and draft them to come work on the current project. So what I was wondering about was how likely it was that the people Trump's backers have found to install who would manage such a deportation have been selected for the ability to manage the task, or locate specialists who can.
I don't doubt there's going to be a lot of "this is paranoid" in response to this, so before anyone jumps to that I am by no means claiming this scenario is likely - just taking OP at his request of "how the worst could work" and leaning into what "the absolute worst" could look like.
So, its 2027, and Trump is dropping hints to staff that he'd really rather stay in power. Maybe Don Jr is doing his best Hunter Biden impersonation, making a ton of money selling presidential access to foreign governments and corporations, maybe there's some family scandals that he's able to keep at bay as president but won't have the leverage for the moment he leaves office, maybe there are some bigly projects around the border that he just doesn't trust a successor to close the deal on - maybe, like all bigly men he doesn't trust any successor- maybe he just likes being the center of attention. But whatever the reason, he starts mentioning to staff that the two-term limit is bullshit and smart government just has to be more important anyway. Someone leaks it to the press, but when it hits the news smart people on boards like this assure everyone that it's paranoid liberal fantasy, Trump would never do that and at any rate he can't because lots of checks and balances from the administrative deep state to the courts would stop him.
Then, some plucky lawyer floats a memo Trump's way, along the lines of the Eastman Memo. https://edition.cnn.com/2021/09/21/politics/read-eastman-memo/index.html. The thrust of the memorandum is that although the text of the 22nd amendment clearly states that "[n]o person shall be elected to the office of the President more than twice," it makes no clear ruling, and there is no supreme court precedent, regarding whether the same person can remain President in cases of a *national emergency*. And this is something for which there is plenty of precedent from around the world, hasn't that darling of the Democrats, Volodymyr Zelenskyy, been governing in a perpetual state of emergency without elections since 2022?
The more "establishment" contingent in Trump's staff initially dismiss it, but initial polling for JD Vance is abysmal compared to various expected Democratic challengers. Not only that, but one or two of the more aggressive-minded of Trump's staff make sure the memo gets to him, and now he's pushing his team in private to get behind the memo. In public, he's making much more brazen declarations about various "emergencies" which might interfere with operation of a proper election. Once again the liberal talking heads, once again the very smart right-leaning commentators explaining how this is paranoid leftist fantasy. Trump derangement syndrome.
But it gives the Democratic candidate a boost in several key swing states. Now faced with a potential loss in upcoming elections, people within the White House begin asking whether they can afford *not* to try something like the National Emergency approach. After all, what will the left do once they have the levers? Last time Trump left office, they tried to prosecute him.
Trump announces suspension of elections for the 2028 cycle, because the number of illegal immigrant voters has made a proper count of votes impossible. Blue states object and attempt to hold elections anyway, but there's no republican candidate on the ballots there as the party falls in line behind Trump and refuses to nominate a candidate. Red states boycot the election. Purple states are a mixed bag - in cases where Democrats hold power, elections are held, sometimes with JD Vance on the Republican ticket, sometimes with Trump; in cases where Republicans are in more control, purple states are less likely to hold votes, or to declare that their Republican-majority legislatures will select their electoral slates, which is only responsible given that the President has declared an emergency for the precise reason that elections can't be trusted right now. As a result, slates of electors arrive at the Capitol in a state of fairly complete chaos. 160 electoral votes are absent entirely. The Democratic candidate has 268 votes, while the remaining 110 votes are split between Trump and Vance.
Democrats claim that they have a supermajority of the electoral votes present. Republicans claim that there was no fair election held - you can't just give the presidency out when nearly a third of states didn't vote at all, that disenfranchises tons of Americans, and besides, the number you need is 270, and the Democratic candidate did not hit that number. Democrats retort that of the 110 votes for Trump or Vance, nearly half were selected by Republican legislatures - a real *vote* in those states, some of which were purple to start, would have yielded more than enough to put the Democrat over the top.
When Congress convenes to count the electors, Vice President Vance declares the fiasco too much. There's been no valid vote, so the proper thing is to let the House of Representatives assign the presidency, in keeping with the constitution. He uses the phrase "assign the presidency" rather than "decide the election" intentionally. As stated in the memo, the fine print of the constitution is "[n]o person shall be elected to the office of the President more than twice." He says it will be up to the House to assign the presidency, then up to the Supreme Court to decide whether this assignment constitutes an "election" for purposes of the 22nd Amendment. "We are all following the constitution as best we can in these unprecedented times, so I urge those unhappy with this result to follow constitutional order and seek redress in the courts."
The Republican House dutifully assigns the Presidency to Trump. Democrats predictably bring a variety of lawsuits, and as predicted in both the Eastman and National Emergency Memo, SCOTUS declines the cases on the ground that the issue is a "non-justiciable political question." Welcome to your new autocracy. Of course, Trump's loyalists are extremely offended when someone calls it that. Some assert that there will be a new election as soon as the emergency ends, others claim it will happen in 4 years, but they are uniform in their confidence that everything happened according to the constitution; at most it was "lawfare," but nothing that justifies the hyperbolic claims of extreme leftists that it was a "coup" or something befitting a tin-pot dictatorship. And in any event, even if it *was* a coup, it's still Democrats' fault since their leadership prosecuted him last time he left office. From a certain point of view, he can't afford *not* to hold onto power in perpetuity.
So there's a sample fact pattern. It's incredibly low-likelihood, and I'm sure any Trump fan reading it is likely to declare it totally deranged fantasy. But then, I would have thought the fake electors plan to be deranged fantasy before Trump tried it, and I guess the key question is, "deranged as it is, is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?"
To my mind, that's the key question, and for all other candidates in my lifetime, I've been so comfortably assured that they wouldn't do something as crazy as this whole screed that I haven't had to factor it into my vote. But now I have a candidate where, instead of just a simple "no, they wouldn't do that" I have to go with "extremely unlikely he would do that - he's old, probably wouldn't want the presidency anyway, and there are lots of checks and balances in place that would block it, and..."
And once you find yourself rationalizing how unlikely it is that a guy would eat your baby, how many checks and balances are in place that would prevent him from eating your baby, how he probably wouldn't even want to do so, etc, etc... when you find yourself having that conversation about the guy, instead of just taking as an obvious given that he wouldn't eat a freaking baby because that would be insane, that's a situation in which you pick pretty much any other babysitter imaginable.
What is your opinion of the 2016 appeal for faithless electors? I'm not trying to wage Culture War, this is a genuine question. I thought it was ridiculous and it seems to have ended in, at worst, egg on some faces - the couple of electors who *did* go faithless dumped Hillary, not Trump.
On the other hand, it was an attempt to overturn the results of the election, calling on electors to pledge their votes in opposition to what the result in their state had been. So was that a coup attempt or not?
There seems to be forgetfulness about there having been a lot of "I refuse to accept the result of this election", "not my president", "he only won the Electoral College not the popular vote so he didn't really win". If people can call for the abolition of the Electoral College and replacement by the popular vote, I don't see why calling for the amendment of an amendment about term limits is off the table.
It seems to me that either we go by the literal legal regulations defining the Electoral College and it's function, in which case it seems like the electors are technically entitled to be faithless (at least from some states? correct me if I'm wrong) according to their personal judgement and conscience.
Or we say that whatever the technical rules are, the electors have a duty to faithfully serve what the people voted for based on democratic principle. But if it's the latter, I'm not sure why abstract democratic principle should bind them to the vote in their particular state over the overall national popular vote- the latter seems to me to be a truer expression of the "will of the people" if that's our standard.
Yes, I think either stick with "this is how we decided to select the president" *or* "okay let's abolish the Electoral College".
But you can't have it be "the Electoral College is right if it gives me the result I want but wrong it if gives me the result I don't want". If Harris wins by winning the Electoral College but not the popular vote, I fully expect none of the people crying over how the College is a remnant of slave-holding and should be done away with because it's racist to stick to that principle, even though their candidate won. That's what annoys me most: the claim that "we are doing this out of principle, they are doing it out of self-interest" when it's blatantly obvious that it's self-interest all the way down on both sides.
Supreme Court is ram-rodding through social liberalisation over the objections and stated preferences of many people? Too bad baby, them's the rules and now the New Thing is law and you have to obey the law!
Supreme Court is ram-rodding through social conservatism over the objections and stated preferences of many people? Oh no, it is a racist sexist non-representative body, we need term limits for judges, it should be overhauled! Resist!
Again, if we're sticking with the Electoral College as it's currently established, that includes the electors being actual humans who can exercise some judgement, not just abstracted votes bound to their states. I personally think that system isn't great and a national popular vote would be better, but I don't think it's wrong or hypocritical for people to try to use the current system as it exists to get the outcome they want even if they'd also prefer for the system to be different.
>I personally think that system isn't great and a national popular vote would be better
Personally, I agree. I would prefer a popular vote also.
I'm not thrilled with the idea of "faithless electors". On the other hand, my understanding of what the founders wanted in the first place was for the electors to actually use their minds in choosing which presidential candidate to vote for. There is a decent argument for doing that, but to _actually_ do that, the electors should be running local campaigns along the lines of "Vote for me. I'm great at evaluating job candidates!".
The Electoral College seems the worst of both worlds. We don't get "one citizen, one vote" as we would with a popular vote. We don't get electors using their brains (with presumably more knowledge of government and politics than the average voter) to grill and evaluate the candidates.
The Supreme Court weighed in on it in the wake of 2016; electors are required by law to vote the way the state wants them to, they're representatives of the state not entities unto themselves.
Good to know, although even that only says that states are allowed to punish electors for being faithless, not that electors are by default required to follow the candidate they were selected for.
You see that Colorado was allowed to replace rogue electors during the election, with people who would vote how the state wanted. So, the state can both nullify the rogue vote, and fine the offender.
Because the rules are up to each individual state and not every state legally requires electors to follow the popular vote. I think most states do require this, around 40 of them, but don't quote me on the exact number.
>>What is your opinion of the 2016 appeal for faithless electors? I'm not trying to wage Culture War, this is a genuine question.
No culture war taken, and no issue answering that - I did in the last thread. Cut/paste to save myself the rewrite, and you the lookup:
>I'm perfectly comfortable condemning Clinton's elector scheme in 2016. I don't think she's a person we want anywhere near power either. I was a GWB voter through the 2000s - the first democratic vote I ever cast was voting in the democratic primary in 2008, specifically to keep Clinton off the democratic ticket even though I fully intended to vote McCain in the general. If Clinton was running against Trump this year, I'd be right back where I was in 2016, voting libertarian because all options suck, this time with a dash of "all options suck and both have tried to subvert election results." But Harris is running, not Clinton, so I don't see any responsible use of a vote other than a vote for the candidate I'm confident will transfer power peacefully later over the candidate who as already demonstrated a refusal to do that.
I will supplement that, though, with the observation that what Trump tried is still demonstrably worse than the faithless electors scheme. To hit Trump levels, Hillary would have had to (a) do her faithless electors thing, then (b) when electors opted not to go for it, sic a mob on them to try to force them to heel. To be clear, she did (a), which puts her comfortably in the category of "do not let near power" with Trump, but it's still a faulty comparison, in addition to being the dumb kind of "it should be okay that I shot a baby since another politician shot one too" defense.
>>So was that a coup attempt or not?
I've avoided calling Jan 6 a coup and would do the same for the faithless electors scheme. Those conversations seem to me to be more of a distraction. People get all heated about whether it was "coup" or "I can't believe its not coup," but either way its an attempt at fraudulently putting the candidate who lost the election into the White House. To my mind that's disqualifying, and an argument over whether it's really a "coup" or not because it had a low % chance to succeed, or the military wasn't involved, or whatever is just a sideshow.
>>There seems to be forgetfulness about there having been a lot of "I refuse to accept the result of this election", "not my president", "he only won the Electoral College not the popular vote so he didn't really win".
Statements like that are just sour grapes from voters, and are in a totally different league from actually trying to defraud the electorate and change the election results. I have no problem with Americans of any stripe whining because their candidate lost an election. That's *everybody's* right.
>>If people can call for the abolition of the Electoral College and replacement by the popular vote, I don't see why calling for the amendment of an amendment about term limits is off the table.
Sure, but my fact pattern wasn't an amendment to term limits, it was a seizure of power. I don't think "encourage voters to demand a change to term limits, then follow the constitutional process for pursuing an amendment" is anywhere near the "worst Trump could do" as requested by OP.
This is what I wrote about the 2016 appeal for faithless electors at the time:
"In many ways, for the Electors to elect a President on their own discretion (as opposed to acting as mere proxies for pluralities of voters in their respective states) would be comparable to the Queen unilaterally firing the Prime Minister and appointing a replacement of her choosing: in both cases, they have the legal power to do so, and indeed would be acting consistently with the historical intent and purpose of their respective offices, but both cases would be an enormous break with long-standing practice, and both would be widely seen as democratically illegitimate barring truly exceptional circumstances."
>>Why do you describe your own scenario as deranged? What are the key points that require overcoming poor odds?
Combination of things. I think it came to mind because: (a) it's a very extreme and unlikely example, (b) a president would need to be deranged to try it, and (c) I do think a person who says "I believe with confidence that Trump will do this if he becomes president" is letting their fear lead them to dramatically overstate the odds. Something like this, to me, clocks in at "extremely unlikely but it's fundamentally disqualifying to Trump as a candidate that I perceive even small odds of it"
As far as hurdles go, coordination seems like the biggest one. As Mike Pence has shown, insiders can break the chain. SCOTUS, for example, could have its moment and choose not to go along. John Schilling also thinks there's some Article 2 limitations that make it illegal, but I'm not sure what they are - I've asked him for an explanation there so keep an eye on that thread. That still boils down to SCOTUS.
Yeah, this is paranoid. The Constitution has a fair bit to say about executive power in the event of e.g. an insurrection, and it is very specific about who gets to be president under various scenarios where the nominal election-winner can't do the job, and none of it adds up to POTUS getting an extension just because he declared an emergency. And it isn't necessary for every bit of law that touches on presidential succession, like the 22nd amendment, to include a "yes, we really mean it, even in an emergency" clause, for it to apply in an emergency.
We already know who gets to be president in any plausible scenario, including "Trump was elected fair and square in 2024 and it was physically impossible to hold an election in 2028". And it isn't Donald Trump. That's black-letter law, beyond Donald Trump's ability to change, and too much of the executive and judicial branches will abide by that law for Trump to remain effectively in power. If he tries, he'll fail as badly as he did on 1/6.
The Eastman strategy had at least a fig leaf of legality to it; the Constitution does call on Congress+VP to certify the election results and is silent about many of the details on that. But here you're talking about a scheme that is explicitly forbidden by the Constitution, and supposedly rationalized by something that isn't authorized.
>> We already know who gets to be president in any plausible scenario, including "Trump was elected fair and square in 2024 and it was physically impossible to hold an election in 2028".
Honest question, what would happen in a scenario where it were genuinely impossible to hold an election? Looking at Article 2 nothing jumps out at me as an obvious answer.
>>And it isn't Donald Trump. That's black-letter law, beyond Donald Trump's ability to change, and too much of the executive and judicial branches will abide by that law for Trump to remain effectively in power. If he tries, he'll fail as badly as he did on 1/6.
I realize I wrote a novel, but this goes so exactly to the last 3 paragraphs that I just have to repost below.
>So there's a sample fact pattern. It's incredibly low-likelihood, and I'm sure any Trump fan reading it is likely to declare it totally deranged fantasy. But then, I would have thought the fake electors plan to be deranged fantasy before Trump tried it, and I guess the key question is, "deranged as it is, is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?"
>To my mind, that's the key question, and for all other candidates in my lifetime, I've been so comfortably assured that they wouldn't do something as crazy as this whole screed that I haven't had to factor it into my vote. But now I have a candidate where, instead of just a simple "no, they wouldn't do that" I have to go with "extremely unlikely he would do that - he's old, probably wouldn't want the presidency anyway, and there are lots of checks and balances in place that would block it, and..."
>And once you find yourself rationalizing how unlikely it is that a guy would eat your baby, how many checks and balances are in place that would prevent him from eating your baby, how he probably wouldn't even want to do so, etc, etc... when you find yourself having that conversation about the guy, instead of just taking as an obvious given that he wouldn't eat a freaking baby because that would be insane, that's a situation in which you pick pretty much any other babysitter imaginable.
Trump takes this kind of thing from a simple “no” to an argument about all the not-Trump people and systems around him and how strong they are to stop him in the unlikely-but-implicitly-admittedly-uncomfortably-more-than-zero-chance-he-were-to-try. That we get to, “… and if he tried he would fail,” rather than stopping at “no” marks him, to me, as exactly the person you keep away from power wherever and whenever possible.
>Honest question, what would happen in a scenario where it were genuinely impossible to hold an election?<
I'm hard-pressed to think of a situation that could cause that. We had an active plague in 2020 and it didn't stop them.
Assuming some weird situation where timely elections are impossible but there are still enough people alive to sustain a functioning government, the state would probably have an election shortly after the deadline, and the late results would be approved by Congress and/or the Supreme Court as a one-time exception to the rule. The system wants the system to survive.
"Honest question, what would happen in a scenario where it were genuinely impossible to hold an election? Looking at Article 2 nothing jumps out at me as an obvious answer."
I think you have to look at the Amendments. If you're sitting in front of a decently hyperlinked version, the text of A2 will note that some of it is superseded by Amendment 12, and also Amendment 20, where section 3 says:
"If a President shall not have been chosen before the time fixed for the beginning of his term, or if the President elect shall have failed to qualify, then the Vice President elect shall act as President until a President shall have qualified; and the Congress may by law provide for the case wherein neither a President elect nor a Vice President elect shall have qualified, declaring who shall then act as President, or the manner in which one who is to act shall be selected, and such person shall act accordingly until a President or Vice President shall have qualified."
By this, neither Trump nor Vance would qualify; it's up to Congress to decide what to do, and it did that in 1947 with the Presidential Succession Act. The Speaker of the House would act as President, probably until a proper election finally happens.
Currently, that's Mike Johnson, but it might easily be someone else by Jan 2028. One could imagine a lot of anxiety over how long that Speaker would want to stay in the Oval Office and whether he or she would have influence over how long it takes to get that election going.
The other problem in that case is that if something physically kept the US from having an election, it probably also kept us from electing a new House as well - and thus, a new Speaker. So the *Speaker* would have a legitimacy problem. That would depend on other factors, among which is whether the House was expected to flip in 2028, or whether it didn't but the current incumbent was facing internal party challengers. At that point, we're in speculative history territory, although I feel comfortable claiming the chance of Trump ending up with the Speaker's seat being well under 0.0001%.
Makes sense - thanks for the citation. I agree that the scenario we're talking about is a well under .0001% scenario; part of why I used words like "paranoid" and "deranged."
I guess for me it just boils down to how, of all the reasons we keep citing for why stuff like this wouldn't happen, it's all things *external* to Trump himself. It's not that he *wouldn't* do something like this, it's that various reasons and structures *prevent* him from doing something like this.
Makes it seem less like we're choosing a responsible person to drive the car and more justifying picking a known reckless driver (who may or may not have intentionally run someone over once - it's debatable) who keeps talking about all the "sick traffic weave moves" he'll be pulling out there, on the promise that there's a speed regulator, guy riding shotgun, etc who will hold back his constant urges.
The way I'm trying to argue isn't intended to stipulate that Trump is a known reckless driver. Rather, it is to say that *even if* he's a reckless driver, the other safeties are going to hold.
If Congressional elections are also blocked, then there's no Speaker because all Represenatives' terms have expired. At least 2/3 of the Senate is left (possibly more, since many state governors have the power to make temporary appointments to fill vacancies), enough to form a quorum, so there is still a President Pro Tem of Senate or at least the ability to select one, and as John Schilling noted, the President Pro Tem is next in the statutory line of succession.
If there were no Senate either for some reason, the next in line would be the Cabinet Secretaries in order of department seniority (State, Treasury, Defense, Attorney General, Interior, etc). It's customary for the cabinet to offer resignations at the end of a Presidential term, especially when there's a new President, but the law and constitution allow them to remain in office until they quit or get fired. So the old Secretary of State would become Acting President if they're constitutionally eligible for the Presidency and don't resign at the end of the old president's term.
For some reason I was thinking that a Speaker election might not happen shortly after Nov 4 if the new House retained the same party majority (e.g., Nancy Pelosi leads the 116th Congress and the 117th is still Democrat and there are no serious Democrat contenders).
I imagine there'd be a Speaker-presumptive if the House was widely believed to not flip if elections had been held, but I agree the Senate President would be the proper call, in an improper situation.
If there's literally no election at all, then on January 21 the presidency goes to the president pro tempore of the rump Senate. Vice President and Speaker of the House get skipped because, if there's no election, there isn't a Veep or a Speaker. Each Congress is legally independent of the last, even if it's mostly the same guys in practice, and the last congress's term expires before 1/21 so if you haven't had an election yet then you don't have a House. But Senators serve staggered six-year terms, so you can skip an election and still have a quorum in the Senate if you need it.
“Is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?”
Trump would almost certainly not refuse based on the nature of the plan. Trump lacks any core principles other than self interest.
What could inhibit Trump is significant opposition from Trump’s own staff. The part of your scenario where Trump “starts mentioning to staff that the two-term limit is bullshit” captures the way Trump operates. If, after having been given of what Trump wants, a large portion of the staff still opposes the idea, Trump is going to be extremely reluctant to move forward with the plan. For example, Trump supported Jeffrey Clark’s plan to send out letters to states like Georgia that would falsely claim that the DOJ had “identified significant concerns that might have impacted the outcome of the election” and suggest that the state legislature chose the Presidential electors itself. When Jeff Rosen refused to go along, Trump could have fired him. Instead, Trump abandoned that part of the January 6 plot.
>>“Is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?”<<
>Trump would almost certainly not refuse based on the nature of the plan. Trump lacks any core principles other than self interest.<
>What could inhibit Trump is significant opposition from Trump’s own staff.<
That matches my own mental model for how Trump operates; I just can't get over the foolhardiness of the bargain, entrusting power to someone based on the mental model that they will be constantly trying to abuse it but restrained by others.
"Let's summon Skraldex, Chewer of Manflesh, from his Cthonic lair in the 8th Hell to be the next POTUS - he hates wokeness and the swamp as he hates all works of man, and will be incredibly aggressive and dangerous to our enemies once turned upon them. Granted, he cares for none but himself, constantly hungers for our succulent flesh, and would laughingly hurl us into the abyss if freed, but I'm confident the seals will hold."
I think almost nobody who talks like Trump is going to impose a dictatorship and round up the illegals, gays, minorities, liberals, etc., actually behaves in accordance with those beliefs. I mean, if you think there's a 50% chance Trump wins and a 50% chance, conditioned on him winning, that he sends your kid to a death camp for being gay, you should be either living outside the US or frantically trying to.
Or you need to declare yourself a refugee who fears political persecution, which from the rhetoric I'd expect many people (on both sides) to feel appropriate.
But even for the more traditional sort of visas, there are tens of millions of Americans who could get a visa to a relatively decent country if they applied, and I'm pretty sure I'd have heard if even tens of thousands were actually doing so for political reasons. So either everyone is waiting for the last minute, which given the delays in processing visa applications would be silly, or the "I'll have to flee the country if Trump is elected!" bit is 99.9% empty posturing.
Also, we can look at 2016 when, yep, it was 99.9% empty posturing.
Most people can't just leave the country. And if you think Trump risks being a Hitler-like figure, fleeing the country before the election makes this MORE likely, so there's a question of moral responsibility.
Most people don't live in swing states, so their vote could not possibly affect the election result. Even for others, if you believe there's a 50% chance of Hitler v2.0 taking power in the US, it's hard to imagine you'd stay here and wait to be sent to a camp rather than fleeing the country.
I've never lived in another country. I don't know what I would even have to do to get permanent residency - probably get a job there. Last time I had to get a job it took ~6 months, and that was in the US, which is one of the best economies in the world, and without needing to ask anybody for sponsorship. My whole family and all of my friends live in the US. Everywhere I've ever worked is here. All my money is in US banks. I'm very attached to this country, emotionally, in addition to financially, etc.
People have lots of reasons for not wanting to literally flee the country. Maybe it would be more calculated if I started shifting my assets and planning to move to Canada or something, but fundamentally there's a lot of factors there and I think at a bare minimum it's silly to imply that people don't actually believe Trump might be a fascist-dictator-wannabe because they aren't willing to uproot themselves completely from everything they've ever known. By that logic you might as well ask why everybody doesn't just move out of Palestine or Ukraine.
The number 1 impediment is the Administrative Procedure Act. Trump’s agencies can’t change the official Code of Federal Regulations without going through a complex, drawn-out rule making process. These agencies are largely staffed by DC-area permanent civil servants, who overwhelmingly lean Democrat and will attempt to slow-roll or sandbag the top-down orders they receive.
The number 2 impediment is that Trump’s ability to fire these permanent civil servants is restricted by the Civil Service Reform Acts.
Now, theoretically these laws could be repealed by a Republican congress, but at that point the objection becomes, “I don’t think the party that controls all three branches of government should be able to implement its preferred policy.”
You are the first person to actually address my question. Thank you. Can you add any more. For instance, regarding
< I don’t think the party that controls all three branches of government should be able to implement its preferred policy.
Are there any legal or practical things that do act as a check on this happening, though?
I am really dismayed at how people responding to my post are just sliding into general impression kinds of stuff: Democrats hyperventilated the first time he was elected and nothing too awful happened, people who talk like Trump don't really carry out the ideas they're promoting during a campaign, the Democrats have been doing all the bad shit they are now afraid Trump is going to do, Trump actually has respect for democracy, a vague "others in government would restrain him."
This is the worst I've ever felt about ACX. It sounds as dumb as every other forum right now. Nobody but Daniel has even tried to address my questions: If Trump is elected and promptly installs many loyalists, what checks exist on his doing (1) the things Democrats are most strongly opposed to, which I could list but I think are common knowledge, (2) nutty things like nuking hurricanes, which I assume everyone is opposed to and (3) dismantling the election process, so that he remains in power until he chooses a successor or is overthrown.
Regarding the first point, it's not clear to me what keeps the administration from ignoring the law. Sure, they will get taken to court, any court, even the USSC, will say they can't do this because of the law, but there is no one to enforce the law. Well, the the Justice Department, but I don't expect them to start prosecuting political appointees.
Regarding the Civil Service Reforms Act, Trump circulated an order in October 2020 to strip protections from tens of thousands of permanent civil servants. This is plausible because the law exempts civil servants that the administration decides are political. This order (schedule F) has been mentioned openly during the current campaign as a way to drain the swamp - note that it also solves the first impediment, because all those agencies can be repopulated by Trump appointees.
I think the biggest concern really is the one mentioned above though. That first, the USSC has pretty much said anything Trump does is legal (he's immune from prosecution), and if the administration is breaking federal laws there is not going to be an independent Justice department to prosecute those people. If Trump decides to ignore the Posse Comitatus Act and call out the troops against civilians, he'll be breaking the law, and the courts might agree, but so what? It will be up to Congress to impeach him, and we've seen that the Senate will not do that. According to Mitt Romney, many GOP senators voted against because they specifically feared for their safety or that of their families. So to answer the original poster's question, I don't see that there really are good guardrails.
If Trump arbitrarily and illegally sends a "you're fired" notice to Alice the Democratic Civil Servant, and writes a letter telling Bob the Trumpist Stooge that he's got a nice cushy job as a civil servant, I'm pretty sure Alice still gets her paycheck every week and Bob doesn't. Because that's what the law says, because that's what the payroll computer is programmed to do, and because even if the civil service job that Bob was told to take up is the "send out paychecks to all the civil servants" job, Alice isn't going to tell him the password to the computer.
I suppose with enough effort, Trump and an army of Bobs could bollix things to the point where *nobody* gets paid.
And the bit about how there's nobody to enforce the law against the Trump administration in this scenario, obscures the key point: there's nobody to enforce the law against *anybody* in this scenario, because you're postulating Trump destroying the administrative state. But the only power the Presidency has, is to enforce the law through the machinery of the administrative state. So, yes, if Trump really works at it, Trump might be able to make the office he went through so much trouble to win, into something impotent.
If the idea is that Donald Trump is then going to build a whole new administrative state from scratch, by sending out legally invalid letters to a bunch of cronies saying "Congratulations on your new post in the Trumpian administrative state!", then that's not going to work. If it *could* work, then Trump could just do that now, the legal power of the Presidency isn't at work in this hypothetical, and if that were possible I'm pretty sure someone more competent would have done it long ago. Nixon, perhaps.
"The legal basis for the Schedule F appointment was a section of the Civil Service Reform Act of 1978 (5 U.S.C. § 7511(b)(2)), which exempts from civil service protections federal employees "whose position has been determined to be of a confidential, policy-determining, policy-making or policy-advocating character". I encourage you to read the article.
As for there being no one to enforce the law against the executive branch, I think it's a big stretch from saying "the Justice department declined to bring charges against the chief of staff" to postulating the administrative state has been destroyed. All you need is an Attorney General who is a bit pliable. Or not even that, just a President who assures everyone he will pardon them of any crimes they commit on his part. The administrative state will muddle on just fine in those scenarios.
I think deporting millions of people is within his power, given that said individuals are breaking the law, and the president is in charge of the branch of government responsible for enforcing the law, AND ALSO he is being voted in in large part for the purpose of enforcing exactly this set of laws. As the Supreme Court previously noted, elections have consequences.
While I'm in favor of open borders, I do not necessarily see this as a problem with his administration. The problem there doesn't lay with Trump being elected. Maximally uncharitably, it belongs to the people who are voting him into office for that purpose - but that is maximally uncharitable, and I do see their point. I don't agree the problem is with the immigrants, but rather with the network of regulations and incentives we have spent more than a century building up, and if I were to blame anybody, it would be FDR, although even there we've had nearly a century to fix things so even then the blame is pretty attenuated.
But the long and the short of it is - he definitely has the power to mass-deport people, and while local areas being uncooperative can make this less effective, "less effective" is about the most they can achieve. I don't expect local governments to be particularly uncooperative, however; it only takes a tiny fraction of the immigrants arriving in the border states to turn local governments, whose policies are not designed with illegal immigration in mind, against immigrants. (See the cities declaring emergencies after receiving a few thousand immigrants because their policies turn these people into massive budgetary problems for the cities.)
That's a key thing, here: The opposition to deporting people tends to evaporate, or at least weaken considerably, once illegal immigrants actually start showing up. Attitudes towards illegal immigration depend, at a statistical level, entirely on the number of local illegal immigrants, and Abbot's innovation of sending people where they want to go has been absurdly effective, even as it has been a tiny fraction of the illegal immigrants showing up in, for example, Texas.
Are there better policies we should be pursuing? Definitely. My personal opinion is that we desperately needed to boost immigration a decade ago, and that the mass-retirement of the boomers is a major contributor to the inflation we've dealt with for the past four years. Best time to plant a tree is yesterday, second best time is today. But that's its own topic, and "deporting illegal immigrants" is something we can actually do, whereas reorganizing society to integrate immigrants and align incentives is, alas, something I -don't- think we can do.
Given how fucked society is in general, and how unlikely it is that Democrats or Republicans will be willing to fix the underlying issues (it gores sacred oxen all over the place), treating the symptoms and deporting millions of people might strictly be the best we can hope for. Which is fucking pathetic, but, well, that's where I think we are, and I think a lot of people are aware of that.
So, all of that is to say - I don't think the Democrats can or even are willing to seriously do anything to stop mass deportation. See Martha's Vineyard. Just takes one person upset that an illegal immigrant got public housing assistance before they did to get the attention of the leviathan.
(Now, whether or not Trump himself is actually willing to deport millions of people is another question entirely; I think he'll make a token effort, enough to satisfy his base that Something Is Being Done, and little more on that front; I expect most of his effort will be devoted to preventing more immigrants from getting in, and deporting anybody foolish or unlucky enough to get caught up in the justice system.)
These aren't impediments, granted, just a discussion of the topic before we get into impediments.
Major impediments to Trump's agenda:
Federal unions will fight reductions in their workforce tooth and nail, and likely tie everything up in lawsuits that will take longer than four years to fully resolve.
Regular federal employees, well-versed in the art of malicious compliance, will engage in full force to fight anything they don't want to do.
Federal judges repeatedly stymied Trump during his first term, and will likely do so again.
The current Supreme Court has shown a strong interest in dismantling the powers Trump needs to enact most plausible agendas.
Trump will likely be looking to secure a better legacy, which means he'll likely prioritize popular causes over unpopular causes; additionally, to the extent he's genuinely personally interested in causes he wants furthered, he needs to make Vance a plausible candidate in the next election to continue those causes, which means he can't piss people off too much.
Congress is unlikely to play ball; they will want credit for the popular things, and want to distance themselves from the unpopular things. They're interested in maximizing their reeleection chances, not in putting feathers in Trump's hat at their own expense.
The military isn't going to go all-in for Trump. They may support him, but this support will extend only as far as the self interest of the individual parties (including their political/moral beliefs).
Trump is in many respects his own worst enemy; he'll stymie himself.
I agree with your edit and I'm glad you're asking the question. I admire so much about the way ACX subscribers think, but Trump has seemed to be a big blind spot for the normal fact-finding abilities of the rationalists here. The question here is very clear and almost no one is engaging with it.
But it's not one of the things she's explicitly said she will do, whereas the Trump admin has specifically stated a desire to do many of those things. There has to be some impetus to begin with.
It would be more like asking "what barriers are there in place to keep Kamala Harris from increasing child tax credits, the thing she has said she wants to do"... not "some random other action she has never said she wants to do".
Specifically, re: the illegal immigrants part, Trump's "chief immigration advisor, Santa Monica-born Stephen Miller, has spelled out what that would mean: Trump would assemble “a giant force” including National Guard troops to seize undocumented migrants, transport them to camps in Texas and expel them." - and if you want to go down the rest of the list and look up every claim, there's plenty to back it. You'll find word for word quotes or video interviews.
Sure, we can say "politicians say lots of things they don't follow through with" but I don't think it's sensible to respond to "will the politician do the thing they say they are going to do" with "why on earth would you even think that?"
I'm not sure what sources you would find most valid but if you look up "trump glen beck lock up opponents" you will find the interview where Beck says: “Do you regret not locking [Clinton] up? And if you’re president again, will you lock people up?” and Trump said: “The answer is you have no choice, because they’re doing it to us.” and he has also stated “I will appoint a real special prosecutor to go after the most corrupt president in the history of the United States of America, Joe Biden, and the entire Biden crime family,”
In all honesty, I think like all politicians he says things he has no way of setting into motion. As to whether these are credible statements or not - I likely agree with you in that I think they are not. However, I think it's reasonable to listen to that and for OP to ask "so what's stopping him from doing those things he says he will do?"
I mean, if Kamala Harris said "we're going to open all the borders to everyone", even though to me that is a supremely unlikely and "incredible" statement, would you really be saying "well, I don't see why we should talk about whether or not that's possible, there's no reason to think it's credible"? It's not outrageous to check whether politicians *can* do the things they say they are going to do even before they start to do them - regardless of if it's in the spirit of agreeing or disagreeing with the action.
Well, Curtis Yarvin seems to think that the bureaucracy will stop Trump/Musk from doing much of anything. And keep in mind that Yarvin would probably *like* radical change, so I don't think he's saying this as some form of wishful thinking. You can read Yarvin's full thoughts here: https://graymirror.substack.com/p/elon-versus-the-usg
You don't need to deport them, you need to get them to deport themselves. You do this by making the penalty for being caught in the country illegally a lot worse than the threat of being deported.
My question is about practical obstacles to doing this. For instance, what penalty could be used? If penalty is a fine, what is the penalty for not paying the fine? Seems unlikely that most will depart pronto, especially if penalty is a fine. Seems like not paying the fine won't make their situation any worse than it is already, and meanwhile they get to keep going to their under the table jobs, seeing local friends and family etc. Prison? For millions? Some kind of semi-confined housing? For millions? Also, problem of many businesses really relying on the labor of these people. Businesses would protest, push back and prob would not cooperate, plus if they do lose the workers various things managed by these businesses are going to be in short supply or more expensive, ripple effect etc. I'm not saying it's impossible, just that it seems pretty hard to figure out how to pull this off.
As I understand it, the primary thing preventing this is what the right would call 'The Deeps State', ie career bureaucrats who care about the country and their role in it and will continue to do their jobs properly and fairly no matter what orders they get from on high.
Obviously this is a biased source, but the book 'Toddler in Chief' collects literal hundreds of anecdotes and statements from people who worked directly under Trump during his first term, talking about how they would routinely distract and deceive him in order to ignore the insane things he wanted to do and do normal sensible things instead. Despite the book being on an obvious side, the quotes do not seem to be generally taken out of context or misrepresented, and many top officials form the first administration have gone on the record in longer interviews describing this state of affairs.
The main reason to be worried that a second term would be unlike the first is that the Trump organization and the increasingly-MAGAfied GOP has widely and loudly denounced this state of affairs, and vowed to have massive purges and loyalty tests to make sure that anyone in a new administration would do exactly what he tells them to every time. JD Vance was chosen largely on the strength of statements he's made to this effect, and Project 2025 is seen as basically a portfolio attachment to the resume of prominent right-wing actors who are pledging to do this and want a place in the new administration.
I think it's pretty credible that a second term would be very different and much worse/much more volatile because of this change. But obviously I also am taking a side here.
My understanding of the American political system is that presidents are quite heavily limited. So if you are talking about the LEGAL system of government Trump can't do that much to undermine democracy alone.
This limit means he needs either a) extensive party support, and republican house and senate) to legally enact change or b) HUGE amounts of popular support + coordination to execute some kind of coup. My understanding is that Trump has neither great dictatoral party support, or popular support, and he certainly seems to lack military support.
That all applies to him being a dictator. If you are scared of him deporting ILLEGAL immigrants. I don't know what to tell you. In a democracy, people can vote for policies you don't like - and the democratic thing is to accept that. You should make your case, argue, accept the winning position, and then if you lost, do a better campaign next time.
- The magic only gets them there, it doesn't help them govern. (So otherwise impressive folk with no political experience should have some reason they won't pull a Herbert Hoover)
I'm considering casting my vote for a write-in candidate on the theory that one should try to vote for good things.
It would seem remiss to waste my magical powers on anyone except myself. (I am not actually eligible but I'm just gonna go ahead and magic myself up a Hawaiian birth certificate while I'm at it.)
Obviously, if we can magic solutions up, the ideal politics is:
"Clone Lee Kuan Yew and install him as dictator-for-life in all ~200 countries."
LKY took a country that started poorer than the Philippines, and in ~30-40 years of autocratic leadership, brought it to fully developed, First World status. And he did it with a lot of ethnic diversity, having zero natural resources, and with incipient civil war and external takeover looming most of that time. It's the least corrupt country in Asia, and the most beautiful, least polluted, and well run.
He's the only provably successful "enlightened tyrant" that we can both verify his accomplishments and actually get DNA for - so LKY for World Presidents!
I think it would be relatively hard to generalize the governance of a large city, to governance of a large country.
In Singapore you can plausibly have your hands on every lever of regulation and power, and a single good strategy can bring significant wealth to the whole city. The same is not true for the United States.
I think in this context we have to consider Singapore's location as a pretty valuable natural resource, although his accomplishments are still pretty impressive.
Singapore reminded me strongly of the United States of the 1990s, including some of the social fissures that would later break apart here. I'd hesitate to say we should definitely replicate it until we see what happens in thirty years.
But arguably, Singapore has had major social divisions (Malay working and under class, Chinese middle and upper, Westerner upper) since the very beginning, and LKY was better than most at keeping everything aligned and progressing even with those divisions. And they were pretty serious, leading to riots and tense situations fairly frequently in his tenure.
Unless you're speaking of other social fissures? In which case, I'm interested in hearing about them.
For an example, social welfare as relates specifically to disabled people; the television programs while I was there were -extremely- careful to simultaneously insist that the way disabled people were treated wasn't okay (seriously, television there felt right out of the 90s in the US, complete with a Very Special Episode about vaping on a daily sitcom), and to showcase ways in which disabled people were being integrated into society - while also going to great lengths to make clear that No Government Money was going towards supporting disabled people.
You don't need to make it clear government money isn't being spent on an issue unless there are people agitating for it to be spent in this way. It seems there is something deeper going on there.
There were other little things. Another show, this one about prison inmates, implied, while very carefully not saying anything to this effect, that the drug punishment laws are too harsh. And a taxi driver claimed all the trees had GPS units installed in them so they could catch people who stole trees (???). A handwritten sign above a potted plant saying it was on camera and anybody who littered in the plant's pot would be prosecuted.
Little things, which in a society where I watched somebody leave an unlocked, signed-in Apple laptop sitting on a table on the side of the street (unthinkable!), suggested that social trust was beginning to fray.
On the whole, mind, I found it quite a nice little society; I'd recommend anybody visit just to see what a well (if, in my opinion, over) governed society looks like. But I'd hold off on assuming that it won't experience the same problems showing up elsewhere in the world, because it feels less like it has found a way to avoid them, so much as that it feels like they're still simmering under the surface.
Arguably, the whole "panopticon" thing is WHY Singapore is so clean and well run and non-corrupt, especially compared to all the rest of SE Asia. I think it's actually a pretty good exhibit on the level of strictness and authoritarianism you need to have a clean and orderly society with a lot of ethnic diversity *and* a very high Gini index. Most countries so clean and crime free (Japan, Scandinavian countries) get by with a lot of ethnic homogeneity and low inequality, but Singapore is proof you CAN have it, it just takes a lot of effort and a panopticon society.
I agree on the overgoverned point, btw. But it's a tradeoff too - if they didn't overgovern so much, they'd be a lot more like Ho Chi Minh or Manila.
I think your disabled point and drug severity point is just regular politics, but agree there is a meaningful split in opinion.
But yes - anyone interested in what an incredibly well run (if overpoliced) society looks like, definitely check out Singapore!
Of prominent American political figures: probably Jared Polis. There are some people I can imagine doing better*, but they're bigger risks and I think someone with a good track record as a major state governor is safer.
*Sam Altman maybe? He seems unusually good at running effective organizations and also good at political maneuvering, and while he's less trustworthy and less worried about AI risk than I'd like for an AI guy he's still better about both than the average politician.
Troll answer: Along Levy (get us some real cost-effective train systems in place). Or that guy who built the entire Madrid metro in four years. Or Deng Xiaoping (use a time machine to get him during his exile period).
The capacity of the Spanish State to deliver infrastructure is under estimated and goes against a Northern European bias I used to have. I feel it’s better than modern Germany and far better than the U.K.
It hasn’t yet helped with economic growth, although these things matter in the long run.
Yes. Alon Levy has also noted that while their ability to build infrastructure is phenomenal, their actual operations (and maybe integration?) are weaker, leading to the Spanish AVE never really getting the kind of ridership the french TGV gets despite the more extensive network.
Sam Altman might be my strongest *anti*-pick—he is already extremely dangerous with the amount of power he has. He would be even more dangerous as president. Sam Altman has so many terrible qualities:
- is trying to build an AI that might kill everyone
- repeatedly lied to his board and then fired them when they tried to fire him
- repeatedly lied about how OpenAI's structure was going to keep it in check (first the independent board who got fired and replaced, then the Superalignment team that got disbanded, then the non-profit parent that got disbanded even though that was supposed to be legally impossible)
I agree that he's unusually good at running organizations effectively, but he runs them toward bad ends. That makes him an exceptionally *bad* candidate, not an exceptionally good one.
So my point is that level of cynical power grabbing politics is just the baseline I expect from elite level politicians (and if we picked a president who didn't have that he'd get played by politicians who do anyway), so it's not a minus there relative to the baseline.
(Plus, it would have the benefit of distracting him from OpenAI)
I've been hearing all kinds of "Elon is a Nazi" talk on Twitter, and not taken it seriously, but if he's installed a "block all Jews" button that you can click...
Iiuc there's a lot of custom mass block lists going around you can use for twitter (e.g. if you want to block all democrats/republicans/musicians/whatever). They're third party tools, not muskware.
> Paul graham, the guy who blocked all Jews on his Twitter
What are you basing this on? Googling gets me one guy saying 'PG has blocked me and every Jew I know' and implying that Graham is blocking people solely on the basis that they're Jewish. But that guy's feed seems to be ~100% pro-Israel tweets, and it seems vastly more likely Graham is blocking people based on their stance on the war than based on their race or religion.
Un-polemic, willing to acknowledge and attempt to fix the other side's problems. Supports the correct foreign policy, understanding the immense role of US in world order, but still cautious and dovish, more so than neoliberals. Extremely smart, hard worker, MPhil international relations and JD (law), important for domestic and international aspects of Presidency. Already very experienced in the gritty details of the top echelons of government. A true political technocrat.
Former speaker of the Texas statehouse Joe Straus, or former Texas Parks and Wildlife director Carter Smith (the latter an appointed position, not an elected one, but it is a political job in that one must go begging the Legislature all the time, and stopping the bubba libertarians trying to "sell all the parks" etc.).
Idiosyncratic and obscure but I have my reasons: chiefly because they represent a vanished world where normal people enter state politics and the state party machinery then elevates those people to a higher level, which machinery appears to have completely broken down in many states so that normal, intelligent people with some ambition need not apply.
On the premise that they need a decent amount of experience in politics/government and I'm just looking for competent, well-intentioned leadership, maybe Ron Wyden? Although he's on the older side as well.
Anthony Blinken? He seems like a very bright, very thoughtful, quite chill guy who knows a lot about a lot, prioritizes his family, and has a sense of how both the US government and the world work and don't work.
A key thing in his favor is that in four years as Secretary of State he has not been faulted much by either side, which seems like magic in itself. Since I know nothing about his thoughts on domestic policy, I get to infer them.
It seems like you're asking because you want to actually vote for someone who isn't Trump or Harris, and in that case I'd suggest that voting for one of the other candidates on the ballot is a better idea than voting for a write-in candidate. Given that the chance of the vote actually making a difference in the election is a rounding error either way, the expectation is that at least you might hear "x,000,000 people voted for the Libertarian Party nominees" in the news, but you probably won't hear "x0 people voted for Scott Alexander". Given that, Chase Oliver would be my top choice, followed by Jill Stein.
I'm aware that this is (close to a form of) strategic voting. I dislike that but it's a necessary evil in any electoral system, but especially the US system.
If I don't have time to go do a ton of research then off the top of my head I would say Penn Jillette. Unlike any career politician I know of, I trust him to be intellectually humble and seek out true expert counsel, and to be skeptical and careful about what to believe from his advisors. I also trust his ultimate values and goals as much as any other public figure I can immediately think of.
Along that vein, Kyrsten Sinema has fairly consistently ranked at or near “most moderate” in both the House and Senate. She went independent after falling out with the Dem caucus and is not currently running for re-election.
Pair her with Romney as an elder statesman and you’ve got a centrist ticket that has declared they are done with this shit in both directions. Has a certain charm.
If Romney had run in 2016, he'd have just further split the pool of primary votes than went to Bush III, Rubio, Cruz, Kaisich, and I think I'm forgetting someone but not matter - that split gives Trump the win with or without Romney in play.
I'd say go with the zeitgeist and pick a sequel of some kind.
Alexander Hamilton has living descendants. Jimmy Carter could theoretically serve a mythical second term. There are yet more Kennedys you could pick, even if you don't want RFK Jr.
Since every movie and book and TV show is a sequel, why not bring it to the Presidency?
If it weren't for the apocalyptic AI threat looming over all of politics: Probably Robin Hanson, despite his rather eccentric views on many things. Having proper incentives makes a lot of things easier, and he's really good at figuring those things out. Markets are smart.
(Given that AI issues do actually far outweigh all other issues: Anyone willing to dump every scrap of their political capital into halting AI globally and indefinitely, plus has sufficient skills in politics/communications/PR/diplomacy/etc so that they'd have the necessary influence in the first place. But that's less fun to go into.)
In the process of reading the Iliad right now. It’s amazing but I can’t help feeling there’s so much here I‘m missing that a more trained eye could show me. Barring just taking a class on it, anyone have recommendations for reading the Iliad and other classics and getting the most out of them?
I'd personally start with Dan Simmons' Ilium (and maybe the sequel), which is a very readable and high fidelity take on the Iliad embedded in another sci-fi story. After you have that high level view of the various motivations and characters as translated readably from a modern context storyteller, then go back to the original source, ideally with some ancillary texts others will recommend.
Uhm, this is a great book that I admire a lot. But I doubt very much that it helps with reading the original Iliad.
I would rather recommend the opposite: *After* reading the Iliad, consider reading Dan Simmon's Ilium, because that is a great Sci-Fi novel whose humor you can only appreciate if you know the Iliad well. Basically it's a fan-fiction to the Iliad, and just like HPMoR it loses a lot if you don't know the original story.
I actually did it this way though - tried to read the original Iliad, read through it but largely bounced off of it mentally, and then later read Ilium.
Ilium contextualized and grounded everything in a way that the original Iliad doesn't do at all, the Iliad was written from a really different culture and mindset, and depending on the translation, the archaic language can be a hangup too.
When I reread it after reading Ilium, it made a lot more sense to me. But that could definitely just be me, maybe it's bad advice generally.
The Emily Wilson translation published last year is snappy, expressed in iambic pentameter, and uses modern language, if the Gutenberg versions are too crusty.
Less high-fidelity I’m sure, but “Song of Achilles,” the story of the Trojan War from the perspective of Patroclus, is one of my favorite books of fiction. I thought it was fun, beautiful, touching, and meaningful.
I think the fastest way is probably to listen to a course that goes chapter by chapter.
I'm very frustrated (though I suppose it's only fitting) how much knowledge of epic literature is trapped away either in oral form or old dusty books. Not because that makes it off limits to me but because I feel like there is a real appetite for deep reading of books that reward that kind of thing.
Depends on what exactly you want to get. If it is about the story, then I can warmly recommend the retelling in "Gods and Heroes of Ancient Greece" by Gustav Schwab. It doesn't change the content, but in terms of language it meets halfway between the original and modern language. So you still get a flavor of the archaic language, but it is much more readable. It covers the Iliad along with many other ancient tales.
But of course, you will not get to see the original verses this way.
Joel Christensen's substack (all about the Iliad) is surpassingly good. Would link but linking in substack is as difficult as everything else about substack.
Find a translation that clicks for you, prose or poetry, and try to imagine what is happening in all its gravity. I know that this sounds trivial, "like duh, that's what reading is," but Homer narrates action and dialogue, not what happens in the characters' heads. You have to infer a lot about what the characters think and feel from by synthesizing what they say and do with your knowledge of their arc.
The scene in Book 6 with Hector, Andromache, and Astyanax on the wall moved me to near-tears when I looked through the characters' eyes instead of focusing on the description of the event. It's not suspenseful - you already know how the story ends - but try to feel what they must be feeling in that moment and you sense the crushing weight of it all: choosing between safety and duty; family and city; and the burden of leadership held by the shoulders of heroes who are more than mere men.
There’s free online courses on YouTube, often by world renowned professors teacher at Yale, Harvard or another impressive university.
Besides that I’d recommend to read through many of the Greek Classics. Iliad, Odyssey, 3 Tragedians, Aristophanes, then circle back to the Iliad with a different (maybe more difficult) translation. Maybe Herodotus, Thucydides, Xenophon’s Hellenica too.
More context on the Greek culture will give you a much stronger basis to understand the Iliad. Much information is assumed, many stories are told in tragedy that would have been basically common knowledge for anyone listening to the Iliad. It’s impossible to get full context, but I think it’s better to get it from other Classical Greeks than even the best written introduction of best taught course.
Your milage might vary - I find the The Great Courses match my desire for infotainment quite nicely when I'm curious about a particular academic topic but coming in pretty cold (for me at least, it's almost always about the right level of introduction).
I typically just listen to them when I'm in the car by myself, if i just listen while I'm going about my life, I find that I knock a lot of them out pretty easily.
Again, you might be looking for something different.
Try HomericEpicPodcast.substack.com, and The Homeric Epic Podcast. He's up to Book 9 and it's a banger for history nerds and use-of-language nerds and lots more.
Emily Wilson's translation of the Iliad has an excellent preface and translator's note that really helps bring out some details and themes.
This is what I used when reading the Iliad to my 4yo and he loved it.
Occasionally heavy handed (to the point of being incorrect) but overall excellent and probably as close as a translation gets to 'reading with an expert pointing out the good stuff'.
Unfortunately, Wilson's Odyssey is much worse than her Iliad.
Joen Christensen also has a series on notable passages you may enjoy:
I’m using the Wilson translation actually. I know it’s taken a fair amount of flak on here but I think it’s actually a really good translation that makes the story quite accessible to modern readers, and I think most of the sections people point out as examples of why it’s a “woke” translation or whatever are just thin points in the translation that aren’t representative of the effort as a whole. I also agree; her notes and introduction are excellent. Sad to hear her translation of the Odyssey isn’t up to par.
She started a substack by the way which is an enjoyable look at the inner workings of translating.
Her Odyssey is ...fine. But the notes and introduction are less extensive and illuminating, the meter starts off stronger than the Iliad's average but lacks drive for long portions, and overall it just seemed flatter to me.
It makes an enormous amount of things in the US more expensive than they should be in an effort to prop up various interest groups, like ship builders, that mostly no longer even exist.
It's like the debt ceiling, a pure deadweight loss law that no other country does. It's not really that big of a deal as the US isn't short on money.
Maybe if he committed to that a few months ago. The existing ferry company would then donate to whoever supported the status quo. There isn't really much money in it for Trump for him want to do that.
Maybe it would work better for a Democrat appealing to Alaska over the issue, although I'm not sure how big of a deal the Jones Act is there.
Oh the other hand, maybe the winning strategy is to commit to repealing the act *right before* the election is held. With the right timing, the ship building companies wouldn't have time to significantly contribute to the opposing side, but Hawaiians would change their vote.
I would like to see you soon on the front lines in the fight against the evil yoke of the Jones act. Let's free the American spirit from its shackles. Thank you.
It sounds like an antiquated piece of legislation that's profiting some groups (US owned shipping companies) at the expense of others (Puerto Ricans and Hawaiians). Maybe it should be repealed.
I still don't see why it's particularly important. If it's repealed what's the net impact? A few million people save a few hundred dollars per annum? This doesn't outweigh the urgency of the climate emergency for me, for example.
It's a law that stops a certain market operating efficiently - why do you think it's evil? And the American spirit, whatever that is, doesn't really seem to be enslaved by shipping law. What's the moral basis for your argument here?
Well I may have been tad exaggerating to rally support for the issue.
But I guess the potency of repealing the Jones Act for me is that it is unambiguously bad. So everyone should be able to rally around its repeal. We should be able to come together as one Comment Section.
I mean perhaps it also relates to the climate emergency as if we repeal the Jones act, cargo ships will be more efficient and hence reduce CO2 emissions.
I am hoping once the Jones Act is repealed, other similarly unjust legislation will be on the chopping block. In that sense, it's the defining issue of our generation for me.
Again I hope to see you on the front lines in the struggle against the Jones Act. Thank you for your interest.
Thanks for explaining - I'm a little more bought-in to your argument now.
Unfortunately I'm neither a US citizen nor based in the US. Despite what Fox News might have you believe NEITHER side has offered to fly me in to vote.
Hey, I work at a thinktank (Balsa Research) that's currently focused on repealing or significantly amending the Jones Act.
The Jones Act requires all goods shipped between U.S. ports to use vessels that are U.S.-crewed, -owned, and -registered. As of January 2023, the entire Jones Act fleet consists of just 56 tankers and 37 cargo ships (1), mostly serving Mainland <> Hawaii, Alaska, and Puerto Rico routes. Given the U.S.'s limited shipbuilding capacity, these numbers are unlikely to change significantly.
This single law has effectively eliminated domestic coastal shipping in the United States (again, there exists all of 37 domestic cargo ships! for the entire country!). While we do have significant barge traffic (2), especially on inland waterways, most domestic freight that could be moved by modern cargo ships is instead transported by rail or truck - methods that are OOMs more expensive and carbon-intensive.
It causes a bunch of other specific harms that I can get into (sluggish WTIV production that will cause us to fail our offshore wind targets; actual safety concerns for sailors; worsened response capacity in wake of environmental disasters such as hurricanes), but all of that pales in comparison to the fact that, again, it's an act that has rendered domestic coastal shipping by cargo ships effectively illegal in a country with two entire coasts.
Balsa Research will have a more comprehensive analysis of the Jones Act available by year's end - it will be announced on Zvi's blog :>
Would it be reasonable to think that if domestic coastal shipping was rejuvenated, that some rail lines could be repurposed to passenger transport? That would be nice.
Isn't rail shipping more efficient than boat shipping though?
Wikipedia claims they're approximately equal: "On average, a gallon of fuel allows one ton of cargo to be shipped 180–240 mi (290–390 km) by truck (e.g. @ 6–8 mpg‑US (2.6–3.4 km/L) 30 ton load, 450 mi (720 km) by railway, and 514 mi (827 km) by barge. "
I wasn't able to find the source for their numbers, though
Thank you for your extremely important service. I will be trying to follow Balsa Research. I am glad I am not alone in my crusade against the corrupt and vile Jones Act.
I think the Jones Act was a stupid idea that is causing a good deal of harm to no good end, but "most important issue of our lifetime" is either completely bonkers, or in need of an extraordinary justification that you haven't provided. And making the unsupported hyperbolic claim, makes me much less likely to take your position seriously or follow your links or whatever.
What kind of anthropic reasoning arguments are valid, which aren't, and why? And how sure are you/our best experts(?) about this? Are there good reading/watching materials online on the subject?
And for a more specific and concrete doubt, take the doomsday argument:
1. There have been ~100 billion people.
2. Suppose you are exactly the 100 billionth person to be born. Assuming 1, this is approximately the case.
3. You are way more likely to have been the 100 billionth person if there was only ever going to be 100 billion people in history, then if there was going to be 500 billion people in history.
4. By Bayes, this means it's more likely that you will be the last person to ever be born then that there will be 400 billion (or any number that matter) more people after you.
This *seems like* evidence that the end of Humanity is nigh. I've read (including from Scott) that this is counteracted by the principle that you are more likely to be born at all (in any order) in worlds where there will be more people throughout history -- which indicates that the end is not nigh.
Both the concept that "you are more likely to be chronologically late" and "you are more likely to exist in timelines where there are lots of people like you" seem to make sense, but I also have a vague but strong feeling that there is some sort of double-counting of arguments happening here. I'm unsure if we can use one to counter the other and end up with a uniform distribution, even if the probabilities seem work out. What is happening here?
3 is just flat nonsense. By that logic there's only one person, because think of the odds of being the first of 500 billion versus being the first of one. Therefore the human race will end 65 million years ago.
As stated it was somewhat garbled compared to other times I've heard the Doomsday Argument formulated. The more usual explanation is that if there are N people who ever live, a randomly selected person would more likely to be vaguely in the middle of the timeline than very close to the extreme ends.
The idea is to sample a set of sequentially numbered elements in order to estimate its total size. If you sampled thousands of elements, you could get a pretty good estimate of the size of the set since the mean of the serial numbers will be about half the size of the set. Unfortunately, we only have one independent random sample, so our error bars are going to be huge but that still gives us a point estimate of the size of the set.
So knowing you're the 100 billionth person to live gives a point estimate of 200 billion people living total in history. But the error bars at 95% confidence are stupendous, as you could still be within the middle 95% if 2.5% of humanity lived before you or if 97.5% of humanity lived before you. So the doomsday argument (if valid, and if I'm doing the math right) implies that there's 95% confidence that between 2.5 billion and 4 trillion people will live after you.
> if there are N people who ever live, a randomly selected person would more likely to be vaguely in the middle of the timeline than very close to the extreme ends.
I don't think that's the case. If we assume that the N people are uniformly distributed across time, then a randomly selected person isn't more likely to to be anywhere specific in that period of time, neither middle or end.
If the N people have a distribution that grows with time -- like the actual distribution of humans being born -- then any randomly selected person is more likely to be born towards the end of the time period. Hence the argument I made.
You would only get this "more likely to be vaguely in the middle" effect you describe if the distribution of births per year was something like a bell curve. Which it isn't.
To clarify, "vaguely" is doing almost all of the work. A randomly-chosen person would be 95% likely to be in the middle 95% of the data set, or 50% likely to be in the middle 50% of the data set.
So if your "random" person is the 100 billionth person, that suggests 95% confidence that there are between 102.5 billion and 4 trillion people who will ever live and 50% confidence that there are between 125 billion and 400 billion people who will ever live.
The argument falls down at multiple points. Firstly, everyone will only ever be able to see how many people there are in their past and present, so if you assume that's the median, this would tell every single person that humanity will end at roughly double its current size. From Adam and Eve to Pericles to Herbert Hoover, the equation is the same, and the numbers are always different.
Secondly, it assumes people pop into existence at random, when reproduction is entirely cause and effect. What are the odds you were born when you were? They're the odds of your parents having a child. The odds of two people willing and able to have children. having children, is ridiculously high. To play the median game is to argue that "you" can exist independent of the biological process that made you. Nobody born today could have been born five thousand years ago, because their parents weren't around five thousand years ago.
Thirdly, nothing says humanity isn't an infinite set. What's the median of infinity?
This whole line of reasoning strikes me as an argument of comfort. People aren't comfortable with giant numbers. They don't like thinking their time will be looked back upon with the same polite indifference we have for the Iron Age. So they create justifications for why they're actually much more important than that; civilization is peaking, right now, definitely. This cup in my hand holds 12 ounces, therefore there's only 24 ounces in the ocean.
I'll grant your second argument as being reason to doubt the doomsday argument. Serious consideration needs to be given to what the most valid reference class should be, and I haven't heard much discussion for that. You sound like you're proposing the individual as being the only sound reference class; I don't think I'd go that far, but I will provisionally accept that "all humans" is not necessarily a sound reference class for the doomsday argument.
The first argument is only an argument against taking the point estimate too seriously. A lot of people who talk about the doomsday argument take the point estimate way too seriously, and you are making a good argument against them, but I am not one of those people. Take all people who have existed or will exist and apply the argument to them, and if the correct value is finite, 95% of them should have the correct number within their 95% confidence interval.
It is conceptually possible for humanity to be infinite, yes. I have my doubts as to physical possibility, but I grant the conceptual possibility, and in that case the real value will be at the infinite end of the tail of the confidence distribution. We can't absolutely rule it out, any more than we can rule out the world ending next Tuesday due to false vacuum decay.
Bottom line is, I agree that taking the doomsday argument to conclude that humanity is likely to go extinct in the next century or so is silly. At most, I'd conclude that humanity is rather unlikely (but not vanishingly unlikely) to progress to be a Kardashev 3 civilization before we cease to be recognizably human. I'm inclined, based on triangulating with other priors, to expect that the actual number is likely to be towards the upper end of the confidence interval or perhaps a bit past it.
That same line of reasoning could have been used 99,999,999,999 times before, and it would have been wrong every single time. An argument with that kind of track record is entirely useless for anything, except maybe as an example for flawed reasoning.
Edit: I guess the most important takeaway is that "something something probabilities something Bayes therefore conclusion" will lead you astray. Bayes' rule is a mathematical theorem, and unless applied with mathematical rigor, it won't be more reliable than "my gut says X".
Indeed, the Doomsday argument (without taking into account the second argument, which is the Self Indication Assumption) would indicate that "the fewer people there are, the more likely". Maybe this can be salvaged if we consider that, yeah, it's a big Universe, most time groups of intelligent living beings appeared, they were few in number, and then died out. We just happen to be living in one of the few big groups.
But my point here isn't to defend any of the arguments I described in my original comment. My response to you is: ok, if that was wrong, where is the mistake in reasoning? The fact that this reasoning would have been wrong 99,999,999,999 times before doesn't tell us where the mistake is.
One possible answer I lean towards is "this argument is actually correct, but there are so many *other* factors coming from empirical observation that this argument's contribution to the probability distribution gets drown out. Is this something you believe?
Your example is wrong because I, a person reading your comment, am not in any sense being sampled from a random distribution of all people. Indeed, to derive 2 from 1, you assume this not to be the case. But then at step 3 you assume this is the case.
I would say most examples of anthropic reasoning one sees online are wrong, or at least under supported, because they nearly always handwave over the part where all the possibilities are equally likely.
I don't see how point 3 assumes anything aside from what is stated in points 1 and 2. Can you elaborate?
And about random sampling: well, yes, my comment is selecting for the people who are reading comment in this thread, not uniformly among everyone in history. But, in principle, anyone inn history could have just come up with the argument by themselves, so the argument -- not the post -- is still randomly selecting from Humanity..? Maybe? Maybe stone age humans wouldn't have had the culture necessary to come up with this argument? Does that change anything? What about neanderthals, do they figure in? So many questions.
You say (1) there have been about 100 billion people. I haven't checked, but I'm prepared to believe you.
Then you say (2) that I'm around the 100 billionth person to have existed, which assumes that, I, the reader exist about now and I'm not reading your comment from a colony in another galaxy a million years in the future.
Then you say (3) that I'm more likely to be the 100 billionth person if there were only a 100 billion people than if there were 500 billion people, which assumes that I, the reader, am equally likely to be each person who will ever exist.
It seems to me that some sort of dualism is underlying the argument, as if consciousness freely floated over matter, so that your consciousness could equally well have attached to any other human. But this is quite wrong: you could only ever have been Michael A. Katt. You have to actually perform a random sample, not just conceptualise it.
Damn, good point. I find that I can switch between a mode of thinking where your position is correct, where "of course I could only ever have been me", and a mode of thinking where you're wrong, where I truly could *hypothetically* have been anyone else. But these are both very informal, gut-feely modes of thinking, we should formalize what we're saying.
Classical probability theory, and the way that we often talk about probability in real life is frequentist. If I throw the dice 6,000 times I expect to see approximately 1,000 occurrences of each face, and as the number of throws goes to infinity, each face occurs 1/6 of the time. That's approximately what we mean when we say that the probability of any particular outcome in a throw of a dice is 1/6. What's key is that you've enumerated all possible outcomes and assigned a probability to each of them. This is black-and-white undergrad maths stuff.
We also say things like "Trump has a 55% chance of winning the election." You can't run the election infinitely many times, so this is a different type of statement. Essentially you're quantifying your uncertainty about an event - let's not get into the philosophy here. It's implicit here that Harris has a 45% chance of winning, and you've again got a satisfactory probability distribution.
What's the distribution in your question above? The space of all people who have ever existed, uniformly? Clearly not. What you're doing is basically reading the nearest car number plate on the street outside, saying 'There are 100 million cars in the US, what are the chances of me reading this one?' and assuming you've witnessed lottery-winning levels of unlikelihood.
Actually, if you pay attention, the way people use probability in practice is often incoherent from a mathematical perspective. And saying 'Bayesian' doesn't solve all the problems.
My goal when I explained the doomsday argument in with my comment wasn't to mean "and I believe this makes sense and is true", it was more to give an explicit example people could dismantle and point to the faulty part, is there is one. In this case, I understand you to to be pointing to this as the faulty part:
> What's the distribution in your question above? The space of all people who have ever existed, uniformly? Clearly not.
The distribution towards the future isn't assumed by the argument, it is actually the thing that the argument is trying to determine, based on both the previous distribution and the observation that the person making the argument was born at the time they were born.
> What you're doing is basically reading the nearest car number plate on the street outside, saying 'There are 100 million cars in the US, what are the chances of me reading this one?' and assuming you've witnessed lottery-winning levels of unlikelihood.
We are witnessing a 10^-8 probability event every time we look at a unique license plate from a pool of 100 million. Yes, this is a very very low probability. But it's not a surprising event, since every counterfactual event -- looking at any other license plate -- would have had the same low probability. An event isn't surprising when it's low probability, it's surprising when it's probability is small *compared to* other possible events.
What the argument is doing is analogous to looking at a "4321" license plate and thinking "if there are 1000 license plates, I just looked at a 1/1000 prob. event. If there are 10000 license plates, I just looked at a 1/10000 event. The former is 10 times more likely than the later."
In general people seem to be horribly confused about anthropic reasoning, which originate from not properly understanding probability theory and its limitations. Common mistake goes like this:
1. Be not randomly sampled from a set.
2. Assume that you are randomly sampled from the set, anyway
3. Arrive to a wrong conclusion
or a bit more generally:
1. Participate in a probability experiment A
2. Assume that you participate in a probability experiment B, that is not A
3. Arrive to a wrong conclusion
Sadly, the confusion is so pervasive that almost noone is thinking about anthropics reasonably. Most mainstream anthropic theories are just arguing which crazy bullets to bite, instead of trying to find a systematic ways not to do it and arrive to correct conclusions.
The two approaches you describe are the Self Sampling Assumption (SSA), where you consider yourself as randomly sampled from all possible observers (in this case, all people that will ever live) and the Self Indication Assumption (SIA), where instead you should weight across all possible observer-moments, adjusted by the probability of that moment (in this case that means that you'd weight worlds with more people heavily).
If you've heard of the Sleeping Beauty paradox, SSA results in a thirder position and SIA in a halfer position. (EDIT: Other way round, see correction below)
The Doomsday argument uses the SSA, and doesn't hold if you take the SIA (because the SIA weights timelines with lots of humans more highly). I think that's what Scott's referring to in your linked post.
However, the SIA has its own paradoxes. For example, if scientists had two theories of the universe, one in which the universe would have trillions of lives and one in which it would have hundreds of trillions, the SIA would imply we should prefer the latter just because it has more observers. (This is known as the Presumptuous Philosopher argument).
> If you've heard of the Sleeping Beauty paradox, SSA results in a thirder position and SIA in a halfer position.
The other way around. SIA leads to thirdism, while SSA leads to halfism. And you do not actually need to subscrive to SSA to be a halfer (double halfer) in Sleeping Beauty, but that's a different story.
I tend to think of the whole approach as cursed and probably hopeless. You're trying to best-guess distributions of possibilities that we cannot check even in principle, and the margins of error are so big that you can convince yourself of "reasonable-sounding" numbers, then apply some bayesian formulas, and get overwhelming probabilities about things that we have no actual information about. Count me out.
My take is that all of the above is true, it's just extremely weak evidence and you need to take that into account when updating your priors. It's easier to see this if you apply this logic to more human scale scenarios.
For example, since moving to a new city in August, I have purchased 5 pints of chocolate fudge brownie ice cream. Obviously it's more likely for the pint of ice cream currently in my freezer to be 1 out of 5 total pints that I've purchased than 1 of 500. So I probably won't end up buying very many more pints of ice cream (probably good for my health!)
Estimating my lifetime ice cream purchases from this information alone probably feels pretty sketchy. You would want substantially more information, and it would be easy for that information to change your best guess of how many pints of ice cream I'll end up buying. Similarly, a doomsday argument *just shouldn't change your mind all that much* if you have... basically any other priors about the longevity of the human race.
It is absolutely true that the observation "I am the nth person" is more likely if there are going to be a total of n people than a total of 4n people.
But the implication of Bayes is that observing that should make you increase your estimate the relative likelihood of there being n as opposed to 4n people from your prior, not that it is more likely that there will be n than 4n people - if your prior on 4n people was significantly higher than your prior on n people - which it probably should be - then your posterior may well still be higher.
Currently the implied probability (to the nearest whole percentage point, mid-price) of the following people winning the US Presidential Election is, in the order Betfair/Polymarket/Metaculus/Manifold (numbers 4 weeks ago in brackets):
Trump: 60/66/54/56 (48/49/44/46)
Harris: 39/34/46/43 (51/50/57/53)
UK bookmakers' odds imply 60/40. I understand Nate Silver is at 53/47.
Besides the obvious shift in favour of Trump, it is notable that the gap between Polymarket and Metaculus on Trump's chance of winning has increased to 12pp, up from 5pp 4 weeks' ago. Previously, these were never more than 4pp apart (and not always in the same direction).
If one were to take at face-value the claim at https://www.astralcodexten.com/p/mantic-monday-12924 that Metaculus is more accurate that real money markets specifically at election predictions, it would follow that there is a profitable bet to be made for those who are able to bet on Polymarket.
That is, if Harris's true chance of winning is 46% and you buy 1,000 shares at 33.8c each, you lose $338 54% of the time and gain $662 46% of the time, so you make $122 (less transaction costs) in expectation. To be clear, the most likely outcome is still that you lose all your money.
If you can't bet on Polymarket, but you can bet on Betfair, you do less well, but (on the same assumptions) still have a bet with positive expected value.
You get a similar result if you take Nate Silver to be an accurate forecaster.
A note of caution is that on 4 March 2024, Metaculus had Biden as 45% likely to win while Polymarket had Biden at 32%. With the advantage of retrospect, Metaculus predictors were overconfident that Biden would be the nominee, while Polymarket predictors were correct is assigning a significant chance to the possibility he would drop out (although largely incorrect in predicting who would replace him as nominee).
One possible model would be that Polymarket was well calibrated up to roughly the end of August, but is losing accuracy as the election approaches because of an influx of "dumb money". If correct, one might expect this trend to continue, suggesting that it might be better to wait until even closer to the election before placing bets.
Right, I'm definitely sticking a tenner on Trump! Even if he loses, I will feel the smug glow of moral vindication when Harris keeps her promise to Build The Wall and everyone conveniently forgets this was Wicked And Evil because Trump proposed it.
This is a common misperception. Trump only talks about the wall a lot, but its construction significantly predates him. The border wall goes back to the Clinton administration with the Illegal Immigration Reform and Immigrant Responsibility Act of 1996. During W’s administration, the Secure Fence Act authorized the construction of roughly 700 miles of fencing along the U.S.-Mexico border.
Under the Obama administration, much of this previously authorized fencing continued to be built. In fact Trump recently made a campaign stop at “his wall” to talk about how he was tough on immigration. The impressive section of wall in front of which Trump made his speech was constructed before Trump ever held office, under Obama’s administration.
While it should in theory be possible to arbitrage, the cut that the platforms take, and the third party risk of storing your money with them, is such that a lot of people would consider it not worth it.
I'm not sure what you mean by "funky", but I see no reason to think these bets are anything other than people with a lot of a money who expect Trump to win. The guy who bought 275,109 Trump Yes shares at 99.7c was a new account and I'm sure it was an error. He's since made a large number of trades in the same market at normal prices: his total position is 7,542,442 shares. If anything it's a warning about Polymarket's lack of protection for punters.
Some people can do the arbitrage. Polymarket states that they do not charge fees https://learn.polymarket.com/docs/guides/trading/fees, although their terms are not at all transparent and there likely are some costs involved. Betfair will charge 2% of profit in any given market (on their "basic" reward plan). The spread between the two sites is sufficiently high that you should end with a tidy profit, but the practical difficulties are significant. To do it properly you would need to hedge the currency risk. The sites also have slightly different resolution criteria which in theory could result in them having different outcomes (e.g. if AP, Fox and NBC all call the election for Harris, but Trump doesn't concede and ultimately Congress certifies Trump to have won, then it will count as a Harris win on Polymarket and a Trump win on Betfair). Nevertheless, I would expect the gap between the real money markets to close. I note that Betfair has moved to 62/37 and the bookies have moved to 62/38 since my first comment, which is the direction I would expect the correction to go given the extraordinary volumes on Polymarket.
ETA Bloomberg are wrong to say that the French guy bet $45m: he currently holds 47m *shares* across the 4 named accounts, which he bought for a total of about $26m. I'm guessing he held 45m shares when they checked.
Why would you assume that they *expect* Trump to win, as opposed to *wanting* Trump to win?
Prediction markets have not grown up to be the knowledge equivalent of stock markets, where serious people go to *make* money. Prediction markets are, for everyone but a handful of nerds, a place where people go to *spend* money. Rather like Las Vegas - loads of fun if you're into that sort of thing, and maybe good for your status or self-image if you don't lose *too* much, but if you're a Yankees fan who thinks the Dodgers are the 60/40 favorite in the World Series, you're still not going to bet on the Dodgers because that wouldn't be any fun.
And, like Las Vegas, that does make opportunities for smart, disciplined nerds to make a bit of money around the edges. But to extract useful signal from that, you'd need a way to distinguish smart-disciplined-nerd money from status-seeking-whale money. Do you have one?
In a reasonably well functioning prediction market, the disciplined nerds are going to jump on any pricing discrepancies caused by the dumb whales.
The best evidence that this is happening is that all prediction and betting markets around the world are currently in agreement that Trump's chance of winning is closer to 60% than 50%. If the markets were vastly inefficient and dominated by dumb money we'd see a wider range between different real-money markets. Only the fake-money markets are close to 50-50.
We don't yet have a well-functioning prediction market. We have well-*administered* prediction markets, which can be mostly counted on to e.g. pay out as called for by the propositions, but that merely a necessary and not a sufficient condition for a well-functioning market.
For that, you also need a large community of well-informed, well-capitalized traders with experience in the market at hand, and it's too young for that. There are a few disciplined nerds who will nibble happily and profitably at any price discrepancies, but not enough to arbitrage those discrepancies out of existence,
I don't think those things are as distinct as you suggest. Wanting Trump to win, they convince themselves that he will. Similarly, I don't think your Yankees fan *does* believe the Dodgers are 60/40 favourite. They can give you twenty reasons why the this is the Yankees' year, so it looks to them that the bookies are selling dollars for 40c.
I'm sorry to say this, but I think people do go to Vegas to make money. They usually fail, but that doesn't mean they're not trying.
What I am very tentatively suggesting is that the smart disciplined nerds congregate at Metaculus, so one can read off the true probabilities from there and bet accordingly. But also, you don't need a general rule if you are confident you've spotted the whale in this particular game, which it kinda seems like you are.
Looking at bets over $100k on Polymarket, there have been 13 in the past 24 hours, by 6 different users. Totals are $1.879m for Trump to win, $132k for him to lose, $325k for Harris to win. So it does look like whales are breaking for Trump.
I am tempted to do exactly that - bet a small amount, as in, something like monthly subscription to this blog, on Harris on Polymarket. Does anyone by any chance know what are the costs on Polymarket? I don't own any relevant crypto. I've asked Perplexity and it answered with this not-very-quantitative and not-very-encouraging lecture (Polymarket's own info is even less informative):
"Cost Structure on Polymarket
Trading Fees:
Winning Trades: Polymarket charges a 2% fee on net earnings from winning trades. This means that if you win a bet, 2% of your profit will be deducted as a fee15.
Losing Trades: There are no fees for losing trades, which makes it more appealing for users who may not always win their bets1.
Transaction Costs:
Polymarket operates on the Polygon blockchain, which generally has lower transaction fees compared to Ethereum. This helps keep costs down for users engaging in frequent trading13.
Users may incur gas fees when depositing or withdrawing funds, but these fees are paid directly to network miners and not to Polymarket itself2.
Deposit and Withdrawal Fees:
No Direct Fees: Polymarket does not charge fees for deposits or withdrawals, which is advantageous for users looking to manage their funds without incurring additional costs13.
However, when transferring USDC (the currency used on the platform), users may encounter gas fees associated with the Ethereum network if they are not using Polygon2.
Liquidity Provider Fees:
When trading, users might also pay small fees to liquidity providers who facilitate transactions on the platform. These fees are generally minimal and help maintain market liquidity34.
Market Prices:
The prices of shares on Polymarket reflect the probability of an event occurring and can fluctuate based on supply and demand dynamics within the marketplace45."
Then I found out that "gas fees" are apparently measured in a unit called gwei, and then I've exhausted my attention span
I would also be interested to know how this works out, i.e. if you currently have some fiat currency in a bank account, what is the total transaction cost of buying USDC, moving it to Polymarket, placing a bet, collecting your winnings, converting to fiat and redepositing in the original account?
At one point I thought I would try it to find out but I was unable to deposit USDC to Polymarket; when I contacted Polymarket they informed me that trading is not available to UK users (although there is still nothing in their Terms to indicate this).
By way of update, currently we have Polymarket at 65/35, with Betfair at 62/37, which matches my prediction that the gap between the two real money markets would narrow. I don't necessarily expect them to get closer than 2pp: my gut feeling is that the inefficiencies are sufficiently large that a difference of that magnitude could be sustained indefinitely. Meanwhile, Metaculus is at 52/48.
I note that over the past 24 hours, there have been 9 bets over $100k on Polymarket, by 8 different accounts, and all of them have been bought Trump Yes. It therefore seems to me that the proposition that whales are breaking for Trump still roughly holds, so I still weakly expect (ceteris paribus) the odds to shift in Trump's favour between now and the election.
"In this particular case, of course, we know for a fact (!!) that Russian intelligence was engaging in cyber-warfare to influence the US election. We also know that Russia has both the will and the technological ability to tamper with foreign elections using vote-stealing malware—indeed, it nearly succeeded in doing so in Ukraine’s 2014 election. Finally, we know that Trump, despite losing the popular vote, surprised just about everyone by outperforming his polls in three crucial swing states—and that within those states, Trump did systematically better in counties that relied on electronic voting machines than in counties that used scanners and paper ballots."
Scott Aaronson, November 2020:
"These past few weeks, my comment queue has filled with missives, most of which I’ve declined to publish, about the giant conspiracy involving George Soros and Venezuela and dead people, which fabricated the overwhelmingly Democratic votes from overwhelmingly Democratic cities like Philadelphia and Milwaukee and Detroit (though for some reason, they weren’t quite as overwhelmingly Democratic as in other recent elections), while for some reason declining to help Democrats in downballot races. Always, these commenters confidently insist, I’m the Pravda-reading brainwashed dupe, I’m the unreasonable one, if I don’t accept this."
Unlike so many others, he does often sound like someone trying to take a principled position and to clarify the things he agrees with his opponents on (e.g. he does say in both posts that he's been long and consistently in favour of hand-recounting every election). And yet even HE is able to express almost completely opposite attitudes to similarly unsubstantiated "my side didn't really lose" conspiracy theories. Most people are a MILLION times even worse.
When is this kind of brazen hypicrisy going to stop? And more importantly, when are people going to realise that this kind of thinking is at the root of everything that's wrong with politics? Every single person who talks with disbelief and fear about the way Trump behaves needs to ask themselves if their reaction to stolen election claims was completely different in 2016 and 2020, and if it was to take a good look in the mirror and understand that everything they hate about Trump is a product of the society they've created.
> By far the most common taxonomy of fascism you will see in most academic circles is that advanced by Umberto Eco in his essay “Ur-Fascism,” copies of which you may easily find online. Like most things academics like, it is complex and a bit fuzzy.
> As Eco notes, fascism as a set is somewhat like the series of “abc bcd cde def” in which all of the elements are clearly a family and yet in freely remixing core elements, it is hard to identify a single set of necessary components.
> instead of a single definition, Eco proposes a taxonomy of 14 points which collectively make up the ‘type group’ of fascism. An ideology doesn’t need to have every point in order to qualify as fascism, but the more criteria it satisfies, the more firmly it fits into the definition. Likewise, individual points can also be fuzzy or borderline – it is the cumulative fit that matters – after all, most ideologies have effectively 𝗻𝗼𝗻𝗲 [emphasis original to Devereaux] of these attributes, so an ideology that clearly has many and sort of has a few more actually fits the mold quite well.
So far so good. But these are the 14 points:
1. The cult of tradition, particularly a syncretic traditionalism that latches on to various iterations of an idealized past, even mutually incompatible ones.
2. The rejection of modernism, specifically, in Eco’s mind, a rejection of “the Enlightenment, the Age of Reason…seen as the beginning of modern depravity.”
3. The cult of action for action's sake.
4. Disagreement is treason.
5. Fear of difference.
6. Appeal to a frustrated middle class.
7. The obsession with a plot.
8. The deceptively strong/weak eternal opponent. This one is complicated, but fascism conjures an ‘enemy’ who is at once too strong (thus requiring the power of the fascist strongman to defeat and whose continued existence can justify continued mobilization and authoritarianism) and yet also degenerate and weak: thus the Nazi view that Jews were racially inferior, weak and cowardly, but also that they secretly controlled all of the world’s most powerful countries.
9. Life is permanent warfare, as Eco puts it, “there is no struggle for life, but rather, life is lived for struggle.”
10. Contempt for the weak.
11. The cult of heroism.
12. Machismo.
13. Selective Populism: fascism claims to speak for ‘the People,’ but in reality only some of the people and not through democratic, majority-rules systems; rather the Leader channels the Common Will which is taken as the Voice of the People, even when it contradicts the actual votes of the people.
14. Newspeak, which Eco identifies both in changing the meaning of words, often inverting them, but more broadly, I think Eco is getting at here what Orwell sums up in his famous line, “The party told you to reject the evidence of your eyes and ears. It was their final, most essential command.”
[He provides discussion of each point, at the link.] In conclusion:
> Fourteen out of fourteen; some are clearer and stronger fits than others, but every element is present to a significant degree. Keep in mind, this is the sort of taxonomy where a regime which, say, satisfied 10 or 12 out of the 14 would still be generally regarded as fascist and the ideology we might call ‘Trumpism’ 𝙛𝙞𝙩𝙨 𝙖𝙡𝙡 𝙛𝙤𝙪𝙧𝙩𝙚𝙚𝙣.
And this looks completely insane to me. Let's just stipulate that points 8/9/11/12 don't apply to the US Democratic party. The other ten clearly do. And I'm pretty sure that five of these points, #s 1, 3, 5, 6, and 13, apply to every political movement that has ever existed, modulo slight cultural translations. What happened to "even one of these would be alarming"?
1 (or 5, I think) would not apply to the Civil Rights, the European Socialists or Social-Democrats, not even the Modernists, the beginnings of the United States, or whatever the American political consensus was in the 1990s?
3 also seems to not apply that often (but my perspective may be twisted as a non-American). Saying “oh well, this seems important issue, surely we should study this more before doing anything rash” seems to have been uncontroversial for a long time.
This is why there are so many think tanks putting ideas out there and publishing studies rather than physical tanks vying for political control (obviously, the second part is a jest).
Note ACOUP’s elaboration on this: for Eco, the worship of action for action’s sake is coupled with a disdain for learning and thinking. It’s not “we’ve heard all sides and thought it through, now is the time to take action”. This is, in spirit, closer to “let’s get all these useless [sneer] thinkers back to doing real work”.
I also disagree about the universality of 6 and 13, but it’s harder for me to articulate precisely why.
Oh? What are some movements that don't or didn't promise to make people better off? The message "actually, people are doing fine" is the antithesis of a movement.
That's why I said "modulo slight cultural translations". If 94% of the population is subsistence farmers, then "appeal to a frustrated middle class" will be "appeal to a frustrated peasantry". There's nothing special about the "middle class" as defined by American usage.
> One hundred years of delay have passed since President Lincoln freed the slaves, yet their heirs, their grandsons, are not fully free. They are not yet freed from the bonds of injustice. They are not yet freed from social and economic oppression. And this Nation, for all its hopes and all its boasts, will not be fully free until all its citizens are free.
Martin Luther King:
> On our part we must pay our profound respects to the white Americans who cherish their democratic traditions over the ugly customs and privileges of generations and come forth boldly to join hands with us.
Martin Luther King, same speech:
> Let us march on ballot boxes (Let us march. March) until brotherhood becomes more than a meaningless word in an opening prayer, but the order of the day on every legislative agenda.
> Let us march on ballot boxes (Yes) until all over Alabama God’s children will be able to walk the earth in decency and honor.
> There is nothing wrong with marching in this sense. (Yes, sir) The Bible tells us that the mighty men of Joshua merely walked about the walled city of Jericho (Yes) and the barriers to freedom came tumbling down. (Yes, sir)
Joseph Jackson:
> As Christians we are a part of our nation and a part of the struggle of America. America was brought into being to satisfy and to answer the human longing for freedom. There was the urge in man to be related to other men as men without a modifier or any kind of limitation or restriction. There was an awareness of a human kinship deeper than race, more profound than nationality, and more inclusive than any accepted religious creed. In addition to the quest for a new geographical spot there was a search for a new human relationship, a new freedom, and new opportunities. These basic urges inspired the early colonies to brave the dangers of a rough and unknown sea, and seek a land in which they could live as free men and aspire to the highest possible goals of life without the enslavement of the past or being the victims of the determinism of enforced circumstances. They wanted a chance to explore and to search out the meaning of life for themselves, and an opportunity to worship God according to the dictates of their conscience.
> This American venture is powerful but not perfect; ever growing but not grown; and still becoming, but is not yet complete. The kind hand of destiny and the benevolent providence of Almighty God have placed the American Negro along with other races and nationalities in this flowing stream of the nation’s life for which we are justly proud. As patriotic Americans we are devoted to our nation’s cause, and are wedded to its ideals and principles.
Malcolm X:
> When this country here was first being founded, there were thirteen colonies. The whites were colonized. They were fed up with this taxation without representation. So some of them stood up and said, “Liberty or death!” I went to a white school over here in Mason, Michigan. The white man made the mistake of letting me read his history books. He made the mistake of teaching me that Patrick Henry was a patriot, and George Washington… Wasn’t nothing non-violent about ol’ Pat, or George Washington. “Liberty or death” is was [sic; https://www.rev.com/blog/transcripts/the-ballot-or-the-bullet-speech-transcript-malcolm-x ] what brought about the freedom of whites in this country from the English.
> They didn’t care about the odds. Why, they faced the wrath of the entire British Empire. And in those days, they used to say that the British Empire was so vast and so powerful that the sun would never set on it. This is how big it was, yet these thirteen little scrawny states, tired of taxation without representation, tired of being exploited and oppressed and degraded, told that big British Empire, “Liberty or death.”
Moving on...
> 1 [...] would not apply to [...] even [...] the beginnings of the United States
Why do you think the more powerful half of our legislature is called the "Senate"?
Heck, I'm not even sure 8/9/11 don't apply to Democrats (oppression / capitalism are the eternal enemies; we're always fighting a war somewhere to export democracy abroad; the people who fight oppression / capitalism / whatever are our heroes). I can concede 12, but if something meets the other 13, I'm not likely to point at 12 and claim there's nothing to worry about.
My irritation otherwise parallels yours. My reaction was something like "looks like my ham sandwich is fascist". I concluded that I had better wait until I read a bunch of other Eco material before I form an opinion of him, because that alone was hackneyed.
Haven’t the recent foreign US wars (especially of the “to export democracy kind”) been instigated by Bush?
Obama refused to get involved in Syria even after Assad defied the stated “red line” about chemical weapons, if I remember correctly, and tried to appease Iran with the JCPOA (where Trump took a far more hawkish stance by denouncing it).
Deveraux explicitly states that 8 is a subtle point, and the weak/strong ambivalence does not apply as straightforwardly to capitalism or oppression.
Political movements usually *want* something, and have to struggle with people who oppose them. But 9 refers to the glorification of struggle in and of itself as the one purpose of life.
An anti-capitalist fights capitalism because life will be better after. Washington fought the British for American self-rule (or at least no taxation without representation, or so the myth goes). Johnson didn’t fight the Vietnam War because war is glorious and to make men out of the hippies, but for geopolitical reasons (the Southern ruler was an ally of the US, the ideas of containment and roll-back, the desire to protect the South Vietnamese population from communism…).
(Also, it seems wrong to conflate the Democratic party with the anti-capitalists? A serious anti-capitalist would hardly admit to any remotely significant economic difference between the Democrats, the Republicans, the Libertarians, and European-styled social-democrats. I’m extremely skeptical of any material influence that the “cultural anti-capitalists” that show up online could wield.)
11: I’d be surprised if Eco thought that all heroes (or cults) were alike. My guess is that there’s a very specific notion of heroism he has in mind, closer to the models of Hercules or El Cid (heroes that are great by war, conquest, or their raw strength), rather than a builder (in a wide sense), an artist, a thinker, or someone who spearheaded a great charitable effort.
Bush only started Afghanistan and Iraq with Democratic Congressional approval. Obama sent CIA operatives and air support into Syria. Clinton sent troops into Bosnia. Point being, it's not the case that Democrats never do this, and meanwhile, they're still our #1 exporter of rhetoric about fighting against oppression and racism and sexism and so forth.
Devereaux's explanation of 8 describes a duality of "too strong" but also "degenerate and weak", "often just a nebulous 'they'". That sounds like systemic racism to me.
I admit that "we're always fighting a war somewhere" (for its own sake) wasn't quite hitting me as a textbook example of "life is permanent warfare", but then you rephrased it as "glorification of struggle in and of itself as the one purpose of life" and I instantly thought of "revolution for the hell of it" and people who activists and look down on anyone who isn't as "sheeple" who "aren't paying attention".
I don't think it's fair to justify Washington's and Johnson's wars and then not look at the justifications for Bush's or Trump's continuation of them. We should either look at both, in which case none of these is fascist, or we ignore them all and point out their existence, in which case they all are. Which is again why I don't like this litmus test.
I don't think every Democrat is anti-capitalist, sure. But that's the side the anti-caps camp on. If we're looking for people who profess that eternal struggle, we're not going to start looking on the GOP side.
I'm sure Eco didn't see all hero cults the same, either, but if he's going to try to explain to us how to spot a fascism, I don't think he really meant "how to spot what meets Eco's personal peccadillos as fascist and naturally this will be an objective test". So I'm going to go by the picture I think he's trying to paint. I originally just trusted the comment above to tell me, but if I go to the original text: "The Ur-Fascist hero craves heroic death, advertised as the best reward for a heroic life. The Ur-Fascist hero is impatient to die. In his impatience, he more frequently sends other people to death." Are we to infer that Democrats don't revere people who will risk their lives to fight oppression / bigotry / capitalism / etc.? Earth First? Weather Underground? The PLO? Gabby Giffords? Heather Heyer? Antifa?
Sorry, I didn’t mean to sidetrack the discussion by bringing Bush up. I felt from your comment that you were essentially accusing the Democrats of being in favor of permanent warfare.
(Although, if I remember correctly, he’s the one who decided it, lied brazenly to get his way, and that did a lot of damage to the US credibility. But that’s irrelevant.)
I think that “permanent warfare” does not refer anyway to “state-on-state warfare”, or at least not necessarily. The Nazis’ designated “racial” enemies did not come from a particular state.
Part of me wants to jest that Umberto Eco was a European leftist, so the difference between Democrats and Republicans would have barely been perceptible to him.
But (one of) Bret Deveraux’s points is that Trumpism is not close to the Republicanism of Reagan, Bush, McCain, or Romney.
The dimension that I don’t find in your comments (but is crucial to the appreciation of phenomenon) is that of scale, or perhaps centrality. Concentration, a chemist would say.
I’m sure that in [sufficiently large group that you personally respect], there will be a lot of [bad things]. The issue is when the bad is prevalent – especially when [said bad thing] it’s part of the group’s identity.
Sure, a lot of political movements will have their extremists, or snub activists that will scoff at anyone that doesn’t embrace and spend their lives fighting for their pet cause.
(Although, again, an anti-capitalist who identifies as a Democrat is about as silly as an astrophysicist who believes in astrology.)
But what I think Eco is saying here is that for a fascist movement, this exaltation of life as a struggle does not come downstream of having extremists or people deeply committed to some aspect of the cause. It is a central tenet of the ideology (perhaps the word “vibe” would be more appropriate if the movement scorns intellectualism). Hence the necessity for an enemy – instead of, say, a challenge or a shortcoming of our societies to overcome.
I don’t understand how “systemic racism” fits as a permanent enemy in point 8, could you please elaborate?
Could we perhaps clarify what your final examples intend to prove? They seem to be a very loose mix without a strong common theme. In Eco’s definition, both the longing and the giving of death seem important.
This is why the two individuals that you cite seem, in my opinion irrelevant to the discussion – they were participating in pretty regular political processes until they were attacked (from a one-second skim of the Wikipedia pages).
The Weathermen seem to have avoided this vibe as well, although, few people would find much to admire in them.
The PLO is an entirely different kettle of fish, since it is literally a terrorist organization.
Antifa is admittedly quite a bit less clear-cut, but it’s also non-central.
If you're going to get in a parting shot, I'm going to have to return volley. :-) (As I understand it, GWB didn't lie brazenly about Saddam so much as tasked intelligence to gather evidence, and reported on what they found, which was going to be something anyway, and did so with the agreement of Congress and most Western nations who had their own intel. Bipartisan level view here is that neither account is agreed upon.)
But yeah, I can't in good conscience characterize the DP as pro-permawar. Pro-perma-protest, however, is certainly on the table. It's just not "machismo" in the Eco sense.
I don't see enough chemical concentration of bad in Trumpism to find Devereaux's claim convincing. Trumpism isn't "unusually bad" nearly as much as it's "made to look unusually bad by a group motivated to do so", and the evidence is hundreds of unfair exaggerations. Meanwhile, Devereaux isn't being scientific; if he were, he'd say, "so, wait a minute" and hold the Democrats up to those 14 points, as Michael Watts did above. And then he'd (I contend) be back to: either everyone's fascist or these points are too underspecified, as I've been arguing here.
I don't see why it should matter whether the life-struggle idea comes intrinsically or is downstream of extremists. First off, fascism itself doesn't arise from a vacuum; any group adopting it is responding to past events anyway. Secondly, we're not supposed to like that idea regardless. So arguing origins sounds like we're arguing whether this thing before us is "technically" fascism or not - as if federal funds will kick in or something if we can secure the credential.
"Systemic racism" is the permanent enemy in the sense that no matter how hard one tries, it'll be there. Its proponents present it as a grime that gets into everything, necessitating nothing short of a radical restructuring - or a permanent cadre of watchdogs, which the left is only too willing to provide in exchange for whatever resources and cultural commitment they determine. Fascism seeks the same sort of sinecure.
I get the sense you're unfamiliar with the history of the Weathermen. They made the news for bombing all sorts of institutions in the name of leftist causes. Bill Ayers' own girlfriend was killed while making one of them. He, and everyone else in that paragraph, are people who either risked their lives or risked others' lives for the cause, and are admired by the left more than by the right for it. This includes some Democrat voters.
"The Ur-Fascist hero is impatient to die. In his impatience, he more frequently sends other people to death."
> I concluded that I had better wait until I read a bunch of other Eco material before I form an opinion of him, because that alone was hackneyed.
I had a different concern; Devereaux explicitly notes that this framework is used by modern scholars who want to discuss fascism as a topic, and based on the contents of the framework, I can't see that being useful for any purpose other than forming an echo chamber.
Having a literal mind, I tend to see "fascism" as being defined by point 4, which is what's emphasized by the name "fascism".
I always saw "fascism" in terms of "fascia" - generally speaking, a casing around a bundle of threads or fibers or cables that holds them together and makes them stronger. Which, by itself, sounds fine and sensible. "Stronger, Together."
The side of it that worries people who are students of history is when someone like Mussolini steps up and claims "everything within the state, nothing outside the state", meaning not just "hey, we'll get farther if we work together" but also "anyone who doesn't is going to get smushed by my fascia and everyone inside it, hint hint".
I notice that practically no one who espouses some version of "Unity" or "Stronger, Together" doesn't seem eager to add "But If You Really Prefer to Go Your Own Way, That's Cool, Too". Well, except for libertarians.
Now, the more I think about that essay, the more I think it would have gotten farther with everyone if treated as "Here's a Rough Sketch of Warning Signs That Your Society *Might Be* Headed in the Wrong Direction", rather than the apparent treatment of "Run This Program on a Society and if it lights up Red, Drop Everything and Dismantle It". And only run it on whatever one didn't like already.
The only reason I think conservatives aren't as susceptible to this is merely that few conservatives read Eco.
As I understand it, the word is derived not from "fascia", a casing around a bundle, but from "fasces" (or the Italian reflex "fascio"), referring to the bundle itself. The idea is that while it's easy to pick up one stick and snap it in half, you can't do the same with the bundle; numbers and alignment lend it strength.
> I notice that practically no one who espouses some version of "Unity" or "Stronger, Together" doesn't seem eager to add "But If You Really Prefer to Go Your Own Way, That's Cool, Too". Well, except for libertarians.
Well, I don't think the libertarians are doing much to espouse unity. They're fine with you going your own way, but they don't have a plan for anything other than that.
The reason no one tries to send both of those messages is that they're inherently opposed.
> The only reason I think conservatives aren't as susceptible to this is merely that few conservatives read Eco.
In a sense, I agree. There's nothing unusual about the impulse to make up some criteria that describe everyone and argue that those criteria show that someone you hate is bad, while glossing over what they might show about anyone else.
Academic scholarship is very slanted, such that errors are going to be overwhelmingly pro-leftist, but that's just because academics are leftist.
This was a funny example of that slant: https://pubmed.ncbi.nlm.nih.gov/34383522/ . ("Authoritarianism has been the subject of scientific inquiry for nearly a century, yet the vast majority of authoritarianism research has focused on right-wing authoritarianism. In the present studies, we investigate the nature, structure, and nomological network of left-wing authoritarianism (LWA), a construct famously known as "the Loch Ness Monster" of political psychology.")
You apparently think that the 2016 and 2020 were the same in ways that Scott Aaronson does not. Perhaps you think that being concerned about Russia, which has sophisticated cyber attack capabilities and a demonstrated intent to interfere in the 2016 election in favor of Trump, is the same as proposing a wild conspiracy involving Soros (who is pro-democracy) doing the same thing. Perhaps you think Scott’s proposed response (performing a hand recount and accepting the result) is the same thing as the Trumpest response (trying to get the election results throw out). I say “perhaps” because you don’t really engage with Scott’s views at all.
You accuse Scott Aaronson of “hypocrisy,” an accusation that he is acting in ways inconsistent with his professed beliefs, indicating that he is being dishonest about those beliefs. Since you don’t identify any actions taken by Scott other than writing in his blog, the accusation is ambiguous. Sure, writing a blog entry is an action, but which blog entry constitutes the action demonstrating Scott’s true beliefs and which is the one where Scott is being dishonest about those beliefs?
To summarize:
1) You make a claim about the similarity between the 2016 and 2020 elections but don’t provide a clear statement of this claim or support it with evidence.
2) You claim that Scott’s views in 2020 are inconsistent with his 2016 views. This claim that is unsupported because it depends on the preceding claim.
3) You claim that this inconsistency shows bad faith on Scott’s part. Even if you could demonstrate an inconsistency, which you haven’t, that still wouldn’t establish bad faith, because people can honestly change their opinions over a four year period.
In short, I believe you have utterly failed to make your case.
I am saying that he is applying wildly different standards to extremely similar partisan claims of election rigging, that both lack any hard evidence and rely entirely on "this sure looks suspicious" vibes. In the case where the claim benefits the Democrats, he describes it in the most charitable and credulous way. In the case where it benefits the Republicans, he strawmans it and points out all its flaws. (I have barely ever heard Soros mentioned by 2020 election deniers--the main thing discussed is always the last minute pandemic-based voting changes.)
You can easily imagine a Republican version of Scott Aaronson (perhaps gay, and having been viciously bullied by his party but still slavishly loyal to them, to match the Democrat version) saying in 2016:
"This past week, I've gotten endless rants about the shadowy conspiracy involving Russia and the FBI, Facebook and the New York Times, all working in concert with a nebulous White Supremacist Patriarchical Agenda to hack machines and fabricate votes for Trump (but for some reason not fabricating enough votes to give him a popular vote win and greater legitimacy). They also confidently tell me that Trump is supported by all the most powerful institutions in our society, and in the same breath that he's "not normal" and "a deplorable" and an unprecedented threat to the security of society. Funny how that works."
And saying in 2020:
"We know for a fact (!!!) that Democrat officials have changed voting rules at the last minute and without proper authorization across the country. We also know that Democrats, despite falling far short of expectations across all down-ballot races, did systematically better in the presidential race, and that their vote was consistently higher in those states that adopted the voting rule changes."
And maybe you can see my point? There are maximally charitable and maximally uncharitable ways of presenting (in these cases bullshit) claims, and he chooses opposite approaches that concord with partisanship.
And I acknowledged that he is more consistent than most in his support for hand-recounts. It's not using the Russia claims as a reason for a hand-recount that's the problem, it's signal boosting bad epistemology that radicalises people and that was later used many times to justify various socially destructive actions, up to and including violent riots, just like Trump.
Your point about hypocrisy is confusing and I think greatly overcomplicating things. I'm using "hypocrisy" to mean applying inconsistent standards to the same thing (claims of election rigging that lack any direct evidence) depending on whether it benefits your side. That's it.
As for "he may have just changed his mind", I haven't seen any acknowledgement of that, and moreover that seems like a way to dismiss any charge of hypocrisy ever. Maybe the moralistic preacher just genuinely changed his mind about the importance of chastity just before he visited the brothel, and changed back as soon as he finished there! It's an isolated demand for rigor.
A few years ago, before Roe v Wade was overturned, I remember reading a Texas law banning abortion but somehow managing to escape judicial contest through some sort of loophole (was it banning the clinics, but explicitly not enforcing the ban, yet rewarding isolated citizens who did, so that no one had standing to contest the law?).
I remember thinking that the legal argument was a bigger deal than the content of the law itself, that it would trigger a flurry of similar laws (which each side would use to ban stuff that they disapproved of).
This doesn’t seem to have been the case, so I’m wondering what happened?
1) was that law struck down as unconstitutional (ie did contesting the law work)?
2) did it become a non-issue after Roe v Wade was overturned?
3) did the legal argument work, but US politicians were not actually willing to sacrifice the spirit of the rule of law to its letter?
4) were there no comparable “political gains” to be made with such laws?
Texas instituted a private bounty targeting abortion (SB 8) and then said they couldn't be sued to stop the law because *they* weren't the ones carrying out the law. Eventually this went to the Supreme Court who narrowly agreed. However the law could still be challenged by anyone who became involved in an actual case under it.
Then Dobbs came down and rendered the whole thing moot and as a result there was never a final finding that the scheme was unconstitutional. But when the Supreme Court was giving Texas a win on standing they also expressed a great deal a skepticism that Texas would win if they decided based on the merits. This probably explains why nobody rushed to pass more laws like it.
I'm sorry that I can't think of examples off-hand, but I feel like I've heard of several laws that were at least proposed since Dobbs that worked on a similar basis. If my memory is accurate, at least a few have been passed on both sides of the aisles.
The problem, as Jeff mentions, is that even though such a law would prevent someone from contesting the law early, it doesn't prevent contesting the law entirely. The original purpose in Texas was more to dissuade people from getting abortions, not that it would ultimately hold up in court. Previous laws were insufficient for this purpose because even lower courts could quickly and easily toss an anti-abortion law under RvW and Casey.
These newer laws that I'm thinking about were not trying to solve the same problem (keeping a law on the books longer) but instead about using governmental resources. By giving private citizens a cause of action, this new type of law uses the same legal technology, but to a different purpose - specifically to enforce laws that take too many resources for the government to enforce on their own. If a negative of a law is "how could we ever enforce it?" then this type of enforcement mechanism is a good fit, at least in theory. I did think of an example that's actually been in place for a number of years - whistleblower protection laws that pay the whistleblower in the event that the government gets a fine out of the organization.
Texas didn't do anything innovative, is the fundamental problem; this particular legal strategy was developed back in the 60s or 70s by, IIRC, Ralph Nader (or at least he used it heavily, if I recall the person correctly).
It has been used extensively by the Federal Government since then, being employed for things like environmentalism and disability rights.
Normally, a government ban on something is enforced by some government agency. If someone who would be subject to the ban thinks it would be unconstitutional, they can file a lawsuit against the agency, and a judge can issue an order to the agency not to enforce the ban.
Texas tried to get clever by passing a law, SB 8, that banned abortion but specifically prohibited any government agency from enforcing it: instead, it could be enforced by any member of the public filing a lawsuit against the abortion-facilitating people. The idea was this way people couldn’t preemptively challenge it in court: suing the government wouldn’t help, since the government wasn’t doing anything. As it happened, the Supreme Court decided that abortion wasn’t constitutionally protected not long afterwards, so Texas was able to just regulate abortion directly without needing to take advantage of this dodge.
Some politicians in Democratic states made some noise about passing their own version that would infringe rights that conservatives like, but as far as I know they never went anywhere much. (The most developed, as I recall, was a proposal for guns in California.)
For whatever it’s worth, while the idea is clever, I don’t think that it actually would have worked, and in my view the abortion-rights freakout was entirely unwarranted.
Is anyone else increasingly interested in cybernetics as serious foundation to build a modern political philosophy?
The existing/traditional ideologies are too rigid and too slow to adapt. Yeah, I get the risks of over reliance on data, dehumanizing citizens, and rule by a technical elite. BUT don’t we already have those things happening. My current thought is the problem lies in the system not being built around real-time data and feedback, which turns its use into an exploitation lever rather than a tool for responsive government.
The ham-fisted swaying during election season highlights how poorly incumbents handle feedback loops—namely, how weak their predictive modeling really is.
I probably got the info from that exact same source, lol.
Yeah, I suspect communists and libertarians both kind of want a machine to mediate human society. Personally I think that's a fair way off and it's just easier for humans to do it for now.
Very funny! Yeah, it’s totally a fair way off which is why it’s worth thinking about now. I want to get people thinking about the nuance that humans are always doing it, it’s just whether we acknowledge that at the outset. So many articles talk about algorithms and AI as these kind of other beings, but they aren’t. They are the product of human intervention.
Some people seem to believe that Allende's Chile was one spreadsheet app away from Singularity. I guess this is one of those things the future generations may want to run some simulations of. I am skeptical because e.g. Venezuela today can make as many spreadsheets as they want; your average desktop PC probably has more computing power than Allende could get in a decade; and yet we don't see any of the promised miracles. Most likely, the problems of socialist economies are caused by something other than the lack of spreadsheets.
Spreadsheets would probably work in some kind of RPG universe where the production methods are fixed, and you can always transform "2 units of leather + 1 unit of rope" into 1 pair of shoes. Then you can use some algorithm to find the optimum production. In real life, not all units of rope are created equal (using a lousy rope will make your produced shoes fall apart), and new inventions happen (things that do not yet have a place in your spreadsheets, but require materials that do).
FWIW, this prompts me to mention the book Complexity and the Art of Public Policy by Colander and Kupers. It harkens back to observation of Scottish philosopher Adam Ferguson in 1767 that there is a category of things that are the result of human action, but not of human design. A quote from the book:
“… we believe that this complexity pattern--- that nonlinear systems can self-organize—is an important pattern for all policy makers to adopt and keep in the back of their minds. Complex systems are continually in the process of organizing themselves, and the way that process works is at odds with attempts to control systems. At best complex systems can be influenced—not controlled.”
Excellent! Added that book to my reading list. Much appreciated.
Conceptually very powerful to be thoughtful of the difference of influence v control. This is exactly what is resonating about cybernetics and the idea that we take feedback loops seriously at the outset.
It had a lot of the anti renewable talking points that ads wrong or disingenuous - subsidies, the output of solar panels on cloudy days , and an argument that’s I didn’t get about solar being cheap because fossil fuel is available as a backup.
However Scott introduced it as a post about the limitations of batteries so let’s stick to that.
Phil said that the “ A lead-acid battery stores about 35 watt-hours per kg. So we'd need about 2,294 billion kg of lead for batteries. There are only about 95 billion kg of lead left on Earth that we know of … The situation with lithium is much worse”.
The problem there is that lead acid batteries aren’t used on the grid or are used very sparingly as an old technology (Chat GPT said not at all).
So even if Phil is right that there’s a similar or worse problem with lithium why even mention an old technology that is not being used. It’s like saying “the AI revolution is dead because there isn’t enough glass in the world to power vacuum tubes needed for GPUs, and silicon isn’t much better” - and then doing the workings out for vacuum tubes.
Afaict (not an expert) iron air batteries solve these issues (at the cost of being heavier and somewhat less efficient, which aren't blockers for grid storage), but most of what I can find on them is by pro-environmentalist groups who may be optimistic.
You're right - lead acid batteries are not the right tool for storing energy for the grid. I'm not entirely convinced by the maths (which is only sketched). But looking at energy densities for Li batteries, you get at least 100 times the energy storage as with lead-acid, seemingly with comparable global supplies of lithium as lead. It seems sodium ion batteries are in development, and obviously there's no issue with supply there.
The idea that you would want to store 24 hours energy supply with batteries is also a bit unrealistic. Our patterns of power use developed with a grid which supplies power at constant levels 24 hours a day. Going forward one would expect there to still be some conventional generation, both fossil fuel and nuclear, but with more variability in supply. Grids will need to be upgraded quite a bit, and we'll move increasingly in the direction of variable pricing to encourage consumers to use electricity when it's abundant and discourage use when it's not. Local power storage will play a part too: I'm looking at solar right now, with the understanding that I could be basically off-grid April to September.
There's an underlying point which is valid, but the argument used to get there doesn't hold up to scrutiny. We're not going to maintain existing usage patterns while building endless battery farms to move forward. But it's true that existing battery technology isn't up to what we will require.
The “cloudy day” argument was wrong anyway as solar panels do produce on cloudy days. However worrying about cloudiness isn’t really that important for large grids, there’s never going to be a day in the US which is full cloud cover across the country.
And while the panels can go anywhere, particularly if we are talking rooftops, the large solar plants will go in the places where cloud cover is low. Night is a different thing.
> Going forward one would expect there to still be some conventional generation, both fossil fuel and nuclear, but with more variability in supply.
Net zero, isn’t zero - there will be fossil fuels so it involves carbon extraction, which to most people isn’t clear. For that reason don’t worry about the doomster arguments that we will have runaway heating even if we get to net zero, because if net zero is possible, net negative is possible.
How bad would the transmission losses be if we went halfway round the Earth? It's always daytime somewhere. Is there any combination of costs where this might make sense?
Agreed. It is an elegant idea (but I don't know the numbers - undersea massive power transmission sounds pricey...). Unfortunately, an additional problem is that transmission lines are intrinsically an extended, soft, hard-to-defend target, and the world has a lot of belligerent states and non-state actors...
It's always sunny somewhere, but does that mean you're building solar panels everywhere? The cost of energy from solar will definitely go up if you have to dig deep into the worst places to build solar panels just to have reliable production. I think the "unlikely but not so unlikely you can just ignore it scenario" is something like a week of unusually cloudy weather in the region that normally produces most of your solar power, leading to substantially less production. Solar still produces power on cloudy days but it's much less (google indicates 1/10 to 1/4 as much as usual). Maybe this sort of thing is much less likely than I'm giving it credit for as solar generation is more geographically dispersed, but I'm not sure "it's always sunny somewhere" is a sufficient answer.
It may be sunny everywhere, but it's not cost-feasible to have panels everywhere. So, yeah. Not only that, but the further you go from the equator, the less efficient those panels are going to be, unless you tilt them, which in turn means spacing them out of each other's shadows.
I feel like there *might* be a decent argument for putting arrays in the Sahara, presumably after paying Algeria, Libya, Mali, Chad, et al. a bargain price for land rights. I don't expect them to power the entire world, but they might put a big dent in the demand, and that might be enough.
I've long wondered what the practical hurdles actually are. I could imagine a few - transport, distance from supply hubs, distance from customers, political instability. But I don't have numbers for any of those.
> The cost of energy from solar will definitely go up if you have to dig deep into the worst places to build solar panels just to have reliable production.
We were specifically talking about the US in this sub thread, since the guy I was responding to was arguing against batteries in the US. I’m not a fan of Europe depending on the Sahara for its energy but wind power is a better bet for Europe anyway.
Solar in the US would generally, if positioned correctly, generate fairly reliable energy. During the day.
There is a company working on placing mirrors on satellites that will direct sunlight to solar plants during sunrise/sunset to extend their working time. So it could be "almost always sunny here in this one spot that we need it to be".
I would have to see some pretty serious math to suggest that it's possible to do this at a scale that would be meaningful without drastically increasing the cost of solar power.
Yeah no idea if it works or is economically feasible. Right now they've only done a proof of concept using a mirror on a balloon: https://www.reflectorbital.com
But there is a town located in a valley in switzerland (I think? norway?) that uses a mirror to reflect sun light over a mountain to give some relief from early sunsets in winter. No solar generation just vibes(?)
Why do you imagine energy *density* is the right metric for grid storage? The batteries are going to spend their entire service life sitting on the same concrete foundation; why does it matter what they *weigh*?
If a megawatt-hour of lead-acid battery storage weighs ten times as much as an MW-h of lithium-ion, but costs half as much, then lead-acid is the winning strategy.
I'm not actually sure if that's the case; the math gets tricky when we're talking about total life-cycle cost. And there are technologies other than lead-acid and lithium-ion to consider. But I see way too much hype about lithium-ion batteries with way too little explanation of why they're the right answer. Lithium-ion batteries usually *are* the right answer if you need to *carry* them somewhere, because of the energy density. But that doesn't apply here.
They were referring to whether it was possible to have sufficient battery storage with the resources viably available to us. If there’s too little lead (94/2.3k billion kg) then it doesn’t matter if it’s more economical per joule - we can’t do it. If lithium has lower total mass requirements due to energy density, then it’s possible to do at all (1/100 the required mass at the same viable material leads to 94 billion kg available and only needing 23).
>It seems sodium ion batteries are in development, and obviously there's no issue with supply there.
What is the oxidizer in this chemistry? I'm fond of sodium/sulfur, since _both_ elements are very abundant - but there are some problems with side reactions producing polysulfides. I'm not sure what the status of trying to avoid that is.
>we'll move increasingly in the direction of variable pricing to encourage consumers to use electricity when it's abundant and discourage use when it's not.
_Yetch_! That puts the burden of grid management on households instead of on large organizations which are far better able to absorb the management costs, both physical and informational.
I'm no expert on battery chemistry. I'm optimistic that smart people will figure it out.
As much power as you want when you want it is something we're used to because we don't see the costs it imposes. I'm suggesting that the grid management company needs to move from seeing its job as matching supply to whatever demand happens to exist, to shaping the demand to the supply.
It needn't be at the level of households. Although, e.g. my hot water is heated 2-4am, in a tank that's lagged to keep it warm for at least 24 hours. It wouldn't be a major inconvenience to me for the grid management to determine when this kicked in to better balance the load, for example. I'm not suggesting they tell me what time I can boil the kettle for tea.
>I'm suggesting that the grid management company needs to move from seeing its job as matching supply to whatever demand happens to exist, to shaping the demand to the supply.
This makes the power supply less useful to households. The degree of enshitification depends on how onerous the "shaping demand" is, but it _is_ enshitification.
I don't know about that. Ireland has a good number of power suppliers now, and each one offers multiple tariffs. It's up to the consumer to find the one that works best for them. E.g. I can choose something like a constant 30c per kWh charge or a 45c daytime but *free* power in the period 12am-4am. I'm told a good number fo people with solar take the later tariff, charge their batteries for free overnight (for the morning draw) and can reduce their power bill to a tiny fraction of mine. Is that enshitification, or just the free market?
>I'm told a good number of people with solar take the later tariff, charge their batteries for free overnight
implies that those people are:
- paying the capital costs for solar
- paying the capital costs for their batteries
- maintaining their solar
- maintaining their batteries
- setting up the proper timing (seasonally dependent) for switching between charging and discharging their batteries, and switching back and forth between solar, charging from the grid, discharging into household consumption
- watching carefully if the tariffs change, reconfiguring the whole thing in response to tariff changes, and possibly being left with a "stranded asset" if the batteries stop being worthwhile.
Now, I could see an argument for distributed solar (though one of the comments in this post said that rooftop solar cost about 3X what a solar farm costs). Solar _has_ to occupy an area proportional to the power it supplies. In a built-up area with scarce land, you either put solar on rooftops or you transmit it from somewhere else and eat the resistive transmission losses.
For batteries, though, I don't think this makes sense. One can put the batteries in a cube, with a footprint scaling as power^(2/3), and you get all the economies of scale from having a _small_ number of people who are professionally skilled at the care and feeding of the battery cube, which can serve a huge number of customers. There are also economies of scale in constructing the batteries themselves. The electrodes just scale with power, but all of the packaging, power conditioning, monitoring, etc. has economies of scale.
> Yetch_! That puts the burden of grid management on households instead of on large organizations which are far better able to absorb the management costs, both physical and informational.
The only burden on households is to decide to charge the car, or do the washing, or even increase the heating or air conditioning when the electricity is cheaper. And no doubt there will be apps for that.
>The only burden on households is to decide to charge the car, or do the washing, or even increase the heating or air conditioning when the electricity is cheaper.
Those are not minor. A washing machine needs to be manually loaded and unloaded before and after its use. Constraining when it can be used constrains when the person using it has to be available for those tasks (there is _some_ flexibility if one pre-loads it - but leaving wet clothes sitting in the machine afterwards is a bad idea.)
>And no doubt there will be apps for that.
The app cannot load or unload the washer, nor ensure that the car is not needed for transportation at the time the power is cheap.
These burdens are indeed enshitification, as I wrote earlier.
Yes leaving wet clothes is a mistake but the idea is to load up before and the washer and dryer will start when on time. People will know when the cheap electricity is available a few days in advance. With solar that will be a lot of the day in summer, with wind it can be anytime day or night. There’s a lot of time to charge a car.
Nobody will be forced to do this, but many people will.
Many Thanks! To put it another way: If e.g. solar was the dominant energy source and electricity prices are set much higher at night, the situation approaches one of rolling blackouts, though not quite as severe. I'd rather have the utilities maintain battery farms, or use nuclear for baseline loads.
Even with traditional power generation we use power storage. Usually pumping water to high reservoirs then letting it flow through generators when more energy is needed.
That kind of storage isn't great at adapting quickly but allows you to even out peaks and valleys while a smaller amount of batteries can be used for unexpected high demand.
Any argument of the form "X wont happen because we will run out of Y natural resource" has proven wrong over and over again for the past 150 years. Oil, nitrogen, food, etc. Humanity always finds a way to overcome the limitation.
You've heard of the utility monster. Now get ready for... the Deontology Monster!
It crawled from the depths of hell, answering the call of a ritual which has entrusted upon it the summoner's moral code. It then proceeded to follow the code to the letter, but in a manner which wreaked havoc upon our world.
It put itself in situations where it could save many people, but only saved those who were destined to become murderers. It found people who could be aggravated into assaulting it and then killed them in self-defense. It taught the values of liberty and progress to members of fundamentalist households, selecting those most likely to get ostracized and harassed by their family. It initiated insurrections against evil tyrants and replaced them by benevolent rulers, but only ones whose successors were even worse than the previous dynasty.
Can a creature like this be called good in any meaningful sense? Is there any true deontologist, real or hypothetical, who can summon it safely? Find out this Halloween in - The Deontology Monster!
Could you be clearer about what specific deontological rules the monster is following in your examples? The only one that's clear is killing in self-defense; I don't understand the "rule" in any of the other cases. So evaluating this argument is rather difficult.
I admit this hangs together better than I thought. I want to ask if the creature is choosing on the basis of these consequences, or if it’s just really unlucky. If the former, lots of the moral rules I have seen described would say that otherwise-good acts done from evil motives lose some or all of their good character. But those rules are virtue-based. I’d be interested to see if any self-described deontologist writers thought people’s motives were relevant to moral rules.
I asked Claude Sonnet to riff on Unsong characters as various flavors of moral monsters inspired by your comment:
"The Comet King seems closest to a Deontology Monster, though with elements of the others. His core characteristic is his absolute commitment to duties and moral law, particularly the imperative to stop Hell's torture of souls. He embodies the terrifying logic of pure moral duty taken to its ultimate conclusion - he's willing to sacrifice his own happiness, his marriage to Robin, and potentially the welfare of living humans in service of what he sees as an absolute moral imperative to end Hell's torments.
What makes him particularly deontology-monster-like is that his actions follow from seemingly valid moral reasoning: If we have a duty to prevent torture and suffering where we can, and Hell represents infinite torture, then any finite sacrifice would be justified to stop it. His logic is cold but hard to refute within a deontological framework. Like a true Deontology Monster, he reveals how universal moral duties, taken absolutely seriously, can lead to seemingly monstrous conclusions.
Other characters that show "ethical monster" traits:
Uriel has elements of a Virtue Ethics Monster - his dedication to truth and order is so absolute that it becomes destructive. His virtues of precision, dedication to cosmic order, and commitment to mathematical truth are scaled up to angelic levels where they become almost incomprehensible and potentially catastrophic from a human perspective.
Thamiel could be seen as a twisted Utility Monster - he argues that he maximizes some form of "utility" by ensuring sinners receive their just punishment, though this is more of a perversion of utilitarian logic rather than a pure utility monster case.
The most interesting case might be Metatron, who in some ways combines aspects of all three monsters. His divine nature and connection to pure abstract good creates new moral duties (Deontology Monster), manifests perfect virtues that become alien to human understanding (Virtue Ethics Monster), and generates overwhelming divine utility that could theoretically outweigh human concerns (Utility Monster).
The broader theme in Unsong seems to be how transcendent beings - whether they're the Comet King, angels, or God himself - can embody moral principles taken to such extreme levels that they become almost incomprehensible or horrifying from a human perspective while still being internally consistent. This maps well onto the "monster" framework where ethical principles that work at human scales break down when pushed to absolute extremes."
The fact that any deontological rulebook that is smaller than the universe will eventually recommend something bad is something the deontologists presumably already know. This creature isn't a counterpart to a utility monster, it's a counterpart to the alignment problem.
There is a counterpart to utility monsters in deontological systems that actually get used: god. "What if there is a creature that enjoys so much that the greatest good becomes whatever it wants?" -> "What if something wrote all the rules, such that the right thing to do became whatever it wants?"
Ha ha ha, you think you can foil The Deontology Monster?
Matthew 10:
"21 Brother will deliver brother over to death, and the father his child, and children will rise against parents and have them put to death, 22 and you will be hated by all for my name's sake. But the one who endures to the end will be saved.
34 “Do not think that I have come to bring peace to the earth. I have not come to bring peace, but a sword. 35 For I have come to set a man against his father, and a daughter against her mother, and a daughter-in-law against her mother-in-law. 36 And a person's enemies will be those of his own household. 37 Whoever loves father or mother more than me is not worthy of me, and whoever loves son or daughter more than me is not worthy of me. 38 And whoever does not take his cross and follow me is not worthy of me. 39 Whoever finds his life will lose it, and whoever loses his life for my sake will find it."
Luke 12:
"49 “I came to cast fire on the earth, and would that it were already kindled! 50 I have a baptism to be baptized with, and how great is my distress until it is accomplished! 51 Do you think that I have come to give peace on earth? No, I tell you, but rather division. 52 For from now on in one house there will be five divided, three against two and two against three. 53 They will be divided, father against son and son against father, mother against daughter and daughter against mother, mother-in-law against her daughter-in-law and daughter-in-law against mother-in-law.”
Revelation 19:
"11 Then I saw heaven opened, and behold, a white horse! The one sitting on it is called Faithful and True, and in righteousness he judges and makes war. 12 His eyes are like a flame of fire, and on his head are many diadems, and he has a name written that no one knows but himself. 13 He is clothed in a robe dipped in blood, and the name by which he is called is The Word of God. 14 And the armies of heaven, arrayed in fine linen, white and pure, were following him on white horses. 15 From his mouth comes a sharp sword with which to strike down the nations, and he will rule them with a rod of iron. He will tread the winepress of the fury of the wrath of God the Almighty. 16 On his robe and on his thigh he has a name written, King of kings and Lord of lords.
17 Then I saw an angel standing in the sun, and with a loud voice he called to all the birds that fly directly overhead, “Come, gather for the great supper of God, 18 to eat the flesh of kings, the flesh of captains, the flesh of mighty men, the flesh of horses and their riders, and the flesh of all men, both free and slave, both small and great.”
I'm not sure that really is analogous. The utility monster is supposed to demonstrate the perverseness of maximizing utility. The deontology monster presumably should demonstrate the perverseness of maximizing rule-following.
So it seems like a deontology monster would be more like, "it nobly goes out of its way to save people from would-be murderers, following the principle of stopping murderers being good, but only when it knows those people will themselves become would-be murderers that it can save more victims from. It is lethally effective in self-defense, but only when it knows its attacker is part of a large clan which will attempt to avenge that death on the deontology monster and 'forcing' it to defend itself more. It is diligent in working for reform and revolution against evil regimes, but only assisting those movements with poorly chosen policies or leaders which would eventually turn out to be even more evil and in need of reform/revolution." It carefully follows all its deontological rules like saving victims or killing only in self-defense or fighting oppressive regimes, but ensuring that it will follow those rules as much as possible in the future.
I think a deontologist would answer something like 'a perfect deontology with a perfect set of rules would not be exploitable by such a monster. Any human construction of the rules *is* exploitable, but *any* human-enumerated moral system is exploitable in some way for much the same reasons.'
The most famous form of deontology is Kantism, and Kant's fundamental principle is famously that "nothing in the world—indeed nothing even beyond the world—can possibly be conceived which could be called good without qualification except a good will." The Deontology Monster clearly doesn't have a good will, by your construction, because its will is to increase murder, assault, ostracism, harassment, and dictatorship.
> It put itself in situations where it could save many people, but only saved those who were destined to become murderers
If it can see the future, then the creature's threshold for murder suddenly expands massively. No ordinary person can say, "If I do X, then Y will incidentally murder Z, and therefore doing X is murdering Z" because we treat individuals as independent agents who can make whatever decisions they like, not as objects that can be triggered to act as easily as nerves control muscles and muscles control weapons. Additionally, for ordinary people, intention doesn't always match outcomes. If the monster has absolute prescience, then all actions leading to deaths become acts of murder. So no, it's not Good. In fact, I think it could be argued that basically any entity capable of seeing the future is (under deontology) extremely Evil.
+1 for this. Everything else seems difficult to define, like project 2025 - given the very obvious ideological slant of almost all institutions, far-reaching personal changes by themselves are quite reasonable and whether it's authoritarian depends entirely on whether the replacements actually are Trump/republican loyalists or just simply not democrat loyalists.
That seems like way too narrow a metric. As a ludicrous example, if he ruled by fiat, suspended civil liberties, but appointed Don Jr to succeed him, that wouldn't count.
I'm really not sure this is operationalized enough for a bet to resolve cleanly. Suppose Trump spends 2027 going around saying "I want another term, the American people want me for another term, but the haters and losers say I'm disqualified on a technicality. I guess we'll see what happens." Then he never attempts to file his candidacy or take any official action related to running. I don't think this is terribly far fetched, and I don't see whether this counts as attempting to retain power.
>Conditional on him winning this election, will he attempt to retain power in 2028?
As an index of Trump's intentions, this is kind-of tricky given his age. Maybe, as a prediction market it should be conditional on both Trump winning the election _and_ on Trump being in reasonably good heath (more coherent than Biden is today) in 2028...
>That’s the best one I can think of. It’s quite likely he will appoint MAGA types to cabinet positions, move the federal bureaucracy significantly right, and centralise executive power.
Sigh. Given the power of the Federal bureaucracy, I'd rather see it moved right by having the heads of (cabinet level?) departments elected by the voters. Yeah, this would take a constitutional amendment, and we don't seem to do that any more... :-(
Why is the list in project 2025 especially authoritarian? Most of what I've skimmed through in the project 2025 doc is pretty vanilla republican stuff (some good, some bad). I'm sure there's some worse stuff I missed (it's like a thousand pages and I only skimmed), but the median point in it doesn't seem particularly authoritarian.
Having read the entire plan, for me it's the unitary-executive stuff that is most troubling: all federal agencies being placed under the POTUS's direct personal control, specifically including eliminating the guardrails which limit a president's ability to hire/fire the top officials in the Dept. of Justice. That would result in a POTUS being able to (just one example of many) order the head of the Census Bureau to have the census come out a certain way and just keep firing them until he found the one willing to do it. Combined with the recent SCOTUS ruling on presidential immunities, the POTUS would be literally "above the law" in a way that the US has never seen before and the Framers explicitly wanted to prevent.
Second most troubling for me, though I could be persuaded to move it up actually, is that as a secular person who does not think government should be carrying out any particular religion's worldview or precepts, Project2025 repeatedly laughs in my face. "Biblicly based" comes up a lot.
If you are strongly pro-choice, a whole long section of Project2025 is a complete nightmare. It tries to deny that it's taking steps towards a national ban but no sensible person would believe that (and the anti-abortion groups sure don't).
Internet neutrality would be gone, with tech companies that in any way allow the transmission of pornography [to be defined in a biblically-based manner] being summarily shut down on a "prove your innocence" basis.
Those are the lowlights for me personally, YMMV of course. For one counter example, while I disagree with the plan's approach to immigrants and immigration that's mostly within the bounds of policy disagreement. Though creating a new national superagency with plenary prosecutor+judge powers at all borders, not subject to legislative control, does reinforce my negative reaction to the unitary-executive stuff.
My Project2025 summary to my wife was, "These guys basically want us to have a Christian Bible-enforcing king", and not a figurehead constitutional monarch like the UK has. Whether that counts as fascist is unclear to me (political science definitions of that word are pretty squishy), but it certainly is authoritarian.
>Your best bet is to just choose some organization that rates nations for how democratic they are and bet on whether they'll pronounce him a dictator within 4 years. Not necessarily reflective of reality, but it's probably better than nothing.
Difficulty: All the organizations I'm aware of that do that are based in the US or benefit significantly from US money, making it very difficult for them to declare the US a dictatorship even if its true.
This is probably the best option. Esp since it has a variety of different indices (Deliberative Democracy Index; Egalitarian Democracy Index; Electoral Democracy Index; Liberal Democracy Index; Participatory Democracy Index) and also a large number of subcomponents.
My only concern is that the indices dropped quite a bit in the first Trump term, so there might be some bias. The drop seemed to be a bit more than merited, at least at the eyeball level. Eg Deliberative Democracy Index was the highest ever in 2015, but in 2019 was the lowest since 1971. (But at least it didnt behave like that silly Polity score did) Perhaps averaging a number of similar indexes would result in something with less volatility? Or perhaps just choosing a low enough score as the threshold would be sufficient
So how do you rate the Democracy of "we're not going to give people the choice to choose who runs, we're telling them this is the candidate, and in some states we try to have a non-aligned candidate removed from the ballot, but trust us, we're the defenders of democracy"?
A one-party government is generally not considered very democratic, but in some American states that's the rule (Republicans shouldn't even bother running, the choice comes down to Democrat A or Democrat B). Trying to get candidates taken off the ballot because you fear people may exercise their choice to vote for them instead of Your Guy/Gal isn't very democratic, either.
If we're going to do "Is the USA sliding into becoming a non-democracy?" then apply it to Tweedledum as well as Tweedledee.
Instead of waging the culture war at every opportunity, why don't you go to the V-Dem website and look at their methodology? That might answer your question as to whether that would effect the ratings of a given country. I would start at pp 94-95 of their codebook (https://www.v-dem.net/static/website/img/refs/codebookv111.pdf) re the sub-component "Candidate selection-national/local". Assuming, that is, that you are actually interested in assisting Scott in his inquiry.
gd, popping up with "Is X the most evil person?" is partisan, whatever way you slice it. Maybe X is, but Y doesn't seem to be 100% spotless either, and if you're really worried about the possibility, then you should consider all relevant persons, not just "I don't like X so I'm saying X is the most evil".
How do I know this random website you recommend is neutral and not partisan?
EDIT: Okay, had a look at the code book. University of Gothenburg, so that gains points from me for being less likely to be partisan on American grounds.
0: No. No-party or single-party and there is no meaningful competition (includes situations where a few parties are legal but they are all de facto controlled by the dominant party).
1: Not really. No-party or single-party (defined as above) but multiple candidates from the same party and/or independents contest legislative seats or the presidency.
2: Constrained. At least one real opposition party is allowed to contest but competition is highly constrained — legally or informally.
3: Almost. Elections are multiparty in principle but either one main opposition party is prevented (de jure or de facto) from contesting, or conditions such as civil unrest (excluding natural disasters) prevent competition in a portion of the territory.
4: Yes. Elections are multiparty, even though a few marginal parties may not be permitted to contest (e.g. far-right/left extremist parties, anti-democratic religious or ethnic parties).
Scale: Ordinal, converted to interval by the measurement model.
Data release: 1-11.
Cross-coder aggregation: Bayesian item response theory measurement model (see V-Dem Methodology).
Date specific: Election-specific dates (v2eltype).
Citation: Pemstein et al. (2021, V-Dem Working Paper Series 2021:21); V-Dem Codebook (see suggested citation at the top of this document).
Years: 1789-2020"
Seems like "2 - Constrained" describes efforts to get other party candidates off the ballot, do you agree? Now remind me, who tried doing that?
Again, you seem to want to wage the culture war, in a thread about how to help Scott operationalize certain phenomena for the purposes of developing a prediction market question. Not cool.
Please try to engage in good faith. One of the huge strengths of ACX is the loose moderation coupled with a community that actually tries to communicate, and by defecting on the communication norms, in the long term you mostly just push towards more restrictive moderation.
>Republicans shouldn't even bother running, the choice comes down to Democrat A or Democrat B
This is a silly argument - it's not like anyone bans Republicans from running, those states are just far enough left that Republicans aren't popular. There are also states that are far enough right that Democrats don't really have a chance. There's no conspiracy here, it's just that the 50 states have a wide enough range of positions that you can't have a nation-wide party that aligns with all of them at once.
By your logic, Vermin Supreme is the most oppressed candidate of all time, because he's never had a realistic chance of winning, ever.
That isn't necessarily a bad sign. Two boring candidates with centrist policies and sensible advisors campaigning for the votes of a sleepy electorate is a better outcome than Nazis vs Communists with record turnout.
I agree that my suggestion isn’t perfect. As I saw it, if people feel democracy is threatened, they might be more likely to vote if they think it’s a necessary to disrupt whatever plans Trump might enact (e.g. to soften the impact of him changing eligibility rules to favour his own party). I think most of the people engaged enough to do that would be voting against him already anyway, but I’m not sure.
With that said, I think I’d probably see both parties choosing centrists as being more likely if both parties agreed Trump had gone too far. It resembles the fact Biden campaigned on being centrist, and the fact that Australia’s main party leaders after the Rudd-Gillard-Rudd-Abbott-Turnbull fracases have been more boring than the ones before that period. If Trump is unpopular for non-authoritarian reasons, I’d expect the Democrats to move left by a median voter argument, since their base will feel vindicated that the other side is bad.
Many Thanks! I'm not sure if I'm reading the sign of your metric correctly.
>As I saw it, if people feel democracy is threatened, they might be more likely to vote if they think it’s a necessary to disrupt whatever plans Trump might enact
So are you saying more turnout is an indicator of people feeling more threatened (perhaps re democracy, perhaps re other parameters)? That sounds plausible, and is analogous to the Nazis vs Communists situation in pre-WWII Germany.
( FWIW, I, personally, see both Trump (with Jan 6th) and Harris (with her proposal to "regulate" [censor] speech as being threats to a democracy with civil liberties. Regrettably, they can't both lose (barring some really improbable events). )
Not a perfect metric, but "will he be denounced as authoritarian by (specific, named) people currently on his side (e.g. the republican house speaker or "at least X members of his starting cabinet ") is an interesting one.
(Advantages: more flexible about catching directionally vague things. Disadvantages: can have both false negatives (he could go authoritarian and keep their loyalty) and negatives (he could lose popularity and get called out despite not being authoritarian).
I had an interesting back-and-forth with chat gpt about this, by asking it to build a score card for evalutating democracy/authoritarianism, then feeding it examples of eroded democracies and finally about which Trump's proposals align with stuff that happened in those eroded democracies.
My main takeaway for the most telling signs:
- Politicize the DOJ and the FBI
- Schedule F proposal to make it easy to replace civil servants
- Changing electoral laws unilaterally by invoking the so-called "2020 rigged election"
The media pose an interesting question. Except for Fox, the media seems very close to acting as a Democratic Party organ. Amongst other problems, it is a miserably bad watchdog when watching Democratic politicians.
Any suggestions for balancing this? Require that the GOP/Dem ratio in the reporter pool be within X% of the ratio in the general population? Something else?
I think the curtailing of media metric applies mostly to "official" or "established" media, and its leaning is irrelevant, the idea is to show that an authoritarian regime doesn't accept criticism and oppositon. In other platforms, like social media, it's a tricky metric. Example: a lot of non-authoritarian regimes have some limits for what can be said in such platforms, at least more limits than in the US.
>I think the curtailing of media metric applies mostly to "official" or "established" media, and _its leaning is irrelevant_, the idea is to show that an authoritarian regime doesn't accept criticism and oppositon.
[emphasis added]
I disagree strongly. When a media channel leans strongly towards the party with power (as the New York Times does with the Biden/Harris administration), one can't _tell_ whether the government would accept criticism, because the media isn't delivering criticism. Now, right at the moment, the existence and toleration of Fox _is_ an indication of the current government's acceptance of criticism, but that network is largely alone in this.
>In other platforms, like social media, it's a tricky metric. Example: a lot of non-authoritarian regimes have some limits for what can be said in such platforms, at least more limits than in the US.
I agree that First Amendment protections (which I support) are stronger in the US than in most other nations. "Non-authoritarian" is a very fuzzy category. Great Britain has imprisoned people for politically incorrect comments on social media. I view that as evidence that they are sliding into authoritarianism.
Most consolidated democracies have hate speech laws, actually the US is a notable exception; Norway for example ranks 9.8 in the democracy index and has hate speech laws, whereas the US ranks 7.8 and doesn't.
>Most consolidated democracies have hate speech laws, actually the US is a notable exception
And I _STRONGLY_ want it to remain an exception, with First Amendment protection for free speech.
I can't speak for Norway, but in the UK, "hate speech" laws have been used to attack people who e.g. merely note that transpeople retain their original chromosomes, or who merely call to "reject white guilt". I consider those uses of "hate speech" laws to be grotesque and sliding into authoritarianism.
One of the amusing things has been seeing the New York Times denounced as a right-wing organ and maybe, who knows, harbouring Fascist sympathies itself for being insufficiently denunciatory of Trump and even covering the news in an even-handed manner.
Many Thanks! That is bizarre... I get the NYT daily summary, and it is glaringly biased towards the Democratic Party. They pretty much phrased their campaign coverage from the point of view of "Let's look at this as an optimization problem in how can Trump be defeated" - as if that goal were a given.
Whether Trump has fired 10-20% of federal bureaucrats by 2028.
It's what Curtis Yarvin wants him to do, which is a decent metric for how authoritarian someone is: (https://graymirror.substack.com/p/elon-versus-the-usg). I think it's also something Trump wants to do quite badly and it's what Project 2025 is about, which means it might actually happen. And I think it fits what came out of your Dictator Book Club; a hallmark of authoritarianism is destroying other power bases until only yours is left, whether that's a somewhat democratic Erdogan in Turkey wiping out the Kemalist military or Xi Jinping in authoritarian China using anti-corruption to wipe out rivals.
I'm not sure if this fits your frame though. I think a Trump who can fire significant amounts of the federal bureaucracy is pretty unavoidably more authoritarian but about half the country would see that as restoring democracy, not undermining it.
Also, more than happy to defer to someone more knowledgeable on the specific percent of the federal bureaucracy to fire or whether specific groups, like senior managers, are more important.
Julius Caesar, basically, although you'll find variations of this concept of mob democracy->tyrants/kings through Greek & Roman political thought.
The core concept is pretty much always the same. There's an entrenched, unpopular oligarchy. The general population is too uncoordinated to create/elect their own oligarchy so they give all their authority to a single individual, usually a successful general. That general then destroys the entrenched oligarchy but then has no rivals, there are no "check and balances" from other institutions because, well, those were all rotten and the people empowered a general to destroy them. Theoretically a truly great leader like Cincinnatus would then relinquish power and restore an oligarchic balance of power but in practice when guys like Sulla and Diocletian tried it, it failed.
Hey, man, Caesar isn't increasing centralized power like Diocletian here, building a huge bureaucracy, he's centralizing power like "Hey, we need to stop all these provincial governors and generals from murdering each other in brutal civil wars" a la Marius and Cinna and Sulla and Pompey and Caesar himself, plus all the Latin revolts. Constant, internal warfare between chief members of the government is, like, the primary feature of this period. The strong oligarchic government of the Gracchi period is in full disarray at this point.
If you don't understand this theory, awesome. If you do and you're just not convinced by it, cool, I'm not your guy.
It's because your model is lacking a little bit of nuance. Reducing government scope and size tends to mean reducing government scope and size *for the parts of the Government you think should be reduced*.
This is so sort of obvious that it's hard to see, and I feel like I'm insulting you by writing it but I don't mean it that way. The parts of whichever government processes that Trump and his associates like are unlikely to be reduced. But the parts they dislike are. This creates an unbalanced, adjusted power structure. Autocratic tendencies increase because though there may be abstract reduction in the size or scope or scale of "government", the individual component that represent a certain factional bias is likely to remain or even increase.
You can avoid this with a principled generalised stance of "reduce everything equally", but I think you will agree with me when I say that this is unlikely. I do not think that Trump (or Musk or whoever) have a principelled libertarian stance when they say they wish to cut the government excess. I think they have regulation they'd prefer not to have to deal with and so want to get rid of it.
To extend that example, consider something as relatively pedestrian as "Food inspection services". A relatively large component of government spending goes to managing food safety. You can therefore reduce a good amount of Government spending by reducing, dismissing or getting rid of those departments. But this is where we get into advanced political interactions that are hard to model out fully. As some utter toy numbers, not to be taken seriously, if there are 10000 Food Inspectors in the country and you cut headcount by 80% you are left with 2000. This sounds fantastic on paper if your goal is solely to "decrease the scale, scope and size of government".
If we start with 10000 Federal Agents and we end there as well but now we're running a 10000 Fed Agent / 2000 Food Inspector mix on the spectrum, the outcome of this will be total compliance with the mandate of "reduce government" but it will lead to a distinctly more autocratic process for the functioning of the government *that remains*. It might also, depending on size and scope and scale, have effects that are longer lasting. There could be some optimal amount of food safety inspectors per 1000 people, and by shifting the balance around, you then start having listeria outbreaks, e. coli, whatever. The government keeps trucking along, we have the same amount of guns / tanks / border patrol / telecommunications / whatever but the agency in charge of food inspection is now significantly compromised.
Again, toy model, but it illustrates the point. Then you have to write this out larger, to the fact that government is multipolar and very large. If you reduce IRS headcount and slash IRS funding you will get more tax fraud. This is unavoidable. There is a certain amount of tax fraud occuring every year, you need an unknown value of X IRS employees to detect and prosecute it, and those agents are now no longer available. We might get effeciency gains elswhere. I am positive we will. But there remains the basic fact that government mandates and taskforces and insitutitons need people to function.
Now we go back four steps.
You are a billionaire, and the Government ability to regulate food safety, the tax code, traffic regulations and let's throw in something like agricultural, whatever, is now significantly reduced because of massive budget reductions, downsizing and firing.
The private individual that is a billionaire remains a billionaire. A lot of people may benefit. Perhaps there is benefit to slashing the above departments. However, in a legal state operating by a lawful system, if the agencies involvedin regulation those law processes are knocked out of action, backlogged or unable to carry out a function, individual rule of individual free will and capability starts reigning supreme. We've veered over here to libertarianism. Which is a wonderful idea, and I support it fully, but it does not start here from first principles and a blank slate. The deck is already stacked - the McDonalds board of Directors no longer need fear surprise food inspections so they start slashing food safety requirements for their large corporation because the regulations that dictate what is or is not allowed are now not being updated and a corporation reasonably takes the steps the law mandates that it should, and here we are: the gradual push of individual incentive have now resulted in a world in which there is a somewhat increased chance your cheeseburger is going to make you sick, you have little recourse because you are not a billionaire, and we have transcended through "freedom from overzealous regulation" into "the individual free choice of individuals, outside of communal oversight, but mediated by the unofficial powerbrokers and movers and shakers that have accrued the means to shift the social balance".
There will then be push back and reorganisation and such-like. This is where it becomes useful to remember that quite often "regulation" or "zealous government agency", absolutely not an unthinking unalloyed good, still represents in the abstract and quite often in the specific the communally pooled ressources of the polity. Everyone gets together, pays XXP% in tax, and YYC% goes to Food Safety Inspection because the larger collective and commune decides that's a better idea for individual freedom. A million 1 buck donations is 1 million bucks, which offsets and checks the 1 million I inherited from my grandparents. That's how you get society.
There are politically reasonable theories that the natural state of man is freedom and that full freedom from meddling beaucratic middle managers will result in the exhultation of the human spirit.
There are also fairly decent counter-arguments that the historic trend, for all of human history, has been towards beaucratic middle managers, processes and laws being required to dictate, at times, the way people act to together to ensure some basic, underlying communal freedom. The shift away from feudalism is a shift towards a legal state where the various organs can meditate the power of strongmen.
So yes, decentralized power is by its nature more difficult to collect, and that's a fine truism. But the power here does not start decentralized. It's already been collected and pooled and adjusted. The playing board is not randomized. You can risk loosening the wolves by reducing the watchdogs. History - which you cite - is rife with examples of tension, horror and decay increasing because of the reduction of power in a centralized authority simply because without compliance from some central force, power becomes what people can project. A German Robber Baron rules the road by right of sword. A small town sheriff that ducks federal corruption investigation has pooled power already. Large multinational corporations that produce many a wonderful and fine good occassionally include ingredients in their food products that are deleterious to human health above certain values, and unless we want the "free market" to sort out the process of what food producer to risk buying from, we do need some basic rules.
Try not to forget here that if we veer in the other direction - towards freedom, reduction in government, lack of regulation, Do What You Can For There Are Less Laws, the friction that happens instead is the econonomic friction of the inefficiency processes of people dying. The free market will probably eventually sort out the issue of McDonalds Cheeseburgers making all their customers sick. Eventually, they'd have no more customers. This isn't a got'cha, it's just the basic logic of how markets work, but the path towards "have no customers" is paved with the millions of half poisoned burgers they, in this scenario, have sold on the way, half by intent, half by accident. That's not freedom. That's just endemic risk taking in every venture and a constant appeal to the hope that enough other people have suffered for the kinks to have been worked out. It doesn't even have to be malicious. Federal food safety standards as they are currently applied is ane evolving body of doctrine that comes from the hundreds of years that people have been killed by eating things. But you want to reduce the size of government. So you slash 80% of the FDA. And the IRS. And the DMV. And sure, Education too, why not.
So what's left, in this admittedly cooky example? Your personal millions, the soldiers you can buy with it and severely reduced pathways for various sorts of friction to be mediated by the consensus mechanism we call "society".
It works great if you already have all that power that never, ever starts as loosely distributed as the logic requires. And now some guy in Nevada is raising a militia and enforcing [the thing you don't like] on a large area. And we could deal with that, but we slashed 80% of the budget for all government services and now here we are.
In freedom, we found the old chains. History shows that centralized powers and processes tend to reduce the fractious friction of a far flung realm, least you invite in the horror that is individual warchiefs, bandleaders and strong men running the local area as how they see fit.
It won't be mad max, of course. I'm not talking "The apocalypse" here. Could just be the freedom of the free market to store your burgers at suboptimal temperatures because Food and Safety has 3 guys to inspect the entire country, and they'll never know, and if you get fined, you can just not pay and it'll never get enforced and . . . on and on.
It's not a simple linear relationship between "cut government" --> " avoid fascism". Rule by corporate forces and strong men is Fascism too, in the classical sense.
That's fair and well reasoned. I just happen to disagree quite a bit with your assessment though. You're conflating larger scale bits for smaller scale bits. Old man Benito in Italy broke up some monopolies and I'm with you there, but we call it Italian Fascism - or classical fascism! - because it's the classical model. It has a few more particularities than your dictionary definition. And it does in fact precisely lead us to a rule of corporate forces and strong men subsuming the larger will into autocratic, entrenched power interests.
They take a while to get there, I grant. There's a lot of "crushing external power blocks" on the way out there, because ideally you'd want the whole region under your control as a good autocrat and that's the thing I'm pointing out is the core of the issue. The struggle of a thousand sub-polities for scraps of power they can carve out because they have the possibility to do so looks a lot like anarchism and anarcho-tyranny but it might as well be marching bands of people in uniform and the rule of autocratic, despostic strongmen with particular preferences!
But that's all politial theory, and we could be here for hours slinging terms, so I'll call that I think we've probably just got different feels for what constitue the components of a fascist movement and leave it there.
But I will sidestep into your defintion, and raise a paralle point. You say that the left uses it so much that it glosses over what it actually means and boils the whole word down to "Icky" as some kind of unthinking reference to Bad Thing. Yeah, I'm with you that the leftist throw the word around a lot. I think the rightist do too. I also happen to think - and I'll doubletap this one to note it's just an opinion - that it is sometimes an accurate assessmet. By any particular definition, surely ". . . centralized autocracy or dictatorship, suppression of dissenting beliefs, and subordination of individual interests to the common good. . . " is the exact and compete political description of a platform that chants "lock them up", threatens to prosecute political opponents and fairly reliably threatens to arrest journalists, along with a few other parlances of the hour like kicking out the foreign immigrant hordes and promising that the nation comes first before all other things.
Is modern scale American Fascism going to look exactly like 20th Century Italy or Spain, or that other place, Georgia, that's the one? No. Of course not. But do I (as a random sample of a random observer) find it meaningfully accuate to describe stated policies of some political parties as "Fascistic"? Yes. Which is presumably why I do it, and I'm at least somewhat careful about the term. Maybe some hard core ravening leftists aren't (quite a lot of them aren't), but from my perspective the term gets thrown about because it fits.
So lets dance back to the larger point, and bear with me here, because it's a bit of a subtle one:
You might reasonably say that "nothing you/I described really points to an increasing centralized authority and subsuming the individual interest to the common good" in the examples above.
I say I disagree, because the process described is a general example of how such a thing happens over time. The people who would end up wanting to squish your interests for the "common good" or centralize authority are the people who step into the vacuum left by, in this hypothethical example, the government that we slashed by some double digit high end percentage. Or it's the natural process of what the government is, after it's been carried out. If, upon taking power, I replace civil servants with those bound to be me by personal loyalty, persecute my enemies, slash funding and components of large organs I disagree with, assume more direct power not balanced out by large organs I disagree with, and push whichever *personal* policies I might prefer, unopposed by a large civic substrate, well, there we are, "Fascistic" or "Authoritarian" are correct terms to use. This holds across all levels, from the random sheriff in some town pushing his weight around to presidents and pretenders elsewhere.
But the coy magic (And this is probably the roof of my real disagreement here, in a sort of generally opposed manner) is that I find your insistence on dictionary definitions of large political tendencies somewhat tendentious? Someone describes a system as "having some measure of ..." and the rejoinder then becomes a got'cha, ah ha, it does not fit the precise definition we are using in this case. We cannot use fascism because the definition I am using it to mean is not the definition you're using and that makes you the slightly snarker prankster.
There's *judo* going on here, and I think you'll know it enough to see it: You can decry "the leftists" for over using the term authoritarian or fascist, which rather than have a discussion about the political theories of the 20th century as it pertains to modern elections, instead becomes a way to turn the very invocation of a concept into a signifier that the people claiming something are to be disregarded for being somewhat hystronic. I think that's a bit too pat? Surely you'll have to at least *somewhat* agree that if I were to wander off now and spend 30 minutes collecting a bunch of excerpts of opinions expressed by leading political figures (democratic or republican, I shan't mind, though I have priors about which set is easier to find them from) then some of those excerpts will certainly have a *tinge* of the old Fascism. A bit of the Authoritarian creeping in. A gentle veneer of "an ultra nationalist form of government characterized by a centralized autocracy or dictatorship, suppression of dissenting beliefs, and subordination of individual interests to the common good".
Is it Fascist enough to institute an Italian style minimal wage and plans to enforce America-wide syndicate corporatism? Maybe not. Is it fascist enough to suggest we criminalize unions, strikes and labour movements? Possibly. Are we sometimes veering into a few little uncomfortable touches of moral purity projections, discussions about enforced gender roles, concerns about the corruption of the body politick at the hands of foreigners? Well, you tell me which group talks more about mass deportations, militarize camps and the need to squish individual elements for the need of the greater good. Well. Whether that's national stage or the governor of a state or the plutocrats of a small town, it's more or less the same fascism, and if you don't have enforcement of one style of government there will be alternatives and now we've gone all the way back around. You start slicing too far into the substate of "government" as we have it, and out the other end comes a lot of authoritarianism, fascism, tension and turmoil (And probably anarcho-tyranny too, and some communism communes and libertarian free holds and so on)
It's a complicated field, semantics and political development and societal opinions.
But I think you do a serious disservice to the notions of taking accusations of authoritarianism, or, heavens forfend, fascism, seriously simply because they're *also* sometimes mildly hysteric exaggerations of something where "Icky bad" would have sufficed. Sometimes they're true descriptions of real processes. But we hardly need the social aikido of subtly implying even the terms themselves have no applicability because they're washed out from overuse. Don't forget either - if things need to be leashed to the state or a movements idea of what's right, that's a lot easier if there's no governing overall structure that makes such compulsions harder. We're back to the fascist militia communes of upper Ohio marching in goosestep to enforce their will because somewher along the way, the processes we've spent two-hundred-years building to avoid all that happening got turn down.
You have to think at different scales for these things - otherwise the dictionary definition slide simply allows someone to gather power, crush popular dissent and centralize all power around their own set of preferred policies, but we can't call it XXXXXism because they're calling themselves National Socialists.
Anyway this all got a bit meandering, so I think we're done here.
Maybe split it into pieces, like one could be 'political appointees in DOJ increase from x to y'? Another one could be DoJ drops cases against trump and starts cases on prominent democrats, could have w one about interference in midterm elections, constitutional crisis, etc.
Could you write a question related to any kind of constitutional crisis?
There's a clear SC ruling that gets intentionally circumvented or ignored, for instance. Similarly, if there's a refusal to leave office at the end of the term.
Looking back over the past eight years, was Biden worse for democracy than Trump? It's a tricky question. Trump was worse in some ways, Biden was worse in others. And the ways that Trump was worse were the ways you might predict him to be worse (e.g. attacking the press) while the ways that Biden were worse were the ways you might predict him to be worse (e.g. by co-opting the press).
If you ask the above question then you're going to get a list of things that Trump is likely to do. What you want to do is also ask the question (independently, to a distinct but otherwise-identical audience) "How would you operationalise 'Harris will be an authoritarian' or 'Harris will undermine democracy' in a way suitable for a prediction market question?"
And then what you want to do is to apply the Harris standards to Trump and the Trump standards to Harris.
> in a way suitable for a prediction market question?
A meta comment - that Scott, who's no dummy and is into prediction markets, is asking for advice on this, is a sign of one the problems of prediction markets. Beyond asking who's going to win the World Series, coming up with a good prediction market question with good resolution criteria can be a Hard Problem, like writing good survey questions.
This year I will just get it in November together with the flu shot for convenience.
Last year there was a point in time when the vaccine against a new variant came out. (September or so in Europe? I don't quite remember.) I haven't followed it this year, but I assume that this is now going to happen once or twice a year. The best time to get vaccinated is right after the switch, because then you have the smallest lag between the currently active variant and the variants that is the basis of your vaccine.
I looked into this pretty carefully, and came away with the impression that resistance to becoming infected fades very fast, but protection from severe illness fades quite slowly.. There is still some protection even a year afterwards. Covid hospitalization rates have dropped hugely, and the main reason isn't that there's not much covid around, but that almost everyone has had the virus, had the vax, or both. That should tell you something about how long-lasting the protection of exposure to covid is, whether it occurs by infection or by injection.
Thanks. I'm not concerned about a severe infection but what about long covid? Increased risks of heart attack etc if you catch covid multiple times? That's why I want to avoid it if I can.
Last week, echoing Scott's popular "Beware The Man of One Study" essay, I posted a discussion of new data questioning whether light drinking has any health benefits.
Thanks to a few poorly designed studies, plus a 60 Minutes special from 1991, we've inherited a meme that light drinking (especially if it's red wine) improves cardiovascular health and increases longevity. The new data suggests a different and more subtle narrative.
EDIT: while this still looks like scientific malpractice, there are some problems with the NYT article. See comments below.
As I understand it:
A multimillion dollar federally funded project researching the effects of puberty blockers on children's mental health didn't find any positive effect and the lead researcher is deciding that she therefore is sitting on the data because she doesn't want people to use her study as evidence that puberty blockers shouldn't be used. She did say she will publish the data at some point.
She's claiming that the 95 participants in her study just had really good mental health to begin with and so they didn't have a lot of room to improve. This is contradicted by her earlier report that 'one quarter of the group reported depression symptoms and significant anxiety, and one quarter reported ever having thoughts of suicide.' When asked about this contradiction she said she is 'still analyzing the full data set.'
It is possible that the puberty blockers prevented a decline in mental health. If, as it seems, there was no control group, it's impossible to say. People who got HRT after no puberty blockers are doing worse than the puberty blockers group.
What worries me most is that Olson-Kennedy asks us to trust her clinical experience instead of the data she doesn't want to publish. If her self reported clinical experience was good enough evidence, then the NIH wouldn't have funded her study.
I am surprised that she is allowed to wait to publish, especially since her study was funded by the NIH. I would have assumed that more follow through would be required. Perhaps it is but the New York Times article didn't make that clear?
No control group means it was a bad study, but the worst of all is the propaganda around "would you prefer a live daughter or a dead son?" to persuade parents into taking the puberty blocker then transition bus. If the researcher is wedded to the idea "dead cis kid or live trans kid", then her beliefs mean she must repress her own work lest the narrative be questioned and all the kids not on puberty blockers will then kill themselves.
The article you picked out seems to be the only study linked to in that substack post that actually deals with studying the effects of puberty blockers for children. I'll take a look at it, but one study does not "lots of studies" make. The only other studies she cites are about kids who have not yet taken puberty blockers, how puberty blockers affect height, how effective puberty blockers are, and the study Chen did on the effectiveness of hormone therapy where they abandoned the hypothesis and method of analysis from their pre-registered version, which is indicative of p-hacking and cherry picking (in the pre-registered version they were going to study 8 different variables tracking mental health, in the final study 6 of those variables are gone and three brand new ones are there. Seems like a open and shut case of cherry picking when the study didn't give you the data you were hoping for).
But thanks for the one study though, I'll be sure to read it thoroughly. Unless you have more I recommend you stop telling people that there are "lots of studies with a control group" that "find a benefit of treatment" with puberty blockers.
Thanks for sharing this. It definitely gave additional perspective on the New York Times article and raised lots of questions.
The Substack writer Erin Reed who criticized the New York Times article is covering this from the perspective of a trans rights activist, and there seems to be bias here.
However, Reed does raise some good questions:
> Twenty-eight studies have been published from the Trans Youth Care (TYC) Network study, including from patients on puberty blockers.
So while researcher Dr. Olson-Kennedy has said there is some data still being withheld, some data has been released, and the NYT article doesn’t make this clear. It seems like the papers which have been published are the ones which show results supporting the use of hormones in treating children with gender dysphoria.
> What Ghorayshi fails to mention is that among those who didn’t receive blockers, suicidal ideation was much higher—66%, more than twice as high.
So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers. This would’ve been good for the New York Times article to mention.
I am still wondering why there was no control group. One important comparison we need to see is between people who got puberty blockers and people who didn’t use hormonal intervention.
"inexplicably, this study did not find that the various types of laws restricting trans youth were tied to considering suicide, only attempting suicide. [...] Also, the increased reported suicide attempts were not tied to the first or third year after the passage of these laws, only the second year after the passage."
> So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers. This would’ve been good for the New York Times article to mention.
Unless the study initially listed suicidal ideation as an endpoint, this is likely just p-hacking.
P-hacking exists, but if you don't have a specific analysis showing that it happened in a given case, just looking at any study that disagrees with your intuition and saying 'probably p-hacking' is not adding to the conversation.
If you have a more specific analysis to justify this claim I would be happy to hear it, maybe my accusation is unfair, but it's impossible to tell from your one sentence post.
This isn't "a study that disagrees with my intuition" though, it's simply not a study.
That's _if_ the study didn't initially list suicidal ideation as an endpoint, as I said. I'm confused about exactly which study is being referred to by this sentence.
True that it's not referring to a study directly; it's comparing the results of the unpublished study in the article, to data about a natural control group from another study by the same group:
>A total of 95 youth were enrolled in the GnRHa cohort. Mean age was 11.22 years (SD=1.46), and the majority were white (52.6%) and designated male at birth (51.6%). Elevated depression symptoms were endorsed by 28.6% of GnRHa cohort youth, and 22.1% endorsed clinically significant anxiety. About a quarter (23.6%) endorsed lifetime suicidal ideation, with 7.9% reporting a past suicide attempt. A total of 316 youth were enrolled in the GAH cohort. Mean age was 16.0 years (SD=1.88), and the majority were white (62%) and designated female at birth (64.9%). Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt. Life satisfaction was lower amongst both cohorts compared to population-based norms.
So it's not p hacking because there's no t-test.
It's just looking at a natural control group and saying 'with puberty blockers depression/anxiety/ideation were around 28/22/24%, without puberty blockers rates were around 51/57/66%.
We can't do a t-test on those sets of numbers because we are just getting summary statistics rather than the data set, BUT given that it's a 30-40% difference on ALL measures, it seems likely to be significant and unlikely to be p-hacking.
>>> What Ghorayshi fails to mention is that among those who didn’t receive blockers, suicidal ideation was much higher—66%, more than twice as high.
>So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers.
No, that's not true at all. Those numbers are looking at rates of suicidal ideation (and other things) *before* treatment, and that study doesn't report any results *after* treatment. So the study says that "before" HRT occurred suicidal ideation was 66% for the group that was going to get HRT but no blockers. Which tells us nothing whatsoever about whether people who get puberty blockers do better than people who do not. The study doesn't even claim to tell us that.
People who are about to get HRT at 18, but didn't get any treatment before that, *are* people where puberty blockers were potentially indicated at 12 but who didn't get them.
That is the natural control group for people who *did* get puberty puberty blockers at 12, and are now 18 (or w/e ages are used in the studies).
There’s an argument for that, but it seems like it was being interpreted as “we gave this group puberty bickers and their suicidal ideation is 22%, and we didn’t give this group puberty blockers and theirs is 66%”. I think the OP at least may be viewing it that way, and they wrote “So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers”, which simply isn’t true. It’s possible that the puberty blocker group will end up doing better than the HRT alone group, but that’s the very data that she decided not to publish!
Ok, so both of those options are wrong, let me see if I can pose it simply. Here is the relevant text fro the study:
>A total of 95 youth were enrolled in the GnRHa cohort. Mean age was 11.22 years (SD=1.46), and the majority were white (52.6%) and designated male at birth (51.6%). Elevated depression symptoms were endorsed by 28.6% of GnRHa cohort youth, and 22.1% endorsed clinically significant anxiety. About a quarter (23.6%) endorsed lifetime suicidal ideation, with 7.9% reporting a past suicide attempt. A total of 316 youth were enrolled in the GAH cohort. Mean age was 16.0 years (SD=1.88), and the majority were white (62%) and designated female at birth (64.9%). Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt. Life satisfaction was lower amongst both cohorts compared to population-based norms.
The experimental measure we are interested is self-identified trans kids who got puberty blockers from 12-16 vs self-identified trans kids who did not get puberty blockers from 12-16.
This study indicates that self-identified trans kids who did not get puberty blockers from 12-16 (ie people who are seeking HRT at 16 as their first medical intervention) have very elevated levels of depression, anxiety, and suicidality.
The researcher in question claims that there was no before->after effect of puberty blockers because levels of those traits were about the same before and after treatment, meaning at 16 their levels of depression and anxiety would be 30-40% lower than the no-treatment control group of 16 year olds.
(and note, the 19%/26% for the puberty blocker group is within 5% of the population average for people that age, AFAICT. They're basically 'normal' on these measures)
What happens before->after HRT is completely irrelevant to the study in quesiton, which is only about puberty blockers. We're only citing this study on HRT because the 'before HRT' cohort is a naturally-occurring control group for the 'after puberty blockers' cohort.
I think the reason there's no control group is that this isn't actually a researcher, it's a doctor at a clinic. Basically, this 'researcher' couldn't get a control group of people who wanted but did not get treatment, because their only contact is with people who got treatment at their clinic.
Grant money is available in many cases to physicians who 'do research' by providing care normally and measuring some things while they do it so we can get metrics on those treatments. But the primary motivation of the doctor is generally to subsidize their clinic and provide more care to patients.
Arguably the grant-giving agency should not have approved this grant without some way of getting a control group to compare to. Although arguably they had no way to know that the study would flounder on ceiling effects, because they had no way to know that most kids coming in for puberty blockers at that age are not yet doing very poorly on emotional health assessments.
It could easily have turned out that those kids are already struggling, in which case we would have had a perfectly good study about whether or not the treatment improved their metrics from there.
Basically this just feels like a case where someone tried to get grant money to support what they were already doing, it was a more-or-less reasonable grant to make, unfortunately it turned out that the experimental design didn't work due to ceiling effects that might not have been predictable ahead of time, and so it was file-drawered as bad methodology leading to a null result.
Not great, but extremely normal. The only notable thing being that it touched on a culture-war issue, so a bunch of people decided they could profit by applying isolated demands for rigor to it.
That's not what the article or OP's post is about, though.
The accusation is 'you should have published your null result, the fact that you didn't publish your null result is evidence of politics-driven malfeasance'
And that's an isolated demand for rigor, because null results are rarely ever published, in any field.
The thing is that the doctor in question explicitly told the New York Times that she doesn't want to publish the null result so that it isn't 'weaponized'. That's the evidence of politics-driven malfeasance, I'm not just inferring that she didn't publish the data because of politics.
My post is about several things, including pointing out that there should have been a control and that she's withholding her data for political reasons and that she appeals to her 'clinical experience' which obviously isn't good enough given that people saw fit to fund her study.
Yes, but most researchers that find nothing don't then do an interview with a newspaper about "the study found no effect but I won't publish - not because it found no effect, but because I'm afraid of political backlash".
If these studies are being used to prop up "this is medically necessary treatment, without it children will kill themselves, do you want a dead son or a live daughter?", then a finding of "intervention did not make a difference" *is* an important result.
If it's "intervention didn't make a difference here, but other intervention did make a difference there", that's also an important result. "Suicidality is not correlated with gender dysphoria, because the patients presented with a complex of psychological ailments" is also important.
But saying "I'm not going to publish a result because I'm clinging to my article of faith and I'm afraid the heretics would use it to discredit my religion" isn't science.
> 'one quarter of the group reported depression symptoms and significant anxiety, and one quarter reported ever having thoughts of suicide.'
A few things to note here:
1. ">Data from the recently released Teen National Health Interview Survey from July 2021 to December 2022, reveals that 21% of adolescents ages 12-17 report experiencing symptoms of anxiety in the past two weeks, and 17% said they had symptoms of depression."
The kids coming in before treatment were very close to the population average on depression and anxiety measures already. So if after treatment they were still the same as teh general population, you'd have ceiling effect and a null result. You could only get a result here if puberty blockers made them *significantly better* than the general population for their age, which is asking a lot.
As for the metric of 'has ever had suicidal thoughts,' obviously it's impossible for any treatment to improve that metric (unless it involves time travel).
2. The same linked study gives rates for 16 year olds who never got puberty blockers and are now pursuing HRT:
>Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt.
Basically, this is the correct control group for the study at hand - self-identified trans kids who got puberty blockers from 12-16 vs self-identified trans kids who did not get puberty blockers from 12-16.
If it's true that the kids in this study had the same mental health ratings before and after puberty blockers, then this indicates a huge improvement relative to this control group.
Overall: the thing is, this is a negative result; there was no effect of treatment, nothing happened. Negative results are rarely published; this gets called the 'file-drawer-problem' and there are reasons to dislike it (though I think they're over-stated because of how we design our studies), but it is 100% common and near-universal, not some crazy thing that only this person has done.
The only surprising thing about not publishing a negative result is that anyone noticed and cared about this one, and the culture-war reasons for anyone caring in this case are pretty obvious.
Add to this that there are already plenty of studies like this with actual control groups that show a benefit for treatment, and the story feels pretty clear-cut. One study produced no results, it was a culture war issue where the non-result would obviously be misrepresented by culture-war outlets eager to make hay, many other better studies on the topic are already available in the literature, so they decided to just move on.
In a perfect world where everyone pre-registered results this type of thing wouldn't happen. But in our world it happens all the time, and focusing on it for culture war reasons is an isolated demand for rigor.
And, really: This is NY Times writing a hit piece on someone related to the culture war. Given what we've seen from them here in this community, do you expect them to be fair and objective?
Yeah I agree the NY Times article is not great. Erin's 'fact check' was also full of twisty words and culture war tactics. I've said above why I don't think this was just a routine 'let's not publish this null result'.
I just wanted to respond to:
> If it's true that the kids in this study had the same mental health ratings before and after puberty blockers, then this indicates a huge improvement relative to this control group.
This could be entirely right. However, if we imagine that only the 50% most gender dysphoric group of 11-16 year olds who do not receive puberty blockers go on to seek HRT, and more extreme gender dysphoria is correlated with higher rates of suicidal ideation and suicide attempts, then this could explain the higher rates in the 'control' group.
So we don't *know* that these groups can be compared, unless we know that nearly all the kids who received puberty blockers were highly likely to seek HRT later in life without receiving those blockers.
It's currently fashionable on Substack to post your life story as a series of bullet points. E.g…
At 20, I dropped out of university.
At 24, my boyfriend dumped me.
At 27, I was treated for depression.
They are overwhelmingly about failure and/or very depressing. How come no one writes happy life stories? Is it shame? Embarrassment? Respect for the sad people? I just wrote my story and it was mostly about happiness and success. My wife chastised me for showing off.
Why is it acceptable to post about sadness and failure but not happiness and success?
I think its because when a stranger does better than you, you will feel worse, and when a stranger does worse than you, you feel a bit better about yourself. Therefore, writing about happy life stories increases bad feelings in the readers and consequentally, it's discouraged. In my opinion context matters here, its much easier to be happy about a friend's success than a stranger's, so imo one should tell the happy stories to one's friends and not to strangers on a public platform.
I think you are right about not making readers feel worse about themselves. Now that you mention it, my wife gets upset when a friend does well if she is not also doing well.
I am always very reluctant to share stories of success with friends. I can tell sad stories to them though. I have two blogs now and about 90% of my followers here are strangers but it is the reverse over on WordPress. I'm very reluctant to tell person stories of misery here while I tell them all the time on my WordPress blog.
It's possible to tell many stories from the same events. A few years ago I made the explicit choice to focus on the generative, dynamic, "what would I be pleased to see someone writing in an obituary" presentation, rather than a "woe be me all these things went wrong and my enemies were out to get me and I should be pitied" framing, and wrote a life summary like that. Using the first kind of framing made a big difference to how I felt about myself. Others seem to find motivation in the second kind of story, using anger at injustice to fuel activism.
I think we in N America, Europe, Au/NZ have allowed incentives to creep in to prefer the second kind of story in public. I think this is damaging. Both kinds are valuable. This extends to the stories we tell of our collective memory. The Progress Studies conference seems to be an explicit attempt to use the first kind of story to talk about our history, whereas degrowth, imminent climate catastrophe, prepping for the apocalypse, and regressing to an imagined version of 1950s Ireland all seem to be examples of the second.
Framing is important. I prefer the framing of sad stories as 'I survived' or 'I learned something'.
I think it's bad too for young people to hear sad stories from their elders. If they read about striving and success, I expect they are more likely to aim for it. If their expectation is that the future is all failure and depression, they might start life with that expectation for themselves.
So how can we get zoomers to pay attention to biographies of people who were not defeated by adversity? Does this material now have to become podcasts or videos? Do we need to petition Kate Beaton to bring back Hark! A Vagrant, pretty please with Jane Austen/Sojourner Truth sprinkles on top?
The Art of Manliness had a post a couple of years ago about how young men used to expect a time of adventure in their twenties and gives the examples of Earnest Hemingway, Steve McQueen, Sean Connery etc going off to join the navy or a Gold Rush or to fight in a civil war.
Since the 1990s, the expectation is that young people will go to college, get a job and buy a house. Those adventures probably had more than their share of sadness — but they were still adventures. Perhaps younger people no long have that mix of good and bad to look on. Now it’s all boring and bad.
It's possible (as Tolstoy implied) that unhappiness simply makes better stories.
The yellow press thrives on stories of gory transgressions. True crime is wildly popular. I understand that American daytime talk shows used to do brisk business with convoluted family drama. Even people who are quite successful tend to dwell in their memoirs on the challenges, not the intervals of smooth sailing.
"Now it is a strange thing, but things that are good to have and days that are good to spend are soon told about, and not much to listen to; while things that are uncomfortable, palpitating, and even gruesome, may make a good tale, and take a deal of telling anyway."
I wonder if that trope comes from self-help books? Those I read back in the day did like to punctuate the text with little anecdotes about "Bob was a successful businessman. But at age 40, Bob's marriage failed. By 45, he had lost his business. At 50, he was a homeless wino sleeping in the gutter. But by following the techniques here, by 55 Bob had cleaned up, stopped drinking, started a new and even more successful business, and got married to a former Miss Cornpone twenty years younger than him!"
Just that today people prefer to leave out the "And at 30 I turned it all around" part? Or maybe they're not 30 yet, so there's still time for the turning it all around.
When I worked in a bookstore, a woman gave me a Russian currency note for no particular reason, after telling me she had just returned from there.
Another time, when I was driving a bus, an 8-year-old boy gave me a stamp with a fingerprint on it. 50 years on, I keep them in a glass box on my dresser.
A narrative of success is about cause and effect, and needs to be written in full paragraphs. Because I did X, I had the opportunity to do Y, which gave me the opportunity to do Z.
A narrative of failure feels like a series of disconnected bullet points. This happened. Then this happened. Then this happened.
People who see their lives as a disconnected set of bullet points that just happen to them are likely to be failures.
If the US is considered to have budget deficit problems- how does Italy manage? How has it not collapsed? US debt to GDP is 124%, but Italian debt to GDP is 135%, and Italy is obviously not the world's largest economy or in possession of the world's reserve currency. 10 year bond rates for the US are in the low 4's, for Italy they appear to be in the mid-3's. I do understand that interest rates are lower in Europe at the moment, but these rates are not indicative of a bond market that's very troubled by what's going on in Italy. If huge deficits are such a gigantic problem- why haven't the bond vigilantes come for Italy?
Interesting comparison. Japan's economy hasn't exactly been good though, since the 90s. After an amazing 10-15 years, the Japanese economy pretty much just stagnated. Long term the outlook is grim, as the population ages and fewer people are born. Fewer people will be supporting more elderly, with either more debt or a significant reduction in benefits. More debt when the ratio is already that bad may not be possible.
Italy is bound to the Euro, so it has a lot fewer options than the United States in dealing with the debt.
In theory the US *can’t* default. All our debt is denominated in USD, so there’s always the option of printing your way back to almost-collapse with high, but not catastrophically high inflation. Italy, as part of the EU, particularly with German central bank philosophy being the guiding principle, can not. They have the advantage of northern neighbors potentially bailing them out, but considering most of Europe has its own economic woes, it’s uncertain how far the fiscally responsible north would go to bail out the improvident south.
In truth, no one knows what debt to GDP is unsustainable. Modern Monetary Theory will tell you it only depends on the interest payments, not the level of debt. Austrian Economists will be unhappy with almost any level of government debt that isn’t easily paid back on demand. It’s undeniable that 124% is less sustainable than 100% though, so whatever the real limit is, a higher debt to GDP is closer to that limit.
I agree with everything you've written- my point is, why isn't Italy's bond yield higher? Why isn't the market demanding more of a premium to lend to an even more indebted and fiscally weaker country than the US? Your 1st two paragraphs are reasons for a higher yield
That's a more complicated question with no single answer.
It's a combination of lower economic growth expectations in Italy, currency control preventing free movement of money to other markets, investment requirements (particularly for institutional investors/banks), and ECB policy.
Personally I'd point to the ECB policy as the primary factor. Italy's debt is supported by the ECB and their interest rates are kept artificially low, as the central bank buys many of the bonds that the government puts up for sale. The bond markets of national currencies are anything but free, and the intuitions of markets that would dictate higher vs. lower prices aren't always at play when it comes to national debts, especially those in capital-rich countries like Italy.
If economic growth is predicted to be basically flat in Italy (as it's been since 2008), a 3% bond is more attractive to purchasers when compared to the alternative in the domestic market. In the US, with consistent GDP growth, bonds have to compete with an investment in the market, which drives up bond prices compared to Italy.
At the end of the day, Italian banks and individuals prefer to purchase Italian securities, as there's lower regulatory and political risk. International markets are hardly efficient, and it's a better comparison with Germany (within the Eurozone) which has lower bond yields and is seen as a more secure borrower.
It would take some serious analysis by someone much smarter and more dedicated than I to sort through all the positive and negative influences on bond yields accurately. All us laymen can do is identify at all the different influences, and see what they sum up to, not what they specifically are.
Different countries are hard to compare on such matters. For example, Japan has a much higher debt, but that is not consider problematic by many economists since almost all its debt is held by domestic residents.
Another thing that makes this all the more mysterious is the flatness of Italy's GDP curve over the past decade and a half, while the US has seen considerable growth. One would think that this would mean that the US has considerably greater potential to outgrow its debt.
1. The US is not considered to have especially bad deficit problems by economic experts. This is a much bigger issue politically than it is among technocrats.
2. The US dollar is the default store of value around the globe, and it gives us a uniquely privileged position in terms of borrowing and loaning money. Very few countries can get away with the type of financial shenanigans we can.
The US has a long history of alternate elector slates showing up due to disagreements over the minutia of State election laws. I believe the most recent was from Hawaii in 1960 (Kennedy v Nixon), but as far back as 1876 a second set of Electoral Votes from a State made it as far as Congress.
"Cars have a long history of occasional brake issues, so from a certain perspective this is just normal car problems we ought not make a big deal about."
- Some guy who you just caught cutting the brakes on your car, 2020
A very interesting recent interview with Mike Levin about expanding our definitional boundaries of concepts like "intelligence". https://m.youtube.com/watch?v=6w5xr8BYV8M
One indirect observation that he made (in the context of cancer cells going rogue) was that "selfishness" can be described of as an individual simply having an overly-narrow self-concept. Whereas, if you think of "you" as being a part of a larger collective whole, then being "selfish" means doing things which benefit that collective whole, which superficially looks like altruism.
My immediate reaction to this is to dismiss it as trivial word games. I get what he's saying, there are intellectual blind spots if we keep insisting that AI is "not intelligent", but simply changing the words will not fix this blind spot, because most people still find the distinction between e.g. ChatGPT and a human brain meaningful.
Moreover I default to being suspicious of arguments that boil down to "we can fix this problem simply by changing the language we use".
He definitely does more than just play word games.
I first came across him because of research he did showing that by altering the electro-chemical attractor states of tissues, you can radically change the functioning of an organism, in ways that were previously only thought possible by genetic engineering.
I saw an inspiring art exhibit about the planned but unbuilt projects of Frank Lloyd Wright. Suppose I am a billionaire and I want to build those houses at long last. Given that Wright has been dead for over 60 years and the blueprints are available on the internet, could I just do it, or would I be violating some kind of copyright and/or need to get permission from his estate?
This is largely pointless whining, but I *am* curious about the answer. Why is the US healthcare billing system so fragmented? Two months ago, my son broke his elbow (he's fine, the surgeon did a great job, and five-year-olds heal amazingly). The treatment involved a lot of moving parts, but people seemed reasonably able to talk to each other, and all the procedures ended up in a single electronic chart. But, I've gotten at least half a dozen, if not more, separate bills (which are not in 1:1 correspondence with the insurance claims). Why isn't there some attempt to aggregate the bills? As far as I can tell, everything I paid was genuine, but I can't help thinking that it wouldn't be hard at all to set up a random payment processing website and fleece me for $100 or so (which admittedly isn't the hospital's concern, but it would be nice if I didn't have to worry about it.)
Any time you receive a medical bill you don't recognize/expect, call the hospital to verify it.
Fraudulent bills are, indeed, quite common.
And if we annoy hospitals enough that they stop sending bills which look fraudulent (holy shit I got a bill for an anesthesiologist once that looked LESS LEGIT than any scam I have ever received, the only way it could have been worse was if it were handwritten in crayon), maybe this problem will get a little bit better.
Seriously. Annoy your medical provider. Make their lazy accounting processes cost them customer service hours, and maybe they'll get their shit together.
The problem isn't that I don't "expect" the bill, I was just kind of hoping that, just as I didn't need to care about the anesthesiologist being logistically different from the surgeon or the radiologist or the emergency room or the ambulance *during* treatment, I also wouldn't need to care about it at payment time. For some reason they all insist on sending me separate smaller bills, on top of the large one sent by the hospital as a whole. (I don't remember this being as much an issue when I gave birth, although that may have genuinely involved fewer distinct groups of doctors.)
I was seriously pissed off when I received a steady series of bills, because I deliberately chose not to involve insurance after looking at the bill at the hospital, and deciding the personal cost of the additional hassle of dealing with insurance wasn't worth it. I was definitely not expecting those bills.
It’s harder to travel light as a man, I think. Have you looked at Ex Officio? They make comfortable clothes that are designed to be washed in a sink and dried overnight on a hanger.
Are you going to be on the road the whole time or will you have a base of operations?
They do, but do take a look at the "new" section of Steam for a few days. It's a tough business.
You seem to have the right attitude. Unshakeable self-confidence is a good founder trait, at least when combined with a small amount of doubt (most startups need to pivot, and the pivot can't happen if the founders are hell-bent on a rigid goal). Good luck!
In some town there are whole meetups of indie game developers. There's one in my town. Also a big online scene, including big and little contests.
Make your first game, sell it. Make your second game, sell it. Make your third game, sell it. Etc. The moment your income from games is comparable to your income from the job, quit your job and enjoy life full of making video games.
If you try it the other way round, chances are it will result in you being homeless.
I don't know how much time you got until the end of college, but it's October, so let's assume at least 8 months. If by the end of June 2025 you have at least *three* video games completed, generating a monthly income of at least $100, then I think you have a chance. Otherwise, you don't.
I mean, monthly $100 is peanuts, that by itself wouldn't save you from homelessness. But it's at least something, a proof of concept. It shows that you can complete the product, and you can find a customer willing to pay. Now you "only" need to get 10x better to be able to survive, and that is not impossible; practice should make you better, plus you can keep selling the old games while adding the new ones.
On the other hand, if you can't get a monthly income of $100 by the end of June 2025, it means you are not ready. You need to get a few years of experience first, and that means getting an IT job now. Don't quit the job until you have produced and started selling your third game. (You think it is difficult to make games while having a job? Wait until you have kids...)
If you're going to self promote, you might want to start using capital letters. You don't exactly create confidence that you have a quality product without them.
Scott Cawthon is arguably the second most successful indie game creator ever (after Notch), but it took him 11 years of making games before he made FNAF, his first real success. Over that time period he made 73 different indie games. All that to say, unless you have the funds to support yourself for a decade (or even just 5 years, assuming you're twice as talented and lucky as Scott Cawthon), you shouldn't quit your day job. Scott was working at Dollar Tree when FNAF took off.
Self promotion only becomes relevant *after* you actually made something. You are not there yet. (And my advice was to figure out whether you can even get there.)
Are you a CS major? I spent much of my youth/teens tinkering around with video game programming and then majored in CS. Got great grades and was always working on video game/graphics-related side projects in my spare time. After college I got a job at a AAA video game studio. Even the first year at that studio made me a FAR more competent software engineer than all my past experience combined. Problems that would take me a week became trivial. I'd highly recommend trying to get some real industry experience before taking on such a big project on your own.
I actually agree with your point on AAA games which is why I left the industry... just didn't inspire me like a lot of smaller indie games or older games do. IMO limitation is one of the greatest drivers of creativity. I'd actually love to have my own small indie studio someday.
If you haven't come across it already, I'd check out an engine like Godot.
Best of luck to you and your GPT coworker.
> i gots enough computer learning on account of i took one python class.
Yikes. Based on your responses, this entire endeavor is a joke.
Why wait until after college? Start now, while you still know everything and before life makes you cynical.
I spent a bunch of time making a game about 5 years ago. It barely got any wishlists on steam, and while some of the early stages of development were quite fun, the processing of finishing the last 10% of playing through the same section over and over again combing out bugs was completely demoralising and took forever. I kinda wish I'd spent that time doing almost anything else and, from the research I was doing on the indie game market at the time, that's not an uncommon outcome for a lot of indies.
i'm doing that right now except also having a job which unfortunately gives me much less time to work on the game but on the plus side, food and electricity. mine is a fully text-based game so i'm really aiming for the .01% of the .01% of the .01% of audiences.
I don't know the specifics, but you probably want books on project management. In the absence of a structure large enough to include project managers, you will have to wear a bit of that hat, which includes negotiation, estimation, and presenting a case to a decision-maker.
Eremolalos's experience above suggests otherwise. And there are plenty of movies not made for kids that would absolutely traumatize a kid.
But I don't think this would traumatize him. I'm just not sure if it would be net positive or harmful.
I think you're right to ask (and right that there are absolutely movie experiences that can traumatize young kids, who sometimes don't know or can't fully process that they aren't witnessing real events), and I don't think it's a question that can be settled on the basis of some principle. It's not actually "should we shelter kids from difficult feelings or toughen them up to real life?", it's "is THIS child ready to have a salutary, if sad, experience with THIS movie?" You know your kid and his sensitivity level--does he remind you of you at that age? Does he tend to take things in stride or does he have intense feelings sometimes that don't make sense to you as an adult, does he perseverate or worry to the extreme about things related to loss?
Mine used to have very intense feelings about lost objects, which he tended to personify. It wasn't "I'm sad I don't have this thing anymore" so much as "this thing will not be OK without me to take care of it." (He also once in awhile had a panic meltdown for incomprehensible reasons, eg that a toy was lying at the bottom of a wading pool. Well that one was somewhat comprehensible, it clearly held symbolism for him.) He was an incredible packrat because getting rid of possessions felt to him like, maybe, dumping a pet by the side of the road--you don't do that just b/c it's old and not fun anymore, and in that same spirit of care we had to keep old toys, papers he had scribbled on... and oh my Lord, we left behind a rotting stick at the creek once whose tip was shaped a little like a horse's head and he brought it up for 2 years whenever he couldn't sleep. He's outgrown this completely now at 11, thankfully. All that to say, these things seemed to be proxies for him for a deep aversion to the idea that irreversible loss and sorrow exist. I literally went back and tried to find that damn horse-head stick because he could not. stop. thinking about it. (And believe me I was trying to ease him along into accepting that sometimes things are just gone. And eventually he did.)
So... I probably wouldn't have shown him the kangaroo movie at that age. (I screwed up on a few movies. I wanted him to love The Iron Giant but showed it to him too early and he thought it was sad & scary.) But when I saw him shift over to being less sensitive, which might have been around 7, then I probably would. It's not an either-or question, it's a question of when.
But if none of this rings a bell at all, if it all sounds so unlike your kid that mine just might be a space alien, maybe you should just go ahead! I do think a lot of kids could handle themes like this at 5--or be sad but in a way they can feel is helping them, as maybe you were.
Your story about your son is a good example of why you don't have to deliberately introduce most kids to tragedy and loss. The little losses of their lives feel huge to them. They are very emotionally alive.
Right, their first year your job is to keep them safe. After that, your job is to make them dangerous.
3 year old and a boiling pot, really? I was going to fully agree with your opening sentence but I don't think you have to go *that* hard. Great if it worked out for you and all, but wow, that's really young.
If you remember it fondly then it probably does have value. But had you seen or heard similar stories at that age before you saw it? The context matters. If the first time someone is confronted with death and cruelty is watching the 1942 Bambi, it will seem a monstrous horror. But most kids who see it have already seen birds killed by cats, heard parents explaining where bacon comes from, or headshotted opponents in a tactical FPS, so Bambi is just a little sad, not a searing tragedy.
I agree that context matters. Protecting children by taking away the opportunity to watch a sad movie makes it even harder to watch the next sad movie. Before you know it you have university students asking for trigger warnings.
I saw Bambi when I was five and I cried. I still cry at movies but I know that emotions over fiction pass quickly. I can learn from them and enjoy them even while experiencing strong emotions. Better to learn early, I think.
There's definitely something weird about memories of movies watched as a child. I find (and others I've spoken to agree) that when you rewatch a movie you'll be amazed to find whole sections, whole themes, whole subplots, or whole really obvious stylistic aspects that you have no memory of. This holds even for movies I watched *dozens of times* as a child!
It's really quite bizarre. It makes me wonder if I'd watched an edited version. But no, I didn't, I just tuned things out apparently, and most children do. Just one example: a lot of 90s kids movies are full of slapstick, throughout the movie. I hate slapstick, and yet I remember liking these movies and had no memory of the sheer amount of it. I must have just tuned out the parts I didn't like.
The same applies to bad dialogue and bad acting, though part of that's lack of knowledge of what's bad and good, but not all of it.
As another parent, I don't know anything else about the movie, but I'd strongly consider showing it to him. It's important for your kids to get in contact with the full range of human emotions. In general imo most parents are on the far end of safetyism and children can handle much more much earlier than they think if you actually teach and help them along the way.
Funnily enough just yesterday I've had a related discussion with my wife; Our daughter (3) watched a children's tv show about a family of dinosaurs we haven't seen before, and then we notice that several episodes go like this: 1) the protagonists meet other dinosaurs which are just as intelligent as them 2) they get into conflict 3) the protagonists win the conflict and ... eat the opposition.
At first my wife was shocked, but after a short talk we both agreed that it might be a good counterweight. We had been complaining about some other tv shows she had seen before which often portrayed even dangerous predators as actually nice deep down and that all animals can get along great through the power of friendship. We know even quite a few college-educated adults who have completely unrealistic ideas about animals (along the lines of "bears would never attack humans unless unnecessarily provoked, anything else is animal hating propaganda") and it often strongly biases their politics in a bad way.
Obviously you still need to talk with the kid and contextualize what they see, but if you do they can handle it just fine.
I agree about kids getting in touch with the full range of emotion, but most are quite able to do that without seeing, when quite small, a movie with a tragic death in it. Think about real kids. They cry far harder and oftener than adults do, and while some crying is set off by anger or frustration, lots is set off by grief. Kids grieve when they lose a favorite toy, when they suddenly start missing their parents when at school or with a babysitter, when their feelings are hurt during play -- and also when they see sad things happen. I can remember a few times when I was small and cried in sympathy when some other kid fell down. I saw my daughter do that too when she was small. And while the things kids cry with grief over seem small to adults, they seem huge to the kids. Don't you remember being a kid, how it felt?
My introduction to death was when I was about six, being brought by my mother with her while she went to visit the deadhouse (as the hospital mortuary was referred to), where the body of a neighbour was laid out before being coffined . Everybody knelt and said a prayer, then afterwards there was the funeral and burial, and I saw the coffin being put into the grave and the grave filled in.
I did have a couple of dreams about death and being buried myself after that, but I put it together that hey, when that happens, I'll be dead so I won't know or care.
So I find it hard to think that a movie could be very traumatic as an introduction to loss, when the Kangaroo is hopping away in good health and still alive, just not going to be around Dot any more 😃
Absolutely, though, I wouldn't let six year olds watch horror movies or anything with explicit violence.
I think precisely because it's a smaller problem, it's more likely to haunt a child. Kids often have separation anxiety, and don't commonly have a self-preservation instinct (or any other sense of their imminent mortality). We had a dog die this summer and then got a new one; the conversation with my five-year-old about dog cremation was morbidly entertaining, but having the new dog run away from me on a walk and go missing for fifteen minutes clearly made more of an impression.
This should not be misconstrued as an argument against watching the movie; if nothing else, practice separation is presumably actively helpful in handling real separation anxiety.
This matches my experience
My intuition is that certain kinds of media and themes will shock and/or distress kids the first time they see them, regardless of the age they are when they're first exposed to those themes. I've known people who were exposed to horror movies when they were six and people whose parents sheltered them until they were sixteen, and it seems like they had roughly the same "trauma" response to their first horror movies. While it might be harder for a six year old to dismiss the "trauma" of their first horror movie...I dunno. Teenagers are often extremely good at leaning into and even enhancing their own "trauma" with rationalizations for why scary stuff might be real. I know I managed to be as scared about alien abduction at 14 years old as I was of monsters in the closet at six.
You weren't asking about horror movies, but probably exposure to the concept of loss is similar. It's going to hurt regardless of the age your kid is when they first experience it as a theme. Five seems like a reasonable age to be exposed to the concept of loss and why it makes people sad, especially if it's handled in a beautiful way. And while I'm not a parent, my intuition is that it's better to have initial exposure to the themes of loss via media rather than a sudden shock of it in real life (the sudden death of a pet or grandparent, etc).
Edit to add: Be prepared for your kid to be heartlessly disinterested in your beloved Dot and the Kangaroo. He might be unforgiving of the rough animation, slow pace, etc after being trained on 2020s modern media (presuming you've allowed them to see any).
He may not cry, so go ahead. It's a way of introducing children to the idea of parting and ending of things, and at least Kangaroo isn't dead, she's just leaving to let Dot return to her human life (and it's open ended to the possibility that they might meet again later).
Sometimes people leave (they move away, they die) but while that's sad, it's not a bad thing and you go on with your life.
I think it's important to introduce this stuff to kids early. Tragedy sticks with you, and I look back on the tragic - even borderline traumatic - stories of my childhood as the most enriching.
I was not ready for Bambi when I was 5. I was ready to learn about death, but the movie presented it in the most traumatic form imaginable: the death by violence of a mommy. After seeing Bambi I understood the reality of death better, but it really left a huge dent in my sense of wellbeing. I tormented for years by stories that formed in my head about little animals left in the nest grieving, terrified and starving to death because their mothers did not come back. I I think you should err on the side of caution with kids about matters like that.
When I was college age I taught nursery school for a while part time, and when the school guinea pig died we showed kids her body the next day, and answered their questions, and let them examine her body or pet her (and then wash their hands really well). We also told parents about the guinea pig's death and how we'd talked with the kids about it. I think that was a decent introduction to death for the kids.
Movies hit kids differently I think. My son helped us bury his grandparents dog, and I think that was a positive experience. I would be reluctant to show him Bambi though.
I remember being very captivated by Bambi's father, though, and how he sort of revealed himself to Bambi (and to the child viewer) - which I don't think would have happened without the death of his mother. I remember also that frightful word - "Man!" Which implicit lesson re nature and loss has only grown more true as time passes.
Once I was sitting next to one of the local springs with a den of Cub Scouts among others, I think it was, listening to a park staffer give a little talk about the "spirit" of the springs, a variety of salamander, and she asked the assembled group if they knew what the salamander's chief predator or threat was.
We all sat awkwardly unable to answer for a few moments.
Then a kid piped up bravely, and with something of that Bambi drama: "Man?" And I think we all, adults and children, thought to ourselves, yeah, that tracks.
"Uh, good guess? Actually, it's crawfish", she said.
Dumbo seems like a good precursor to Bambi. It's wrenching but Mother doesn't die.
It was an old favorite from when you were five, so you know for a fact that at least one kid can handle it. The question to ask is whether you think your son is meaningfully different from you in his ability to handle sad stories. In general, I agree with the majority here that it's good for kids to encounter difficult emotions in fiction. And if it becomes too much, you can always pause the movie to talk about the movie and give him a chance to decide whether he continues or not.
It was an old favorite from when you were five, so you know for a fact that at least one kid can handle it. The question to ask is whether you think your son is meaningfully different from you in his ability to handle sad stories. In general, I agree with the majority here that it's good for kids to encounter difficult emotions in fiction. And if it becomes too much, you can always pause the movie to talk about the movie and give him a chance to decide whether he continues or not.
Yeah I think they need the opportunity to experience and rehearse different kinds of emotions in a safe manner while they're developing. If they have a strong reaction then have a discussion with them afterwards to help them process and contextualize the feelings, but I don't think that shielding them from children's movies is going to help much in the long run.
One thing I think is pretty true is that not only *can* most humans have all of the standard suite of human emotions, but that we *will* have them with some regularity, because the brain doesn't like to lets parts of itself just atrophy inactive forever. if you don't have an y appropriate targets for an emotion in your experiences, you will attach that emotion to *something* going on in your life, in a way that may be less appropriate and more damaging than just having an actual correct target.
For negative emotions, movies are probably a good target because they provide accurate contexts to attach those emotions to, while having those events not be something in your own life that you have to constantly fear or obsess over.
I've watched several videos of the SpaceX Super Heavy Booster going straight back to the launchpad, which is one of the coolest things I've seen currently happening in the space program. A question I've never seen answered: Why could they never recover the space shuttle fuel tank like they could with the rocket boosters? It seems like a huge piece of equipment to throw away and replace every single time.
I've wondered that too.
I always assumed it because it stayed attached for much longer it burned up after being ejected. But that's just a guess.
The SRBs (solid rocket boosters) were jettisoned at a speed of roughly 4,800 km/h, while the ET (external tank) was jettisoned at over 28,000 km/h – close to orbit – so reentry was much more violent.
(numbers from Claude, so double-check them before building your own reusable launch system)
If you are going to post LLM output to make factual claims, please do us the courtesy of performing the verification yourself, or otherwise leave out the supposed details. We can all type a prompt into a chatbot. We also don't need more imaginary numbers floating about for search engines to find and become the foundation for future myths.
> If you are going to post LLM output to make factual claims, please do us the courtesy of performing the verification yourself, or otherwise leave out the supposed details.
Would you have felt better if I had posted numbers from a superficial Google search? Or from Wikipedia? How thorough and well-sourced would my verification have to be according to your standards?
> We can all type a prompt into a chatbot.
Then why doesn't everyone? State of the art LLM chat bots are perfectly capable of answering simple questions such as the above, and in great detail – enough details to enable further, independent research and verification, if desired.
+1
I don't think I've ever gotten a truly useful answer out of it, though I haven't tried in a while. The AI worship around here is really annoying and dare I say may blind some people to its limitations.
Ironically, I find the art generators vastly more impressive than the LLMs, despite the former getting far more hate. Of course that may be why.
I think that ChatGPT is useful in finding specialized nomenclature. E.g., if one is looking for a named law or model or theorem, and one can describe what the law/model/theorem is about in layman's language, the LLM can be useful in finding the name of the thing.
On the other end, if one wants to survey possibilities and select them according to some measure, e.g. 20 lowest-boiling inorganic gases, good luck, unless some human has already compiled such a list - even if every candidate is already documented in Wikipedia, in the LLM's training set.
( And I've been steering clear of politically controversial questions, where the RLHF Woke indoctrination is likely to obscure what the _capabilities_ of the technology really are. )
Whatever Google embedded in its search is pretty awful. I just tried
> What is an example of a molecule with an S4 rotation reflection axis but no mirror planes and no center of inversion?
It replied with
>A classic example of a molecule with an S4 rotation reflection axis but no mirror planes or center of inversion is methane (CH4); its tetrahedral geometry allows for three S4 axes, making it a prime example of this symmetry element without additional symmetry features like mirror planes or a center of inversion.
which is just wrong. Methane has 6 mirror planes. In fact, _this_ LLM "knows" this. If I ask it
>How many mirror planes does methane have?
I get:
>Methane has 6 mirror planes.
>Explanation: Since methane has a tetrahedral geometry, you can create a mirror plane by selecting any pair of hydrogen atoms and passing a plane through them and the central carbon atom. This gives you 6 possible mirror planes.
That's funny, because I went to the effort of verifying _your_ assertion that ChatGPT gets this wrong, and it turns out, you're wrong:
https://imgur.com/prV9Mch
That's with ChatGPT 4o.
Old-fashioned Google turns up first that Ripple is some sort of cryptocurrency, then adding in LRL gives me:
https://direct.mit.edu/jcws/article-abstract/23/2/133/101892/Ripple-An-Investigation-of-the-World-s-Most?redirectedFrom=fulltext
"In 1962 the United States conducted its final atmospheric nuclear test series, Operation Dominic. The devices tested were designed and built by the Los Alamos Scientific Laboratory (LASL) and the Lawrence Radiation Laboratory (LRL). During the test series, LRL conducted four tests of a radically new design called the Ripple concept. Tests of the Ripple concept demonstrated performance characteristics that eclipse those of all nuclear weapons designed before or since. For numerous reasons discussed in the article, the Ripple concept was not pursued, but the technology it pioneered has been in continual development—for peaceful purposes—to this day. Until now, very little has been known about these tests and the concept behind them. This article, the result of a multiyear investigation, sheds light on the Ripple program for the first time, allowing for a largely complete account. Included are the origins of the concept and its designer, the technical characteristics, the significant role played by the geopolitical context, the test series in detail, and the cancellation and legacy of the program."
So I'm going with Chastity here since she knew what she meant in the first place and the ChatGPT did not suggest it as one possible answer, and did get the LRC and LRL confused when replying to her.
I just noticed - LRL does not correspond to Langley Research Center (LRC). So yeah, the AI is too stupid to work out that "L" and "C" are different, it's just regurgitating something from its training data.
People are asking questions here and hoping that someone has the motivation to research a real answer (or has expertise to share). It used to be that such questions were accompanied by "and a cursory search came up with these links which leave me confused" or a Fermi estimate, and it would be nice to return to such standards. Adding unverified LLM numbers as answers doesn't help, nor would "my random friend said". True/necessary/kind (2/3) are the tests we are supposed to be applying, right? https://slatestarcodex.com/2014/03/02/the-comment-policy-is-victorian-sufi-buddha-lite/
You might be right but you haven’t proven the LLM figures posted by Adrian wrong yet, and regardless of source that would be necessary to this argument, an argument that I otherwise don’t care about.
Which makes none of your comment true (or rather not yet proven), necessary or kind.
>You might be right but you haven’t proven the LLM figures posted by Adrian wrong yet, and regardless of source that would be necessary to this argument, an argument that I otherwise don’t care about.
We shouldn't have to prove a negative here. For the time being LLMs are simply not accurate enough.
Where LLMs are wrong, which is not infrequently at the moment, they're usually wrong in ways which are not easily apparent to people unfamiliar with the subject running a cursory search.
"Why was the Space Shuttle's external fuel tank not recovered" isn't a complex question which requires some unique insight only shared by five experts worldwide, two of which frequent ACX, nor does it require a Fermi estimate by the Bayesian gurus that upheld the standards in days long gone by.
Looks like "sometimes it's cheaper to throw something away rather than re-use it" is the answer, so far as I can find one.
Good discussion on a Reddit site about this question, we all got side-lined by "is ChatGPT answer good enough?" from the original question, which is "WHY did they not re-use the external tank?"
A combination of "they wanted to shave every pound of weight off" and "plans were there to use them to build a space station but never went anywhere as the adjustments would mean too expensive, too heavy, too much new equipment to make this possible", so it ended up "as light as we can manage and make it throw away to that end":
https://www.reddit.com/r/space/comments/1k4g1o/the_external_tank_from_the_space_shuttle/
This is the very first time I haven't been disgusted by the idea of having upvotes on act, because I would upvote this comment and downvote its parent.
I can see upvotes.
The comment you replied to has one upvote, and its parent has nine.
Yours will have one after I upvote it.
> perfectly capable of answering simple questions
...no, they are perfectly capable of autocompleting a piece of text that begins with some combination of the words you typed in and whatever else the vendor chooses to prepend in a manner that results in a statistical match for text found on the internet.
This is not the same thing, because the internet is full of rubbish, and also because there is nothing in the process to distinguish between "here's the answer" and "here's a piece of text in the style that an answer would be written in, if you were given an answer". Your "numbers from Claude, so double-check them" disclaimer implies you are at least somewhat aware of this, and it would be disingenuous to now claim otherwise.
Hence people specifically wanting a response from a human: yes, humans can also be wrong, make things up and/or lie, but our well trained intuitions for how to detect that stuff at least have some small hope of matching the territory in this case; when an entirely alien mechanism is generating the text and also our mental model is demonstrably mistaken about what it is even doing in the first place, there is essentially none.
> > perfectly capable of answering simple questions
> ...no, they are perfectly capable of autocompleting a piece of text that begins with some combination of the words you typed in and whatever else the vendor chooses to prepend in a manner that results in a statistical match for text found on the internet.
Potayto, potahto. I used to think like you, until I started using LLMs in earnest. Sure, I'm still encountering hallucinations on a regular basis, but the "statistical parrot" mental model falls far, far short of their real capabilities.
> Sure, I'm still encountering hallucinations on a regular basis
Potato, potahto. Outside tech demos, when people ask questions they want actual answers and not hallucinations. It's amazing that the dog can sing, but it's not going to replace my CD collection.
I'm not convinced the "statistical parrot" model is wrong, rather I think that that's a good description for a lot of what people do. It's not a complete model of people and it's not what we mean by understanding, which is why LLMs are such a mixed bag.
If you google household ingredients for washing a floor, you get a bunch of hits for vinegar, vinegar & dishsoap, and vinegar and baking soda combined. The last of these is nonsense, because the 2 active ingredients cancel each other out. I asked GPT4 for ingredients a few months ago and it gave me vinegar and dishsoap. I asked whether adding baking soda would help, and it agreed heartily: "Adding baking soda to your cleaning mixture can enhance its effectiveness, especially for tackling tough stains and odors on linoleum floors."
Jeffrey Soreff, a chemist who posts here frequently, has posted many wrong answers it has gotten from GPT4 for chemistry questions that are easy to look up the answer to. Recently he posted that it doesn't understand what a tetrahedron is -- can't make an image even when he explains that it's a pyramid with a triangular base.
GPT4 often does no more than compile the most frequent google hits, but then it packages them so that they sound authoritative. I don't think either a superficial google search or a chatbot query is adequate for questions like OP's. You have to google for answers and then you poke around and check the one you think is probably accurate. If you don't know how to poke around and check that particular question then you just don't know for sure what the answer is.
It's also kind of rude to chatbot an answer to somebody's question. With the same amount of typing the person could have asked a chatbot this question instead of you. Obviously they are looking for a different source of information.
"rude"? What is "rude" about it? Did I insult anyone? Some people do seem to be offended, though…
I openly stated my source. Feel free to ignore such comments.
Edit: I am actually quite surprised about the general reaction to my lighthearted comment. Admit it – "double-check them before building your own reusable launch system" is at least worthy of a smirk, no?
I notice you don't respond to my main point, examples of inaccuracy. Anyhow, about the rudeness: It's sort of like answering somebody's question by sending them to this: https://letmegooglethat.com
Yes, I did get a smirk out of it, but Victualis is right.
Here is what ChatGPT has to say on the topic:
Using LLM output in online discussions or forums can come across as impolite for a few reasons, especially if it’s clear that the response isn’t a personal one:
Lack of Authentic Engagement: Posting a generated response might make it seem like the person didn’t genuinely engage with the question or community. People generally appreciate thoughtful replies that show understanding and connection with the original question or topic.
Unfiltered or Imprecise Information: Sometimes, LLMs might generate responses that are too generic, overly detailed, or miss subtle context cues that a real person would catch. This can make the response feel like an awkward fit for the conversation and might even be misleading if not carefully reviewed.
Lack of Personal Touch or Effort: Communities often value responses that show effort, nuance, or personal insight. Posting LLM responses can seem dismissive, as though the question wasn’t worth the time to answer individually.
Potential for Misinformation: If people recognize a response as AI-generated, they may also distrust its accuracy. Unless the response is verified, it might not meet the standards of a community that values reliable, accurate information.
Risk of Redundancy or Dullness: LLM responses may sound “robotic” or repeat information already available in standard sources, lacking the freshness or original thinking that people often look for in online discussions.
When using AI-generated answers, giving credit or adding a personal summary can help avoid these pitfalls and maintain the quality of engagement.
Actually, it's not nonsense, I've heard of baking soda + vinegar. You apply baking soda to the grease on the floor, then add vinegar and mop it up. I don't know how well it works, because I've never tried it, but it's not implausible. I think the idea is that this makes the grease lift off, but I'm not sure. Or maybe it's just something that someone tried, and it worked for them.
In any case, what I asked AI about would not have worked. I asked about just mixing it into the wash water along with the dish soap
and vinegar.
> Would you have felt better if I had posted numbers from a superficial Google search? Or from Wikipedia?
Yes. Because those would have contained context and metadata and citations which could be further checked and traced back, and terminate in a NASA PDF or something. Even if they had contained literally the same ex cathedra statement word for word as ChatGPT, you would be no worse off in trying to factcheck it, and the *lack* of all that would have told you something useful: that it is a low-quality source of dubious veracity that may well be wrong. Meanwhile, some LLM obiter dicta kills all curiosity and is the junk food of writing: fattening webpages while providing no nutrition.
Just want to chime in that I agree with Victualis here. Results from a superficial Google search or from Wikipedia would indeed be preferable.
Also, "results from a Google search" should, ideally, not be credited to "I Googled and found this", but "[this site] says...", because the fact that you found it from Google doesn't tell you a whole lot about its reliability. (Google would like the fact that they brought it up to mean something, though.)
Agreed. I tend to include urls with information I find, so that people reading the comment can see exactly where I found the information (and, usually, what organization it is associated with).
On a related note - even a very superficial Google search is often improved by including the name of a plausibly authoritative organization in the search terms. ( Bluntly, I got a bit sick of the back-and-forth on the shuttle H2/O2 tank meta level questions above, so I did a cursory Google search - but including _NASA_ in the search, and then commented, quoting from the NASA site about the shuttle and citing the URL. )
I would trust a google search more than an LLM - at least then I know that at least one real person on the internet believed it.
For anyone that did want to know the numbers, it looks like the above is broadly right. I've done a brief Google but haven't dug especially deeply (though I see a bunch of sites that seem to agree).
The fuel tank was jettisoned after main engine cutoff (MECO) but prior to orbit (https://en.wikipedia.org/wiki/Space_Shuttle_external_tank) - the shuttle then used its online maneuvering system (OMS) engines to get thr rest of the way to orbit.
Speed at MECO was 17,000 mph (https://pages.cs.wisc.edu/~yat/space/facts.htm) which is 27.4 km/h. 17m seems pretty rounded but about the right number, and I can't find any other numbers out there. It's pretty close to the 28k that Claude gave.
This is broadly right; a massive, ongoing complaint in the 80s and 90s from the folks who would broadly be SpaceX employees today was that we constantly threw away the ET when it was almost at orbital velocity and that we should have found a way to push it to orbit and use it for space stations.
Many Thanks! I did much the same thing, and got essentially the same answer.
If you are going to post the output of some system of norms and values, please provide evidence that this set of norms and values has been backtested across millennia of human culture and indeed promotes human thriving. We can all judge others easily based upon our own standards as the absolute correct stance. We also don’t need more imaginary moral systems floating around for impressionable algorithms to find and become the foundation of future moral myths.
I found the comment amusing, and I've already chiseled it on a tablet and buried it in the backyard.
I think this is unreasonably strict. From my experience I'd estimate that LLM provided figures are no less accurate than cursory Google searches. I'd be very surprised if Claude gave figures outside reasonable confidence intervals for questions like this more than 5% of the time.
I think either people heavily exaggerate hallucination rates on cutting edge models because of bias, or otherwise I'm very very curious to see what kind of tortured queries they're giving to get such inaccurate results.
LLMs are unlikely to give figures less accurate than a cursory google search, the step where the error is more likely to enter in here is when you ask the LLM to explain something, and it gives an answer which it justifies based on the presumed relevance of those figures. The presumption in accepting the LLM's answer is that it's more likely than an uninformed person to be generating a correct answer, for which those figures are an appropriate explanation. In areas where the average person doesn't have enough domain knowledge to generate the right answer with some cursory googling, LLMs are wrong quite a lot, but this also makes their inaccuracy hard for the average person to check. It's easiest to check on straightforward factual matters which you're familiar with, but you know the average person is not.
>It's easiest to check on straightforward factual matters which you're familiar with, but you know the average person is not.
For example, I recently asked ChatGPT 4o "Which Valar took part in the War of Wrath?"
The actual answer is that the published Silmarillion doesn't explicitly name any Valar as doing more than agreeing to the expedition, uses language that's ambiguous but can be (and often is) read to imply that some or all of the Valar are directly involved (referring to "the Host of the Valar" and "the Might of the Valar" doing various things in the war), and includes details that are usually read as implying that the Valar didn't accompany the expedition (namely, the Maia Eönwë commanding the army rather than Manwë, Oromë, or Tulkas, and after the final battle Eönwë ordered Sauron to return to Valinor for judgement by Manwë as he felt he lacked the authority to judge a fellow Maia).
ChatGPT answers this question okay. It glosses over the ambiguity of the text, but the overall framework isn't badly wrong, and it offers up some mostly-plausible speculation on how four of the Valar might have been involved. It does mention some stuff that Tulkas did in the Book of Lost Tales (the earliest version of the story), but doesn't seem to notice that that was a BoLT-only part of the story.
My follow-up question, "Did the involvement of the Valar differ in different versions of the story?", intended to tease out the problems in the bit of the answer about Tulkas, resulted in some pretty bad hallucinations. For example, it says that in the Book of Lost Tales, "Tulkas, Manwë, and others were imagined as physically fighting in the War of Wrath." Tulkas did explicitly take part in the War of Wrath in BoLT, but Manwë and the rest of the Valar emphatically did not. Manwë actively opposed the expedition in BoLT and Tulkas, most of the Elves of Valinor, and many of the "Children of the Valar" (i.e. Maiar) defied him and went anyway. ChatGPT also badly overstates Tulkas's involvement in later versions of the story (where he actually isn't mentioned at all) and brings up some stuff that I'm pretty sure is hallucination about Ulmo being explicitly involved in some versions.
I recently posted an example of such an encounter with an LLM (I always ask Google Gemini, because I don't want to sign up for an account, but I already have a Google account).
I asked a question taking the form "here is a couplet from a broadsheet ballad - what does the singer mean by these lines?", and noted in an earlier thread that the answer I received was abysmally bad.
But, of note, I got a response in that other thread saying that I shouldn't be calling that a bad answer because it looks like a good answer if you're unfamiliar with the facts.
It's still not clear to me why that should make the answer better.
I'd say it makes the answer worse! Because if it's *obviously* wrong, you're going to catch that and not propagate it, but if it looks plausibly right, you might be fooled into thinking that it's trustworthy unless proven otherwise.
Along with other types of nerdery more commonly represented on this blog, I'm also a martial arts nerd, and I've spent a fair amount of time asking ChatGPT questions about martial arts. My takeaway is that ChatGPT is quite familiar with the sorts of names people tend to mention in association with martial arts, the sorts of adjectives people use and which styles are most frequently mentioned, but its accuracy in actually answering even basic and straightforward questions related to the martial arts is much worse than even cursory googling. But to someone who doesn't actually know anything about the subject in question, it sounds perfectly credible,
>I'm very very curious to see what kind of tortured queries they're giving to get such inaccurate results.
It doesn't take tortured queries, see my comments at
https://www.astralcodexten.com/p/open-thread-353/comment/74496967
and
https://www.astralcodexten.com/p/open-thread-353/comment/74504922
tldr - in the first case ChatGPT o1 wound up getting the explanation for the color of CuCl4 2- badly wrong, and I had to lead it by the nose to force it to finally cough up the right answer ( detailed transcript of the session at https://chatgpt.com/share/671f016f-3d64-8006-8bf5-3c2bba4ecedc )
in the second case whatever Google is embedding in its searches (Gemini???) falsely claimed that methane has no planes of mirror symmetry (in the course of giving methane as an incorrect answer to my original query)
Just to be clear: I _WANT_ AI to succeed. I would very much like to have a nice quiet chat with a real-life HAL9000 equivalent before I die. It is probably the last transformational technology that I have a shot at living to see. But it is _not_ reliable (nor at AGI) yet.
If I ask an LLM something that is easy to Google, it is likely to give a response that is close to that answer. I seldom ask an LLM for such things, because I usually try searching first based on likely keywords (I often want to go deeper so I need useful further links, not a tepid summary, and this saves time). I probably have a higher prior on incorrect hallucinations than someone who goes to ChatGPT first.
LLMs are currently more likely to be misleading...but perhaps not by a huge margin. Most of the answers I get to web searches are quite wrong, and usually obviously so. (Most of them are so wrong they're irrelevant.) But I ignore the (blatantly) wrong search responses. LLMs tend to give one answer, and when it's wrong, it often isn't obviously wrong.
Automating human judgement about what is blatantly wrong is what we now need.
He reported that they *were* from a Chatbot (and which one) which is the important part. He gave his source. Most web searches don't yield a verifiable source either, and some of them return invented answers. (Not being invented by an LLM doesn't mean they weren't just invented.)
Approximate confirmation of the numbers from
https://www.nasa.gov/reference/the-space-shuttle/
>After the solid rockets are jettisoned, the main engines provide thrust which accelerates the Shuttle from 4,828 kilometers per hour (3,000 mph) to over 27,358 kilometers per hour (17,000 mph) in just six minutes to reach orbit. They create a combined maximum thrust of more than 1.2 million pounds.
Computers at the time weren't good enough to control the descent to the degree of accuracy required.
The Space Shuttle was a terribly suboptimal design hobbled by political compromises. It is a wonder it looked and worked as well as it did. NASA was already ossifying into a terrible bureaucracy, slowly losing its skills and spirit from the glory days of Apollo. Adding fuel tank recovery and refurbishment would have added years and billions of dollars to the schedule and budget, it was not even seriously considered. Even for the modern SpaceX catching the booster is pretty audacious, and it was maybe one second from a failure, according to the Musk's accidental Diablo sound overlay.
Like launching a second Europa Clipper on a Starship in two years to get there before the first one does.
Alas, Starship can't get past Low Earth Orbit, nobody has an in-space maneuvering stage that can fit inside a Starship and take a Clipper to Europa, and neither of those things is going to change in two years even if you tell the engineers to get started today.
It depends what the downside of the Shuttle not having the specific military mission capabilities would have been, had the occasion to use them come up.
There are two parts of this:
1. Could a non-shuttle launch vehicle perform the mission?
2. How much of a luxury was the mission, i.e. what happens if we can't do it at all?
For 1, I understand the answer was mostly yes. Launching large spy satellites (which I understand to be the main driver for payload size and polar orbit capabilities) wound up mostly being done by disposable boosters (Delta and Titan, IIRC) anyway.
The main leftover mission I'm aware of that other launch systems couldn't do was to snatch a Soviet satellite out of orbit and return to Vandenberg. I am not familiar with the thinking that this would be a message or desirable thing to be able to do, so I will tentatively classify it as a luxury mission.
Right, that is a good point. A better (and more expensive during the design stage) approach would have been having a configurable setup where, like with the SpaceX Falcon Heavy, the boosters could be recoverable unless the mission profile forces them being expended.
How many enormous projects are ruined by requirements which are decided in advance, which turn out to be unachievable, but which then can't be changed later on once we learn more? It seems like the answer is "most of them".
If Starship had stuck with its original specs, it wouldn't have worked -- they needed to try a few things and figure out what was practical and what was not. On the other hand, Elon doesn't have a flawless record here either, and the Cybertruck suffers from similar problems where it's a worse vehicle than it would have been if they hadn't made certain dumb commitments at the planning stage.
In addition to what others are saying, the big orange fuel tank was actually the cheapest part of the Shuttle, by far. It had no engines, so it was basically just expensive pipes, tanks, and insulation.
In order to make that reusable not only would significant weight be added, it would also make it more expensive. The savings in reusability would have been more than cancelled out by the lost payload and refurbishment costs.
That is a very difficult technical problem that wasn't technically possible until SpaceX made it work it 9 years ago.
I don't think anyone else does it even today?
As I recall (from reading a fantastic book on the history of the Challenger disaster which I recommend here without any reservations https://www.amazon.com/Challenger-Story-Heroism-Disaster-Space/dp/198217661X); the original plan was to have two part launch system, where the shuttle is first flown on a carrier to a suitably high altitude and then launched from there for whatever it's mission was.
Both parts were envisioned to be re-useable but the cost was well, astronomical.
In this case, the chatbot got it right - the external fuel tank carries all the propellant the shuttle's main engines will use taking the shuttle all the way to orbit (well, except for a small circularization burn with the maneuvering thrusters). So the tank can't be discarded until the Shuttle is at orbital velocity, roughly 8 km/s. At that point, there's no question of it coming back to the launch site or parachuting into the ocean anywhere near the launch site; it's going to come down halfway around the planet.
And it's going to be subject to the same sort of reentry heating environment as the Space Shuttle itself. A simple aluminum tank with just some spray-on insulation to keep the propellants chill before launch, is not going to survive that. A tank which could survive that, would probably weigh enough that the already-marginal Shuttle couldn't carry any actual payload (and certainly not the big military spysats that were part of the requirement).
The only remotely sensible proposal for reusing the Shuttle external tanks was to take them *all* the way to orbit, and then use them as pressurized habitat or propellant-storage elements on a large space station. A single external tank would have more interior volume than all the pressurized elements of the current ISS combined. But nobody had the budget to build a space station that big even if they got the pressure vessels delivered to orbit for free, and their orbits would have decayed long before NASA got around to using them, so they just ditched them in the ocean instead.
The space shuttle fuel tank was just a big tank, it didn't contain anything capital-intensive or fancy like advanced rocket engines. Those were on the Shuttle itself, and those were recovered.
Even recovering the SRBs didn't make sense, because they were only "reusable" in a marketing sense: the cost of fishing them out of the ocean and refurbishing them was greater than the cost of just manufacturing additional SRBs, but reusability was one of the justifications for the expense of the shuttle program so reusable that was deemed.
I mean, the simple answer as to why they couldn't recover the fuel tank was because the entire launch stack was designed around a set of premises, and one of those premises is that that tank was going to be jettisoned and break up on reentry instead of being recovered, and if they'd wanted to recover it that would have required a fundamentally different spacecraft than the one they designed. The Super Heavy Booster is an entire rocket, with engines and electronics and computers and cameras and radios and miles of wiring and sensors. The external tank was just a tank.
It never really got past power point engineering as far as I can tell, but ULA had a proposal for Vulcan that involved basically detaching the engine section and recovering only that - for basically that reason, the majority of the cost of the rocket is the engines and avionics, while the tanks (basically just big empty aluminum cans) are bulky, kind of delicate, and therefore hard to recover.
What would happen to our society if a large-scale, long-term blackout occurred? There is a high chance that it would get quite bad very quickly. Transportation and health services would likely cease to function within a few days, and many people would face food and water insecurity almost immediately. This highlights the urgent need for greater investment in preparedness, as there aren't even exercises to train those responsible for managing such crises. If you're interested in more details, I have written a new post in my living literature review that offers a deep dive into the consequences of blackouts: https://existentialcrunch.substack.com/p/the-consequences-of-blackouts
> What would happen to our society if a large-scale, long-term blackout occurred? There is a high chance that it would get quite bad very quickly.
I think you've independently reinvented the best argument for "prepping."
It's less for the Big One or Zombie Apocalypse and more for longer stretches without power and services brought on by severe weather and / or state inadequacy.
Also the strongest argument for your own solar + battery setup, re Scott's last post.
I like that you're optimistic and think something can be done collectively - I personally believe the only "real" solution is personal and independent, because you can actually prep and get your electricity off the grid through your own efforts without having to persuade a lot of other 'general public' people who don't believe in thinking ahead.
Yes having a backup for smaller catastrophes makes sense, but I generally think that for everything taking longer than ~ 2 weeks, is is important to be able to rely on the state. Otherwise, everything will turn quite bad.
Generally, I don't think long term prepping really works. I've been looking into global catastrophes and societal collapse a lot and my conclusion is that prepping for a duration longer than two weeks mainly buys you the privilege of dying a bit later than the rest.
I think that if the society you live is generally unprepared, most people would die. You might live longer if you are prepped, but I doubt it would be be a great life to have. But, we could plausibly make it through large catastrophes, if we take the time to prepare. And generally, I think we are going in this direction, albeit slowly. For example, more and more countries are starting to consider large scale catastrophes in their risk assessments.
This is a common fantasy but just seems like a way to make your eventual death more cinematic. I've seen Die Hard and Home Alone too, but I don't like my chances against an armed gang, because the defining aspect of gangs is that they outnumber me. I mean, *maybe* I can use my superior planning and knowledge of the terrain to my advantage and fight off dozens of dudes on my own, but that sounds like a fantasy rather than a plan.
If it comes down to a world where we're having gunfights over the world's remaining food resources then I'm probably dead. But there's a bunch of far more likely scenarios where having a bunch of sensible preps will turn a horrible situation into a much more comfortable one, even if the people without preps aren't dying, they're just spending their days standing in line waiting for supplies.
I'm not a real prepper, but I think part of prepping is also training. No matter how much food you have stored, you'll run out eventually. Do you have the skills to obtain more, whether by growing, finding, hunting, or something?
Same goes for everything else: heating, cooling, washing, repair work, etc.
In Gaza, the optimal approach to prepping would probably involve finding out a lot of intelligence about Hamas and then offering it to Israel in exchange for passage to elsewhere.
At some point, you're prepping for the collapse of civilization and that's beyond your resources. But a couple weeks without power is not the end of civilization, it could happen, and having some notion of how you'll heat your home/charge your phone/cook your meals is probably smart.
A comment in support: a few years ago, we lost power for 11 days after an ice storm. Now, the greater area we were in didn't lose power for nearly that long, but because we live in a rural area with a lot of trees, there were hundreds of line breaks and it took a long time to fix them.
In that scenario, the best prep seems to be a car and some money.
Depends. I can see solar panels and batteries, and not doing much at night for a couple of weeks. It wouldn't even be that bad. You should have a bit of food storage, but that should be easily doable in a rural environment (i.e. space wouldn't be a problem).
What do preppers do about water? Seems like that's the biggest barrier if there's a long term catastrophe, unless you live but a river, and it's hard to store a month's worth of water like you can with food
Storing water treatment supplies is easy. Getting access to actual cleanable water is harder (most preppers, I'm guessing, live in rural areas where that's a lot easier)
Not really. https://theprepared.com/homestead/reviews/best-two-week-emergency-water-storage-containers/
Hm, looks like you can last longer than I originally guessed off water containers (though it's still probably the bottleneck).
Oh, what I meant is that a month of water needn't take up a huge footprint. I purchased the The Prepared's Top Pick (https://www.campingworld.com/reliance-rhino-pak-heavy-duty-water-container-203153.html) and the three containers (a 16 day supply) takes up a small corner of the floor in the back of my closet. Another three is utterly doable.
Oh, and I live in just 486 sqft.
FWIW, best practice is to fill them with standard tap water, maybe add a tiny bit of plain chlorine just to be super extra safe, and then dump and refill them every 2-3 years. Because I'm lazy and they weigh 40 pounds, I didn't refill them for 4 years. Then I had to empty them due to water damage from a broken sprinkler and decided to taste the water just to see how it was. It was perfectly "fresh," tasted exactly identical to running tap water.
Many preppers live near streams or springs (or have wells) for this reason as well; small-scale water filtration is remarkably cheap and extremely effective these days, so you don’t really need to worry about the centralized distribution aspect as long as you can access any water at all. The exception would be in the case of significant contamination from pesticide runoff, or industrial chemical dumping, or viral load… none of these are common concerns in the USA.
I get my water from a well. Admittedly I need a bit of electricity to pump it out.
Wells seem like rivers, great if you have one handy but hard to generalize
I can’t really imagine an instance where someone is prepping without a well though… like, are there preppers that have prepped homes in the city??
Sometimes people who live in the city like to be prepped too
On any given day I have a 16-20 day supply of water *and* a reasonably stocked bug-out-bag, Plus a gun and ammo to defend them/myself in my condo, which is 1/2 mile outside of literal downtown Seattle.
I am not at all interested in living in the aftermath of a civilization-ending collapse, so if it looks like it's going that way, that gun will have a different purpose. But I don't see any reason to needlessly suffer or potentially die during local or moderate disasters if I can either literally shelter in place until the lights and water are turned back on, or hike out to an area where I can then be evacuated to my family in california, or friends elsewhere in the country.
If you ration it to a gallon per person per day (plenty for drinking and cooking, but very little left over for washing), a standard tank-based home water heater holds 1-2 person-months of water. It relies on the main water supply's pressure to push it to the taps, but even without that you can still draw water from the drain valve at the bottom of the tank.
If you have a little bit of notice of an impending natural disaster that might interrupt water supply, you can also fill up your bathtub and store additional water there. Although if you use tub water for drinking, the standard recommendation is to add bleach (1/8 tsp per gallon) or water purification tablets to sanitize it.
Relying on a water heater and/or bathtub for potable water in an emergency is ill-advised! Certain kinds of natural disasters or deliberate attacks are likely to taint the water supply coming into your water heater, and even if it doesn't, you have to have both equipment to drain the heater and more storage containers if you need to evacuate.
Yes, it's not a complete solution. I bring it up because there's a wide range of situations where using your bathtub or water heater will give you a fair amount of additional potable water, or at least water suitable for washing.
My water heater doesn't take any special equipment to drain. There's a standard hose bib on the bottom of the heater. I turn the knob and water comes out. If you're using it for its primary intended purpose (flushing out sediment from the bottom of the tank or emptying the tank for maintenance), you want to attach a garden hose first so you can direct the water so it goes to the storm drain instead of flooding my garage. But if you're tapping it for drinking water you just need a clean container to catch the water.
I was going to joke about drinking swimming pool water but I looked it up and apparently that's actually fine as long as you boil it.
Read _Dies the Fire_ by S.M. Stirling. Of course, in the scenario of that novel, not only did the power go out permanently, but all high-energy chemical reactions stopped working. So firearms stopped working, too. YMMV if it's just a long-term power outage. ;-)
Thanx for sharing the link, I learned a lot.
Thanks for the positive feedback!
Seems like there might be one just starting in Cuba. Might also be one this winter in Ukraine.
Agreed, and I think large-scale USA blackouts might happen shortly before or on election day. I can't think of a better time to for foreign actors to sow further discord in the US, if not provoke an actual civil war, and my understanding is that it's relatively easy infrastructure to attack.
Also, I'm a superfan of theprepared.com, which is basically the Consumer Reports of emergency preparedness products, as well as having a lot of general education.
Folk should start here: https://theprepared.com/prepping-basics/guides/emergency-preparedness-checklist-prepping-beginners/
One Second After is a good book about what might happen to society if an EMP wiped out all electronics.
A relevant series of blog posts from the Argentina economic collapse in 2001
http://ferfal.blogspot.com/2008/10/thoughts-on-urban-survival-2005.html
Thanks!
i’m reposting here a question from the old expiring open thread that didn’t get much response, which I’m hoping is because jt went up after the thread went stale.
I know many democrats who are genuinely terrified of Trump winning. The say he will actually do all the worst and wildest stuff he’s just talked about doing so far, because he will quickly install appointees everywhere. He will quickly become a dictator, jailing those prominent people who speak out against him or take legal action against him. I would like to know what impediments there are to this happening.
Last night I googled “impediments to Trump ruining country” and every single hit I got was about how Trump is for sure going to ruin the country. A typical one was a Washington Post article “A Trump Dictatorship is Increasingly Inevitable.” Inevitability doesn’t even *come* in degrees. WTF?
For instance, what impediments are there to Trump’s bringing some charge against Chuck Schumer, winning the case, and getting the guy locked up for a few years? So one thing I wonder about is about legal constraints.
Another is practical constraints. For instance the logistics of rounding up 10 million or so undocumented immigrants seems pretty daunting. Who rounds them up? Where do you put them til they are shipped out? How do you transport them back to where they came from? What do you do if their country of origin won’t take them back? For that and for other proposed Trump plans, it seems to me you need staff who are not only willing to carry out such plans, but are also skilled
at the logistics involved. Seems likely potential Trump appointees are carefully vetted for loyalty, but not for skills (to vet for skills, you need skillful people).
A third is resistance by local governments. What happens if the government of a blue state objects strongly to some Trump plan being carried out? Seems like many things require the cooperation of local officials. Wouldn’t some states refuse to cooperate in carrying out plans they are strongly opposed to? Would Massachuetts, for example, help round up the undocumented immigrants in its state? And then there would be some sort of legal fight about that, but surely it would drag on for a long time.
Just to be extra clear in these contentious times, I am not interested in debating anything having to do with this topic. it’s a request for information.
Later edit: I am horrified by the way people are answering. Dozens of replies have rolled in with no substantive content, just general impression kinds of stuff: People worried last time Trump was president and nothing that awful happened, others in government will restrain him, Trump actually believes in democracy, the Democrats have done all the bad shit they say Trump will do, no they didn't, you are a meanie asshole, you are a poopy head, you eat boogers for breakfast, etc etc.
Finally Daniel actually named 2 things that are impediments. I wish he had had more to say, but his was a genuine answer.
All you smart and usually admirable people are melting down so badly you can't even remember the question. This is the worst I have ever felt about ACX.
It doesn't matter whether you hope there will or will not be lots of impediments to Trump taking the steps he wants to once elected. The question is what impediments will he encounter if he attempts to carry out the policies the Democrats are most strongly opposed to: Legal impediments, practical ones, pushback from local governments, pushback from powerful interest groups.
Still later edit: I looked around some, and this article seems pretty intelligent to me, and honest. Though of course one never knows for sure. It's by a Turkish journalist:
https://www.politico.com/news/magazine/2024/01/28/a-trump-dictatorship-is-possible-but-not-in-four-years-00137949
It's worth noting the Ukraine/Russia conflict was merely less intense during the Trump administration, but went on the whole time. I don't think it makes sense to imply that he deserves credit for the fact that Russia didn't increase the tempo until later, particularly when his own actions mostly involved delaying or threatening to cancel support for Ukraine, and since then he's contributed to a collapse of pro-Ukraine sentiment in the Republican party.
Also he's openly advocating for Israel to go all-out against Iran.
>Trump has much more respect for Democratic institutions than his opponents.
Did he not try to con his way out of losing an election? I agree with your general thrust, but let's not get carried away.
2020 was just a continuation. Trump said the election would be stolen in 2016. Previously he said the 2016 Iowa caucus was stolen. He was asked repeatedly if there would be a peaceful transfer of power and he declined to answer.
There is no world in which he would admit 2020 was legitimate.
"Well, they’d just watched the establishment unite against their guy for four years, impeachment based on dubious pretenses, Russiagate based on the Steele dossier which was transparently a DOJ plant, Obama spying on the Trump campaign, etc. And the media refused to admit that any of it was wrong."
As well as all the back-patting stories later published in the same media about "yes, we *did* co-ordinate to make sure the election was won by the right side".
Shadow campaigns are not anti-democracy when *we* do it!
https://time.com/5936036/secret-2020-election-campaign/
"Their work touched every aspect of the election. They got states to change voting systems and laws and helped secure hundreds of millions in public and private funding. They fended off voter-suppression lawsuits, recruited armies of poll workers and got millions of people to vote by mail for the first time. They successfully pressured social media companies to take a harder line against disinformation and used data-driven strategies to fight viral smears. They executed national public-awareness campaigns that helped Americans understand how the vote count would unfold over days or weeks, preventing Trump’s conspiracy theories and false claims of victory from getting more traction. After Election Day, they monitored every pressure point to ensure that Trump could not overturn the result. “The untold story of the election is the thousands of people of both parties who accomplished the triumph of American democracy at its very foundation,” says Norm Eisen, a prominent lawyer and former Obama Administration official who recruited Republicans and Democrats to the board of the Voter Protection Program."
"They got states to change voting systems and laws". No big deal, not autocratic at all!
"Sometime in the fall of 2019, Mike Podhorzer became convinced the election was headed for disaster–and determined to protect it.
This was not his usual purview. For nearly a quarter-century, Podhorzer, senior adviser to the president of the AFL-CIO, the nation’s largest union federation, has marshaled the latest tactics and data to help its favored candidates win elections. Unassuming and professorial, he isn’t the sort of hair-gelled “political strategist” who shows up on cable news. Among Democratic insiders, he’s known as the wizard behind some of the biggest advances in political technology in recent decades. A group of liberal strategists he brought together in the early 2000s led to the creation of the Analyst Institute, a secretive firm that applies scientific methods to political campaigns. He was also involved in the founding of Catalist, the flagship progressive data company."
Wait wait wait - I thought Cambridge Analytica was the devil's handmaiden for doing the same kind of thing? But of course, they were on the 'wrong' side, while the "secretive firm" was working on the 'right' side this time!
<quote>"They got states to change voting systems and laws". No big deal, not autocratic at all!</quote>
would it depend on the law?
<quote>Why would they think that?</quote>
There really should be 2 columns in your response to this.
I get the impression he just couldn't stomach admitting he lost, not that he tried to stay in power. Compare what real dictators do when they attempt a coup: the military arrests the opposition.
I specifically didn't use the word coup, because, while it might technically be one (although maybe not, I've seen argued that the definition of coup is fairly stringent), it'd a very mild version of one.
But I don't think we can just give him a pass for a pretty naked power grab, particularly with reasons like "he didn't really mean it, guys".
If Trump actually beleived he won and was cheated, his actions take a different tone. I think it's plausible Trump beleives he won because much of his persona is built around magical positive thinking around his capabilities and what the world owes him. Also, there is a >0% chance he *was* cheated in the 2020 election, but of course we will likely never know for sure.
I remember when January 6th happened, my thoughts were something like, "oh, they're finally doing something about being ignored, but this won't end well."
I haven't done in-depth research to find out what election challenges were placed and what they said was wrong with them, but I heard nothing of substance about it from news media, except that all challenges were refuted. The instances of voter fraud that I HAD heard about, such as finding some military mail-in ballots that wouldn't have been counted, were only in favor of Democrats (I had heard of only a few hundred such ballot misplacements).
I mean personally if he honestly believes it because his ego and magical thinking put him wildly out of touch with reality, I don't really see that that's much better than a cynical power grab.
>Biden didn’t talk much about prosecuting Trump, but his DOJ has certainly done so. If you want to know what politicians believe, watch their feet, not their mouths - Trump has much more respect for Democratic institutions than his opponents.
Biden does not control the DOJ and what the DOJ went after are crimes connected to the fake elector scheme which Trump was actually involved in. The DOJ also went after Biden's son and Biden did not stop it, did he? Trump however has pardoned Manafort, Stone and Kushner. All people he has direct ties to. The fake elector scheme alone which led to a number of convictions is tied to Trump directly and is absolutely as undemocratic as you can get.
I think you are either very misinformed on this topic or are willfully spreading wrong information.
This is incorrect - one of the impeachments of Trump was about him asking the premier of Ukraine to give him dirt on both Joe Biden and Hillary Clinton (the latter based on the nonsensical conspiracy theory that her email server was hidden in Ukraine). He also appointed a special counsel just to investigate the prior investigation of himself. He has also, repeatedly, suggested that several of his political opponents are guilty of treason and gone so far as to suggest that the national guard or military should deal with them. https://www.npr.org/2024/10/21/nx-s1-5134924/trump-election-2024-kamala-harris-elizabeth-cheney-threat-civil-liberties
Oh come on now. What in Trump's history or personality suggests credibly that he would actually _de_ politicize the DoJ? Talk about drinking kool-aid, you're swimming in it there.
He dropped prosecuting HRC simply because he hadn't figured out that he'd need to have put his own loyalists into the top jobs; know he knows better.
No arguments here about Clinton or Nixon. The former should have been removed from office and the latter got away with multiple prosecutable crimes. (I get Ford's logic on the pardon, which was much more specific than simply "everybody needs to move on", but it was the wrong call.)
Just for the record though Trump was not charged or convicted for banging the porn star or for paying her to shut up about it. Also the current DoJ is absolutely correct to be trying to prosecute him and others for things like the fake electoral-college slates and the phone call to the GA officials and related actions which attacked the very heart of our constitutional system.
> If you want to know what politicians believe, watch their feet, not their mouths
This is not good advice, we must watch both feet and mouths. What politicians say they will do is, more often then not, different than what they actually do, but it's not unimportant or uncorrelated. What they say has cultural power: it slides the window of what is considered acceptable (to their followers) and determines what their opponents will be thinking about.
Trump saying "I didn't actually lose, there was election fraud" could have been a testing of the waters, a check of which powerful people and institutions would agree with him, and thus hypothetically give him support should he attempt some sort of coup.
I'm not sure that's what actually happened, but the fact that it's possible already proves that, when politicians say things, it's not just words, that is an act with consequences. A similar reasoning can be applied to him saying he was going to prosecute Hillary during his campaign: even if he never was going to do it, he was (deliberately or not) creating a cultural ambient where a president prosecuting their enemies is something somewhat more plausible.
Agreed. Historians routinely use "what politicians say" as one of several proxies for what they believe. The trick is to triangulate multiple sources when available and to read "what they say" critically in light of who they're saying it to and what they're trying to accomplish by saying it.
I don't think you can avoid a values debate in answering a question like this, because parts of your question make implicit values judgements, or at least seem to. For example, you seem to be putting "retaliation against political opponents through the criminal courts" in the same rough category as "rounding up and deporting every illegal immigrant". Note that I personally disagree with the latter policy (and obviously the former too...) but lumping "standard conservative policies", "extreme conservative policies", and "dictatorial actions that threaten democracy" together is something people on the left did throughout Trump's first term and it made it all but impossible for anyone to discuss anything with any clarity. Even here, there've been comments saying things like "Trump did several radical unpopular things like refusing to concede the election and appointing the justices that overturned Roe". Any statement like that (listing threats to democracy and standard right-wing policies you disagree with) is certain to make anyone not already on the left stop taking you seriously.
Note that "you" is generic, not about you personally. I think answering your question with any clarity would require very clearly separating "policies that a lot of people object to" from "threats to the democratic political structure itself".
<Any statement like that (listing threats to democracy and standard right-wing policies you disagree with) is certain to make anyone not already on the left stop taking you seriously..
You could still answer my questions, though, even if you don't take me or my frightened friends seriously. Surely the question itself is worth taking seriously. I restate it below, as a nudge to all the people who are answering based on vibes and general impression kinds of things, or else just starting to fight about who's the biggest asshole, Trump, people voting for him, or people voting against him. My question is what impediments are there to (1) Trump doing the things lefties are particularly horrified by. (immigrant deportation being one, I think the others are pretty well known) (2) Trump doing wacko things like nuking hurricanes (3) Trump disabling the election process, and becoming able to stay in power until he chooses to leave, appoints a successor, dies or is overthrown. By impediments I mean laws, practical difficulties, actions by other parts of the government such as local officials, & pressure from large interest groups (eg businesses who want to keep undocumented immigrants here because they depend on their labor).
One concern now is that Trump's incipient dementia is both quite obvious and a good deal more advanced than Biden's was in 2020. [It's important to compare apples to apples on this topic, not Trump now to Biden in 2024 but to Biden in 2020.]
That doesn't guarantee that Trump will now quickly slide into the full-on dementia stage -- real life is that the pace of that has a lot of individual variation and isn't predictably linear, and also as POTUS he'd of course have top-notch medical care. But sensibly it does create realistic odds for a mental-capacity slide greater than the one that Biden has had while in office. Maybe only a little greater, maybe a lot -- no way to know in advance.
If that does happen it will _not_ quickly or certainly lead to Vance taking charge. The VPOTUS has exactly zero executive authority in our system until the moment he gains complete authority, either by the POTUS' death or resignation or by the full cabinet invoking the 25th Amendment. Trump will never agree to resign so long as he can still speak sentences. The 25A has never been tested, and also would a bunch of Trump appointees have the nerve to do that while Trump is still able to walk and talk? Color me verrry skeptical about that scenario.
If the POTUS is a dementia patient and the VPOTUS is not being placed in charge, and unlike the first Trump administration this one really is full of Project2025 contributors in all the top non-elected positions....that's a basis for taking seriously what the latter crowd explicitly wants to do. And having read their detailed written plan, I take that very seriously indeed.
Reminder: if the Pres fights the 25th, it then requires 2/3 of each House of Congress to remove him. There are definitely cases where this would lead to some very interesting decisions for Dem officeholders.
How so? No Republicans in Congress would vote against Trump, and there certainly won't be close to 2/3rds Dems in either house. So for Dem officeholders it's a freebie, they can do whatever they want on that vote without changing the outcome. Vote yes if they represent a hard-blue district, or not if they don't, whatever.
Doesn't matter though because there's zero chance that a majority of a Trump cabinet would do that, and also Vance wouldn't have the balls to sign on.
I don't think you bit. Even if the stuff in the leftie press is not an accurate representation of what Trump will do, I am asking what impediments there would be to his doing it. Do you know? You can consider answering it an intellectual exercise. Even if Trump isn't going to try to various crazy things, it surely makes sense to consider what guardrails there are against some president doing crazy things. The impediments I have in mind are legal impediments, impediments built into the structure of government, practical impediments, resistance from local government, pushback from big interest groups.
What are your reasons for thinking "what Trump publicly says he'll do" bears a strong relation to "what Trump actually wants to do"/"what Trump will do"?
I get the feeling you see much of the criticism of Trump as hysterical and misguided, but I'm sure you've noticed that he does not have a strong commitment to truth-telling.
"businesses who want to keep undocumented immigrants here because they depend on their labor"
I think a lot of the problems around immigration could be sorted out if there was honesty on this: let businesses like construction and agriculture be open about "we need cheap labour; the natives won't do it, or if they will, they expect better wages and conditions than we can/are willing to provide, so we need immigrant workers". If the liberals are horrified about exploiting minorities, then they'll have to accept that the workers can't come across the border because they *will* be exploited if they do.
Either way, set a target of "who can legally enter as immigrant labour" and *stick to it*. People who break the rules get sent back. Those who follow the rules get a chance for citizenship or at the least, working their way up into better jobs. Decide if Jose the fruit picker gets the vote after living and working in the US for ten years and both sides *stick to that*, no blurring the rules that technically Jose shouldn't vote but... he has a driver's licence and we don't impose ID rules on voters so if he does vote, naughty Jose but nothing is done about it.
I think mostly people are pissed off about the "illegal" part, not the "immigrant" part, despite the attempts to paint any opposition as racism. Put structures in place, let the people see that those are being implemented, and you'll cut back a lot of the objections. Jose follows the process to come here and pick fruit cheaply for MegaFarm Corp, there's a better chance he can integrate into the local landscape. Felipe comes here illegally and gets booted back over the border, everyone agrees the rules are being followed, nobody has a problem with Jose being here and Felipe not being here.
>By impediments I mean laws, practical difficulties, actions by other parts of the government such as local officials, & pressure from large interest groups (eg businesses who want to keep undocumented immigrants here because they depend on their labor).
Laws are very important, but largely an imponderable. AFAIK, every recent administration has done at least _something_ that is arguably illegal or unconstitutional. And yet, I can't tell at what line Federal employees would refuse to follow illegal orders.
One practical constraint that I'm reasonably sure of: There isn't enough prison space in the USA to imprison all of the voters voting for the losing party. Peak total incarceration in the USA in roughly 2008 ( https://en.wikipedia.org/wiki/Incarceration_in_the_United_States ) was about 2.3 million. This is more than an order of magnitude less than e.g. the roughly 81 million votes for Biden in 2020 (and similar numbers for Trump).
Might Trump dissolve Congress? Might Harris suspend the first and second amendments? I have no idea. Both are clearly illegal and clearly physically possible. Would Federal employees go along? I have no idea.
Thanks. Those things you have no idea of? I have no idea of either. And I'm not sure many people arguing about the election do either. And yet people are turbo-powered and savage in their certainty about what should happen and what will happen.
Many Thanks!
ES: 2 cents, educated guess, not very familiar with Trump's administration.
We live in times where, frequently, a politician who explicitly and/or implicitly has some crazy ideas gets elected as head of state. The usual pattern seems to be that, once in power, they actually do implement lots of crazy policies and make crazy decisions, but way, way less then both their followers and their opposition thought they would. It seems that was the case with the first Trump administration: he neither became a dictator, like some anti-MAGA people said he would, nor did whatever the Q-Anon people thought he would do. But he did build around 52 miles of new primary border barriers. (This is not to understate the consequences of the Trump administration, just saying, it's less then what he seemed to promise during campaign.)
I think that, aside from legal limitations to the power of any politician, there is a big social-cultural factor that keeps them from doing anything too out of the Overton window: behind the curtains, there must be a set of people that tell them that it's a bad idea. I imagine Trump must have heard the phrase "sir, this wouldn't work/isn't possible/isn't viable/doesn't work like that" multiple times per day when he was president. I have the impression that this might actually accounts for a significant amount of the total impediments there are to politicians going mad with power.
My model for this is something like the following: assume some president would like to become a dictator (you can substitute that with "implement some radical policy they like"). A politician is either:
- type A: believes that, if they just did what they actually wanted to do, their opponents would use a lot of resources to remove them from power, and not enough of their allies would back them up to preserve them in power.
- type B: does not believe the above.
If the president is type A, they don't try to become dictator. If they are type B, they necessarily need to consult some of their allies and underlings in order to attempt some sort of coup. If an ally is type A, they advise against. If they are type B, they consult their own allies and underlings, and the process continues recursively.
> The usual pattern seems to be that, once in power, they actually do implement lots of crazy policies and make crazy decisions, but way, way less then both their followers and their opposition thought they would.
This seems like the kind of heuristic that works until suddenly it does not. Currently in Slovakia we have yet another government of Robert Fico. Previously he was the kind of guy who has big mouth, but luckily after election only does maybe 1/4 of what he promised. But not this time. He reformed the criminal law practically the day he got elected (reduced the status of limitation of many kinds of crimes; mostly economical but also e.g. rape), and fired all prosecutors and policemen who in the past acted against the members of his party. Crimes bosses that it took literally decades to put in prison are now walking free. He also established some new taxes, and other minor stuff I don't remember. Shortly: mostly harmless, until suddenly he was not.
I suspect that the most dangerous moment of a populist politician is when he starts losing popularity. When he can no longer be certain that he will win another election... even if not the next one, then definitely the one after that. Especially when it seems that if he no longer has the immunity and power to stop prosecution, he might serve some prison time. Then it may turn out that until now he was playing on easy mode (just telling the voters what they wanted to hear, and mostly doing nothing), and now he is playing on hard mode (trying to destroy anyone who looks at him the wrong way, and desperate enough to take any kind of allies).
Just a minor quibble, but
> […] is Increasingly Inevitable.” Inevitability doesn’t even *come* in degrees. WTF?
Come on, it obviously means "approaching inevitability".
When someone says, "my trash can is increasingly full", they mean that it will soon have exhausted its capacity for additional trash, even though "full" is a binary condition, like "inevitable".
Yes, I understood what they meant. But there are adjectives that denote a certain state that does not exist in partial forms -- for instance "infinite," "pregnant," "dead." It irritates me to see words used in ways that blur their meaning. Their architecture is lost, their crystal structure. But my objection to the Post headline is mostly based on something else. It seems like a way to summon maximum horror without actually committing to a prediction. Sort of like saying "one day soon blood may come spurting out of your mouth and ears."
Well and humorously said.
The government was able to deport 100,000s of immigrants in the 1940s, it is interesting that most people don't think the government have the state capacity to do that in the 2020s.
https://en.wikipedia.org/wiki/Operation_Wetback
I'm not sure the government doesn't. But organizing it does sound daunting. Covid made me much more aware of things like the difficulty of transporting things to large numbers of people -- for instance there was a whole nasal swab debacle. Reading up on that stuff, I came to see that for large tasks the government does not routinely do, somebody in government has to locate the people here and there, often civilians, who happen to have the required skills and info, and draft them to come work on the current project. So what I was wondering about was how likely it was that the people Trump's backers have found to install who would manage such a deportation have been selected for the ability to manage the task, or locate specialists who can.
I don't doubt there's going to be a lot of "this is paranoid" in response to this, so before anyone jumps to that I am by no means claiming this scenario is likely - just taking OP at his request of "how the worst could work" and leaning into what "the absolute worst" could look like.
So, its 2027, and Trump is dropping hints to staff that he'd really rather stay in power. Maybe Don Jr is doing his best Hunter Biden impersonation, making a ton of money selling presidential access to foreign governments and corporations, maybe there's some family scandals that he's able to keep at bay as president but won't have the leverage for the moment he leaves office, maybe there are some bigly projects around the border that he just doesn't trust a successor to close the deal on - maybe, like all bigly men he doesn't trust any successor- maybe he just likes being the center of attention. But whatever the reason, he starts mentioning to staff that the two-term limit is bullshit and smart government just has to be more important anyway. Someone leaks it to the press, but when it hits the news smart people on boards like this assure everyone that it's paranoid liberal fantasy, Trump would never do that and at any rate he can't because lots of checks and balances from the administrative deep state to the courts would stop him.
Then, some plucky lawyer floats a memo Trump's way, along the lines of the Eastman Memo. https://edition.cnn.com/2021/09/21/politics/read-eastman-memo/index.html. The thrust of the memorandum is that although the text of the 22nd amendment clearly states that "[n]o person shall be elected to the office of the President more than twice," it makes no clear ruling, and there is no supreme court precedent, regarding whether the same person can remain President in cases of a *national emergency*. And this is something for which there is plenty of precedent from around the world, hasn't that darling of the Democrats, Volodymyr Zelenskyy, been governing in a perpetual state of emergency without elections since 2022?
The more "establishment" contingent in Trump's staff initially dismiss it, but initial polling for JD Vance is abysmal compared to various expected Democratic challengers. Not only that, but one or two of the more aggressive-minded of Trump's staff make sure the memo gets to him, and now he's pushing his team in private to get behind the memo. In public, he's making much more brazen declarations about various "emergencies" which might interfere with operation of a proper election. Once again the liberal talking heads, once again the very smart right-leaning commentators explaining how this is paranoid leftist fantasy. Trump derangement syndrome.
But it gives the Democratic candidate a boost in several key swing states. Now faced with a potential loss in upcoming elections, people within the White House begin asking whether they can afford *not* to try something like the National Emergency approach. After all, what will the left do once they have the levers? Last time Trump left office, they tried to prosecute him.
Trump announces suspension of elections for the 2028 cycle, because the number of illegal immigrant voters has made a proper count of votes impossible. Blue states object and attempt to hold elections anyway, but there's no republican candidate on the ballots there as the party falls in line behind Trump and refuses to nominate a candidate. Red states boycot the election. Purple states are a mixed bag - in cases where Democrats hold power, elections are held, sometimes with JD Vance on the Republican ticket, sometimes with Trump; in cases where Republicans are in more control, purple states are less likely to hold votes, or to declare that their Republican-majority legislatures will select their electoral slates, which is only responsible given that the President has declared an emergency for the precise reason that elections can't be trusted right now. As a result, slates of electors arrive at the Capitol in a state of fairly complete chaos. 160 electoral votes are absent entirely. The Democratic candidate has 268 votes, while the remaining 110 votes are split between Trump and Vance.
Democrats claim that they have a supermajority of the electoral votes present. Republicans claim that there was no fair election held - you can't just give the presidency out when nearly a third of states didn't vote at all, that disenfranchises tons of Americans, and besides, the number you need is 270, and the Democratic candidate did not hit that number. Democrats retort that of the 110 votes for Trump or Vance, nearly half were selected by Republican legislatures - a real *vote* in those states, some of which were purple to start, would have yielded more than enough to put the Democrat over the top.
When Congress convenes to count the electors, Vice President Vance declares the fiasco too much. There's been no valid vote, so the proper thing is to let the House of Representatives assign the presidency, in keeping with the constitution. He uses the phrase "assign the presidency" rather than "decide the election" intentionally. As stated in the memo, the fine print of the constitution is "[n]o person shall be elected to the office of the President more than twice." He says it will be up to the House to assign the presidency, then up to the Supreme Court to decide whether this assignment constitutes an "election" for purposes of the 22nd Amendment. "We are all following the constitution as best we can in these unprecedented times, so I urge those unhappy with this result to follow constitutional order and seek redress in the courts."
The Republican House dutifully assigns the Presidency to Trump. Democrats predictably bring a variety of lawsuits, and as predicted in both the Eastman and National Emergency Memo, SCOTUS declines the cases on the ground that the issue is a "non-justiciable political question." Welcome to your new autocracy. Of course, Trump's loyalists are extremely offended when someone calls it that. Some assert that there will be a new election as soon as the emergency ends, others claim it will happen in 4 years, but they are uniform in their confidence that everything happened according to the constitution; at most it was "lawfare," but nothing that justifies the hyperbolic claims of extreme leftists that it was a "coup" or something befitting a tin-pot dictatorship. And in any event, even if it *was* a coup, it's still Democrats' fault since their leadership prosecuted him last time he left office. From a certain point of view, he can't afford *not* to hold onto power in perpetuity.
So there's a sample fact pattern. It's incredibly low-likelihood, and I'm sure any Trump fan reading it is likely to declare it totally deranged fantasy. But then, I would have thought the fake electors plan to be deranged fantasy before Trump tried it, and I guess the key question is, "deranged as it is, is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?"
To my mind, that's the key question, and for all other candidates in my lifetime, I've been so comfortably assured that they wouldn't do something as crazy as this whole screed that I haven't had to factor it into my vote. But now I have a candidate where, instead of just a simple "no, they wouldn't do that" I have to go with "extremely unlikely he would do that - he's old, probably wouldn't want the presidency anyway, and there are lots of checks and balances in place that would block it, and..."
And once you find yourself rationalizing how unlikely it is that a guy would eat your baby, how many checks and balances are in place that would prevent him from eating your baby, how he probably wouldn't even want to do so, etc, etc... when you find yourself having that conversation about the guy, instead of just taking as an obvious given that he wouldn't eat a freaking baby because that would be insane, that's a situation in which you pick pretty much any other babysitter imaginable.
What is your opinion of the 2016 appeal for faithless electors? I'm not trying to wage Culture War, this is a genuine question. I thought it was ridiculous and it seems to have ended in, at worst, egg on some faces - the couple of electors who *did* go faithless dumped Hillary, not Trump.
https://en.wikipedia.org/wiki/Faithless_electors_in_the_2016_United_States_presidential_election
On the other hand, it was an attempt to overturn the results of the election, calling on electors to pledge their votes in opposition to what the result in their state had been. So was that a coup attempt or not?
There seems to be forgetfulness about there having been a lot of "I refuse to accept the result of this election", "not my president", "he only won the Electoral College not the popular vote so he didn't really win". If people can call for the abolition of the Electoral College and replacement by the popular vote, I don't see why calling for the amendment of an amendment about term limits is off the table.
It seems to me that either we go by the literal legal regulations defining the Electoral College and it's function, in which case it seems like the electors are technically entitled to be faithless (at least from some states? correct me if I'm wrong) according to their personal judgement and conscience.
Or we say that whatever the technical rules are, the electors have a duty to faithfully serve what the people voted for based on democratic principle. But if it's the latter, I'm not sure why abstract democratic principle should bind them to the vote in their particular state over the overall national popular vote- the latter seems to me to be a truer expression of the "will of the people" if that's our standard.
Yes, I think either stick with "this is how we decided to select the president" *or* "okay let's abolish the Electoral College".
But you can't have it be "the Electoral College is right if it gives me the result I want but wrong it if gives me the result I don't want". If Harris wins by winning the Electoral College but not the popular vote, I fully expect none of the people crying over how the College is a remnant of slave-holding and should be done away with because it's racist to stick to that principle, even though their candidate won. That's what annoys me most: the claim that "we are doing this out of principle, they are doing it out of self-interest" when it's blatantly obvious that it's self-interest all the way down on both sides.
Supreme Court is ram-rodding through social liberalisation over the objections and stated preferences of many people? Too bad baby, them's the rules and now the New Thing is law and you have to obey the law!
Supreme Court is ram-rodding through social conservatism over the objections and stated preferences of many people? Oh no, it is a racist sexist non-representative body, we need term limits for judges, it should be overhauled! Resist!
Again, if we're sticking with the Electoral College as it's currently established, that includes the electors being actual humans who can exercise some judgement, not just abstracted votes bound to their states. I personally think that system isn't great and a national popular vote would be better, but I don't think it's wrong or hypocritical for people to try to use the current system as it exists to get the outcome they want even if they'd also prefer for the system to be different.
>I personally think that system isn't great and a national popular vote would be better
Personally, I agree. I would prefer a popular vote also.
I'm not thrilled with the idea of "faithless electors". On the other hand, my understanding of what the founders wanted in the first place was for the electors to actually use their minds in choosing which presidential candidate to vote for. There is a decent argument for doing that, but to _actually_ do that, the electors should be running local campaigns along the lines of "Vote for me. I'm great at evaluating job candidates!".
The Electoral College seems the worst of both worlds. We don't get "one citizen, one vote" as we would with a popular vote. We don't get electors using their brains (with presumably more knowledge of government and politics than the average voter) to grill and evaluate the candidates.
The Supreme Court weighed in on it in the wake of 2016; electors are required by law to vote the way the state wants them to, they're representatives of the state not entities unto themselves.
https://crsreports.congress.gov/product/pdf/LSB/LSB10515#:~:text=College%3A%20States%20May%20Restrict%20Faithless%20Electors,-July%2010%2C%202020&text=On%20July%206%2C%202020%2C%20the,the%20voters%20of%20their%20state.
Good to know, although even that only says that states are allowed to punish electors for being faithless, not that electors are by default required to follow the candidate they were selected for.
You see that Colorado was allowed to replace rogue electors during the election, with people who would vote how the state wanted. So, the state can both nullify the rogue vote, and fine the offender.
Because the rules are up to each individual state and not every state legally requires electors to follow the popular vote. I think most states do require this, around 40 of them, but don't quote me on the exact number.
>>What is your opinion of the 2016 appeal for faithless electors? I'm not trying to wage Culture War, this is a genuine question.
No culture war taken, and no issue answering that - I did in the last thread. Cut/paste to save myself the rewrite, and you the lookup:
>I'm perfectly comfortable condemning Clinton's elector scheme in 2016. I don't think she's a person we want anywhere near power either. I was a GWB voter through the 2000s - the first democratic vote I ever cast was voting in the democratic primary in 2008, specifically to keep Clinton off the democratic ticket even though I fully intended to vote McCain in the general. If Clinton was running against Trump this year, I'd be right back where I was in 2016, voting libertarian because all options suck, this time with a dash of "all options suck and both have tried to subvert election results." But Harris is running, not Clinton, so I don't see any responsible use of a vote other than a vote for the candidate I'm confident will transfer power peacefully later over the candidate who as already demonstrated a refusal to do that.
I will supplement that, though, with the observation that what Trump tried is still demonstrably worse than the faithless electors scheme. To hit Trump levels, Hillary would have had to (a) do her faithless electors thing, then (b) when electors opted not to go for it, sic a mob on them to try to force them to heel. To be clear, she did (a), which puts her comfortably in the category of "do not let near power" with Trump, but it's still a faulty comparison, in addition to being the dumb kind of "it should be okay that I shot a baby since another politician shot one too" defense.
>>So was that a coup attempt or not?
I've avoided calling Jan 6 a coup and would do the same for the faithless electors scheme. Those conversations seem to me to be more of a distraction. People get all heated about whether it was "coup" or "I can't believe its not coup," but either way its an attempt at fraudulently putting the candidate who lost the election into the White House. To my mind that's disqualifying, and an argument over whether it's really a "coup" or not because it had a low % chance to succeed, or the military wasn't involved, or whatever is just a sideshow.
>>There seems to be forgetfulness about there having been a lot of "I refuse to accept the result of this election", "not my president", "he only won the Electoral College not the popular vote so he didn't really win".
Statements like that are just sour grapes from voters, and are in a totally different league from actually trying to defraud the electorate and change the election results. I have no problem with Americans of any stripe whining because their candidate lost an election. That's *everybody's* right.
>>If people can call for the abolition of the Electoral College and replacement by the popular vote, I don't see why calling for the amendment of an amendment about term limits is off the table.
Sure, but my fact pattern wasn't an amendment to term limits, it was a seizure of power. I don't think "encourage voters to demand a change to term limits, then follow the constitutional process for pursuing an amendment" is anywhere near the "worst Trump could do" as requested by OP.
thank you for the thoughtful and well-written answer
This is what I wrote about the 2016 appeal for faithless electors at the time:
"In many ways, for the Electors to elect a President on their own discretion (as opposed to acting as mere proxies for pluralities of voters in their respective states) would be comparable to the Queen unilaterally firing the Prime Minister and appointing a replacement of her choosing: in both cases, they have the legal power to do so, and indeed would be acting consistently with the historical intent and purpose of their respective offices, but both cases would be an enormous break with long-standing practice, and both would be widely seen as democratically illegitimate barring truly exceptional circumstances."
Why do you describe your own scenario as deranged? What are the key points that require overcoming poor odds?
>>Why do you describe your own scenario as deranged? What are the key points that require overcoming poor odds?
Combination of things. I think it came to mind because: (a) it's a very extreme and unlikely example, (b) a president would need to be deranged to try it, and (c) I do think a person who says "I believe with confidence that Trump will do this if he becomes president" is letting their fear lead them to dramatically overstate the odds. Something like this, to me, clocks in at "extremely unlikely but it's fundamentally disqualifying to Trump as a candidate that I perceive even small odds of it"
As far as hurdles go, coordination seems like the biggest one. As Mike Pence has shown, insiders can break the chain. SCOTUS, for example, could have its moment and choose not to go along. John Schilling also thinks there's some Article 2 limitations that make it illegal, but I'm not sure what they are - I've asked him for an explanation there so keep an eye on that thread. That still boils down to SCOTUS.
Yeah, this is paranoid. The Constitution has a fair bit to say about executive power in the event of e.g. an insurrection, and it is very specific about who gets to be president under various scenarios where the nominal election-winner can't do the job, and none of it adds up to POTUS getting an extension just because he declared an emergency. And it isn't necessary for every bit of law that touches on presidential succession, like the 22nd amendment, to include a "yes, we really mean it, even in an emergency" clause, for it to apply in an emergency.
We already know who gets to be president in any plausible scenario, including "Trump was elected fair and square in 2024 and it was physically impossible to hold an election in 2028". And it isn't Donald Trump. That's black-letter law, beyond Donald Trump's ability to change, and too much of the executive and judicial branches will abide by that law for Trump to remain effectively in power. If he tries, he'll fail as badly as he did on 1/6.
The Eastman strategy had at least a fig leaf of legality to it; the Constitution does call on Congress+VP to certify the election results and is silent about many of the details on that. But here you're talking about a scheme that is explicitly forbidden by the Constitution, and supposedly rationalized by something that isn't authorized.
>> We already know who gets to be president in any plausible scenario, including "Trump was elected fair and square in 2024 and it was physically impossible to hold an election in 2028".
Honest question, what would happen in a scenario where it were genuinely impossible to hold an election? Looking at Article 2 nothing jumps out at me as an obvious answer.
>>And it isn't Donald Trump. That's black-letter law, beyond Donald Trump's ability to change, and too much of the executive and judicial branches will abide by that law for Trump to remain effectively in power. If he tries, he'll fail as badly as he did on 1/6.
I realize I wrote a novel, but this goes so exactly to the last 3 paragraphs that I just have to repost below.
>So there's a sample fact pattern. It's incredibly low-likelihood, and I'm sure any Trump fan reading it is likely to declare it totally deranged fantasy. But then, I would have thought the fake electors plan to be deranged fantasy before Trump tried it, and I guess the key question is, "deranged as it is, is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?"
>To my mind, that's the key question, and for all other candidates in my lifetime, I've been so comfortably assured that they wouldn't do something as crazy as this whole screed that I haven't had to factor it into my vote. But now I have a candidate where, instead of just a simple "no, they wouldn't do that" I have to go with "extremely unlikely he would do that - he's old, probably wouldn't want the presidency anyway, and there are lots of checks and balances in place that would block it, and..."
>And once you find yourself rationalizing how unlikely it is that a guy would eat your baby, how many checks and balances are in place that would prevent him from eating your baby, how he probably wouldn't even want to do so, etc, etc... when you find yourself having that conversation about the guy, instead of just taking as an obvious given that he wouldn't eat a freaking baby because that would be insane, that's a situation in which you pick pretty much any other babysitter imaginable.
Trump takes this kind of thing from a simple “no” to an argument about all the not-Trump people and systems around him and how strong they are to stop him in the unlikely-but-implicitly-admittedly-uncomfortably-more-than-zero-chance-he-were-to-try. That we get to, “… and if he tried he would fail,” rather than stopping at “no” marks him, to me, as exactly the person you keep away from power wherever and whenever possible.
>Honest question, what would happen in a scenario where it were genuinely impossible to hold an election?<
I'm hard-pressed to think of a situation that could cause that. We had an active plague in 2020 and it didn't stop them.
Assuming some weird situation where timely elections are impossible but there are still enough people alive to sustain a functioning government, the state would probably have an election shortly after the deadline, and the late results would be approved by Congress and/or the Supreme Court as a one-time exception to the rule. The system wants the system to survive.
"Honest question, what would happen in a scenario where it were genuinely impossible to hold an election? Looking at Article 2 nothing jumps out at me as an obvious answer."
I think you have to look at the Amendments. If you're sitting in front of a decently hyperlinked version, the text of A2 will note that some of it is superseded by Amendment 12, and also Amendment 20, where section 3 says:
"If a President shall not have been chosen before the time fixed for the beginning of his term, or if the President elect shall have failed to qualify, then the Vice President elect shall act as President until a President shall have qualified; and the Congress may by law provide for the case wherein neither a President elect nor a Vice President elect shall have qualified, declaring who shall then act as President, or the manner in which one who is to act shall be selected, and such person shall act accordingly until a President or Vice President shall have qualified."
By this, neither Trump nor Vance would qualify; it's up to Congress to decide what to do, and it did that in 1947 with the Presidential Succession Act. The Speaker of the House would act as President, probably until a proper election finally happens.
Currently, that's Mike Johnson, but it might easily be someone else by Jan 2028. One could imagine a lot of anxiety over how long that Speaker would want to stay in the Oval Office and whether he or she would have influence over how long it takes to get that election going.
The other problem in that case is that if something physically kept the US from having an election, it probably also kept us from electing a new House as well - and thus, a new Speaker. So the *Speaker* would have a legitimacy problem. That would depend on other factors, among which is whether the House was expected to flip in 2028, or whether it didn't but the current incumbent was facing internal party challengers. At that point, we're in speculative history territory, although I feel comfortable claiming the chance of Trump ending up with the Speaker's seat being well under 0.0001%.
Makes sense - thanks for the citation. I agree that the scenario we're talking about is a well under .0001% scenario; part of why I used words like "paranoid" and "deranged."
I guess for me it just boils down to how, of all the reasons we keep citing for why stuff like this wouldn't happen, it's all things *external* to Trump himself. It's not that he *wouldn't* do something like this, it's that various reasons and structures *prevent* him from doing something like this.
Makes it seem less like we're choosing a responsible person to drive the car and more justifying picking a known reckless driver (who may or may not have intentionally run someone over once - it's debatable) who keeps talking about all the "sick traffic weave moves" he'll be pulling out there, on the promise that there's a speed regulator, guy riding shotgun, etc who will hold back his constant urges.
The way I'm trying to argue isn't intended to stipulate that Trump is a known reckless driver. Rather, it is to say that *even if* he's a reckless driver, the other safeties are going to hold.
If Congressional elections are also blocked, then there's no Speaker because all Represenatives' terms have expired. At least 2/3 of the Senate is left (possibly more, since many state governors have the power to make temporary appointments to fill vacancies), enough to form a quorum, so there is still a President Pro Tem of Senate or at least the ability to select one, and as John Schilling noted, the President Pro Tem is next in the statutory line of succession.
If there were no Senate either for some reason, the next in line would be the Cabinet Secretaries in order of department seniority (State, Treasury, Defense, Attorney General, Interior, etc). It's customary for the cabinet to offer resignations at the end of a Presidential term, especially when there's a new President, but the law and constitution allow them to remain in office until they quit or get fired. So the old Secretary of State would become Acting President if they're constitutionally eligible for the Presidency and don't resign at the end of the old president's term.
For some reason I was thinking that a Speaker election might not happen shortly after Nov 4 if the new House retained the same party majority (e.g., Nancy Pelosi leads the 116th Congress and the 117th is still Democrat and there are no serious Democrat contenders).
I imagine there'd be a Speaker-presumptive if the House was widely believed to not flip if elections had been held, but I agree the Senate President would be the proper call, in an improper situation.
If there's literally no election at all, then on January 21 the presidency goes to the president pro tempore of the rump Senate. Vice President and Speaker of the House get skipped because, if there's no election, there isn't a Veep or a Speaker. Each Congress is legally independent of the last, even if it's mostly the same guys in practice, and the last congress's term expires before 1/21 so if you haven't had an election yet then you don't have a House. But Senators serve staggered six-year terms, so you can skip an election and still have a quorum in the Senate if you need it.
“Is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?”
Trump would almost certainly not refuse based on the nature of the plan. Trump lacks any core principles other than self interest.
What could inhibit Trump is significant opposition from Trump’s own staff. The part of your scenario where Trump “starts mentioning to staff that the two-term limit is bullshit” captures the way Trump operates. If, after having been given of what Trump wants, a large portion of the staff still opposes the idea, Trump is going to be extremely reluctant to move forward with the plan. For example, Trump supported Jeffrey Clark’s plan to send out letters to states like Georgia that would falsely claim that the DOJ had “identified significant concerns that might have impacted the outcome of the election” and suggest that the state legislature chose the Presidential electors itself. When Jeff Rosen refused to go along, Trump could have fired him. Instead, Trump abandoned that part of the January 6 plot.
>>“Is the above deranged enough that Trump himself would refuse to try it in order to obtain something he wanted?”<<
>Trump would almost certainly not refuse based on the nature of the plan. Trump lacks any core principles other than self interest.<
>What could inhibit Trump is significant opposition from Trump’s own staff.<
That matches my own mental model for how Trump operates; I just can't get over the foolhardiness of the bargain, entrusting power to someone based on the mental model that they will be constantly trying to abuse it but restrained by others.
"Let's summon Skraldex, Chewer of Manflesh, from his Cthonic lair in the 8th Hell to be the next POTUS - he hates wokeness and the swamp as he hates all works of man, and will be incredibly aggressive and dangerous to our enemies once turned upon them. Granted, he cares for none but himself, constantly hungers for our succulent flesh, and would laughingly hurl us into the abyss if freed, but I'm confident the seals will hold."
I think almost nobody who talks like Trump is going to impose a dictatorship and round up the illegals, gays, minorities, liberals, etc., actually behaves in accordance with those beliefs. I mean, if you think there's a 50% chance Trump wins and a 50% chance, conditioned on him winning, that he sends your kid to a death camp for being gay, you should be either living outside the US or frantically trying to.
Yeah, as the Brexit people learned, it turns out a lot of other countries think _you're_ a foreigner who shouldn't be allowed in.
Or you need to declare yourself a refugee who fears political persecution, which from the rhetoric I'd expect many people (on both sides) to feel appropriate.
But even for the more traditional sort of visas, there are tens of millions of Americans who could get a visa to a relatively decent country if they applied, and I'm pretty sure I'd have heard if even tens of thousands were actually doing so for political reasons. So either everyone is waiting for the last minute, which given the delays in processing visa applications would be silly, or the "I'll have to flee the country if Trump is elected!" bit is 99.9% empty posturing.
Also, we can look at 2016 when, yep, it was 99.9% empty posturing.
Most people can't just leave the country. And if you think Trump risks being a Hitler-like figure, fleeing the country before the election makes this MORE likely, so there's a question of moral responsibility.
At least 20% of the population can leave the country.
I haven't heard of anyone moving, but I'm sure some do.
Most people don't live in swing states, so their vote could not possibly affect the election result. Even for others, if you believe there's a 50% chance of Hitler v2.0 taking power in the US, it's hard to imagine you'd stay here and wait to be sent to a camp rather than fleeing the country.
I've never lived in another country. I don't know what I would even have to do to get permanent residency - probably get a job there. Last time I had to get a job it took ~6 months, and that was in the US, which is one of the best economies in the world, and without needing to ask anybody for sponsorship. My whole family and all of my friends live in the US. Everywhere I've ever worked is here. All my money is in US banks. I'm very attached to this country, emotionally, in addition to financially, etc.
People have lots of reasons for not wanting to literally flee the country. Maybe it would be more calculated if I started shifting my assets and planning to move to Canada or something, but fundamentally there's a lot of factors there and I think at a bare minimum it's silly to imply that people don't actually believe Trump might be a fascist-dictator-wannabe because they aren't willing to uproot themselves completely from everything they've ever known. By that logic you might as well ask why everybody doesn't just move out of Palestine or Ukraine.
Or learning to be a better shot than the kid in PA, tbqh
The number 1 impediment is the Administrative Procedure Act. Trump’s agencies can’t change the official Code of Federal Regulations without going through a complex, drawn-out rule making process. These agencies are largely staffed by DC-area permanent civil servants, who overwhelmingly lean Democrat and will attempt to slow-roll or sandbag the top-down orders they receive.
The number 2 impediment is that Trump’s ability to fire these permanent civil servants is restricted by the Civil Service Reform Acts.
Now, theoretically these laws could be repealed by a Republican congress, but at that point the objection becomes, “I don’t think the party that controls all three branches of government should be able to implement its preferred policy.”
You are the first person to actually address my question. Thank you. Can you add any more. For instance, regarding
< I don’t think the party that controls all three branches of government should be able to implement its preferred policy.
Are there any legal or practical things that do act as a check on this happening, though?
I am really dismayed at how people responding to my post are just sliding into general impression kinds of stuff: Democrats hyperventilated the first time he was elected and nothing too awful happened, people who talk like Trump don't really carry out the ideas they're promoting during a campaign, the Democrats have been doing all the bad shit they are now afraid Trump is going to do, Trump actually has respect for democracy, a vague "others in government would restrain him."
This is the worst I've ever felt about ACX. It sounds as dumb as every other forum right now. Nobody but Daniel has even tried to address my questions: If Trump is elected and promptly installs many loyalists, what checks exist on his doing (1) the things Democrats are most strongly opposed to, which I could list but I think are common knowledge, (2) nutty things like nuking hurricanes, which I assume everyone is opposed to and (3) dismantling the election process, so that he remains in power until he chooses a successor or is overthrown.
Regarding the first point, it's not clear to me what keeps the administration from ignoring the law. Sure, they will get taken to court, any court, even the USSC, will say they can't do this because of the law, but there is no one to enforce the law. Well, the the Justice Department, but I don't expect them to start prosecuting political appointees.
Regarding the Civil Service Reforms Act, Trump circulated an order in October 2020 to strip protections from tens of thousands of permanent civil servants. This is plausible because the law exempts civil servants that the administration decides are political. This order (schedule F) has been mentioned openly during the current campaign as a way to drain the swamp - note that it also solves the first impediment, because all those agencies can be repopulated by Trump appointees.
I think the biggest concern really is the one mentioned above though. That first, the USSC has pretty much said anything Trump does is legal (he's immune from prosecution), and if the administration is breaking federal laws there is not going to be an independent Justice department to prosecute those people. If Trump decides to ignore the Posse Comitatus Act and call out the troops against civilians, he'll be breaking the law, and the courts might agree, but so what? It will be up to Congress to impeach him, and we've seen that the Senate will not do that. According to Mitt Romney, many GOP senators voted against because they specifically feared for their safety or that of their families. So to answer the original poster's question, I don't see that there really are good guardrails.
If Trump arbitrarily and illegally sends a "you're fired" notice to Alice the Democratic Civil Servant, and writes a letter telling Bob the Trumpist Stooge that he's got a nice cushy job as a civil servant, I'm pretty sure Alice still gets her paycheck every week and Bob doesn't. Because that's what the law says, because that's what the payroll computer is programmed to do, and because even if the civil service job that Bob was told to take up is the "send out paychecks to all the civil servants" job, Alice isn't going to tell him the password to the computer.
I suppose with enough effort, Trump and an army of Bobs could bollix things to the point where *nobody* gets paid.
And the bit about how there's nobody to enforce the law against the Trump administration in this scenario, obscures the key point: there's nobody to enforce the law against *anybody* in this scenario, because you're postulating Trump destroying the administrative state. But the only power the Presidency has, is to enforce the law through the machinery of the administrative state. So, yes, if Trump really works at it, Trump might be able to make the office he went through so much trouble to win, into something impotent.
If the idea is that Donald Trump is then going to build a whole new administrative state from scratch, by sending out legally invalid letters to a bunch of cronies saying "Congratulations on your new post in the Trumpian administrative state!", then that's not going to work. If it *could* work, then Trump could just do that now, the legal power of the Presidency isn't at work in this hypothetical, and if that were possible I'm pretty sure someone more competent would have done it long ago. Nixon, perhaps.
Would reclassifying large numbers of civil servants be illegal? According to Wikipedia (https://en.wikipedia.org/wiki/Schedule_F_appointment)
"The legal basis for the Schedule F appointment was a section of the Civil Service Reform Act of 1978 (5 U.S.C. § 7511(b)(2)), which exempts from civil service protections federal employees "whose position has been determined to be of a confidential, policy-determining, policy-making or policy-advocating character". I encourage you to read the article.
As for there being no one to enforce the law against the executive branch, I think it's a big stretch from saying "the Justice department declined to bring charges against the chief of staff" to postulating the administrative state has been destroyed. All you need is an Attorney General who is a bit pliable. Or not even that, just a President who assures everyone he will pardon them of any crimes they commit on his part. The administrative state will muddle on just fine in those scenarios.
I think deporting millions of people is within his power, given that said individuals are breaking the law, and the president is in charge of the branch of government responsible for enforcing the law, AND ALSO he is being voted in in large part for the purpose of enforcing exactly this set of laws. As the Supreme Court previously noted, elections have consequences.
While I'm in favor of open borders, I do not necessarily see this as a problem with his administration. The problem there doesn't lay with Trump being elected. Maximally uncharitably, it belongs to the people who are voting him into office for that purpose - but that is maximally uncharitable, and I do see their point. I don't agree the problem is with the immigrants, but rather with the network of regulations and incentives we have spent more than a century building up, and if I were to blame anybody, it would be FDR, although even there we've had nearly a century to fix things so even then the blame is pretty attenuated.
But the long and the short of it is - he definitely has the power to mass-deport people, and while local areas being uncooperative can make this less effective, "less effective" is about the most they can achieve. I don't expect local governments to be particularly uncooperative, however; it only takes a tiny fraction of the immigrants arriving in the border states to turn local governments, whose policies are not designed with illegal immigration in mind, against immigrants. (See the cities declaring emergencies after receiving a few thousand immigrants because their policies turn these people into massive budgetary problems for the cities.)
That's a key thing, here: The opposition to deporting people tends to evaporate, or at least weaken considerably, once illegal immigrants actually start showing up. Attitudes towards illegal immigration depend, at a statistical level, entirely on the number of local illegal immigrants, and Abbot's innovation of sending people where they want to go has been absurdly effective, even as it has been a tiny fraction of the illegal immigrants showing up in, for example, Texas.
Are there better policies we should be pursuing? Definitely. My personal opinion is that we desperately needed to boost immigration a decade ago, and that the mass-retirement of the boomers is a major contributor to the inflation we've dealt with for the past four years. Best time to plant a tree is yesterday, second best time is today. But that's its own topic, and "deporting illegal immigrants" is something we can actually do, whereas reorganizing society to integrate immigrants and align incentives is, alas, something I -don't- think we can do.
Given how fucked society is in general, and how unlikely it is that Democrats or Republicans will be willing to fix the underlying issues (it gores sacred oxen all over the place), treating the symptoms and deporting millions of people might strictly be the best we can hope for. Which is fucking pathetic, but, well, that's where I think we are, and I think a lot of people are aware of that.
So, all of that is to say - I don't think the Democrats can or even are willing to seriously do anything to stop mass deportation. See Martha's Vineyard. Just takes one person upset that an illegal immigrant got public housing assistance before they did to get the attention of the leviathan.
(Now, whether or not Trump himself is actually willing to deport millions of people is another question entirely; I think he'll make a token effort, enough to satisfy his base that Something Is Being Done, and little more on that front; I expect most of his effort will be devoted to preventing more immigrants from getting in, and deporting anybody foolish or unlucky enough to get caught up in the justice system.)
These aren't impediments, granted, just a discussion of the topic before we get into impediments.
Major impediments to Trump's agenda:
Federal unions will fight reductions in their workforce tooth and nail, and likely tie everything up in lawsuits that will take longer than four years to fully resolve.
Regular federal employees, well-versed in the art of malicious compliance, will engage in full force to fight anything they don't want to do.
Federal judges repeatedly stymied Trump during his first term, and will likely do so again.
The current Supreme Court has shown a strong interest in dismantling the powers Trump needs to enact most plausible agendas.
Trump will likely be looking to secure a better legacy, which means he'll likely prioritize popular causes over unpopular causes; additionally, to the extent he's genuinely personally interested in causes he wants furthered, he needs to make Vance a plausible candidate in the next election to continue those causes, which means he can't piss people off too much.
Congress is unlikely to play ball; they will want credit for the popular things, and want to distance themselves from the unpopular things. They're interested in maximizing their reeleection chances, not in putting feathers in Trump's hat at their own expense.
The military isn't going to go all-in for Trump. They may support him, but this support will extend only as far as the self interest of the individual parties (including their political/moral beliefs).
Trump is in many respects his own worst enemy; he'll stymie himself.
I agree with your edit and I'm glad you're asking the question. I admire so much about the way ACX subscribers think, but Trump has seemed to be a big blind spot for the normal fact-finding abilities of the rationalists here. The question here is very clear and almost no one is engaging with it.
But it's not one of the things she's explicitly said she will do, whereas the Trump admin has specifically stated a desire to do many of those things. There has to be some impetus to begin with.
It would be more like asking "what barriers are there in place to keep Kamala Harris from increasing child tax credits, the thing she has said she wants to do"... not "some random other action she has never said she wants to do".
Specifically, re: the illegal immigrants part, Trump's "chief immigration advisor, Santa Monica-born Stephen Miller, has spelled out what that would mean: Trump would assemble “a giant force” including National Guard troops to seize undocumented migrants, transport them to camps in Texas and expel them." - and if you want to go down the rest of the list and look up every claim, there's plenty to back it. You'll find word for word quotes or video interviews.
Sure, we can say "politicians say lots of things they don't follow through with" but I don't think it's sensible to respond to "will the politician do the thing they say they are going to do" with "why on earth would you even think that?"
I'm not sure what sources you would find most valid but if you look up "trump glen beck lock up opponents" you will find the interview where Beck says: “Do you regret not locking [Clinton] up? And if you’re president again, will you lock people up?” and Trump said: “The answer is you have no choice, because they’re doing it to us.” and he has also stated “I will appoint a real special prosecutor to go after the most corrupt president in the history of the United States of America, Joe Biden, and the entire Biden crime family,”
In all honesty, I think like all politicians he says things he has no way of setting into motion. As to whether these are credible statements or not - I likely agree with you in that I think they are not. However, I think it's reasonable to listen to that and for OP to ask "so what's stopping him from doing those things he says he will do?"
I mean, if Kamala Harris said "we're going to open all the borders to everyone", even though to me that is a supremely unlikely and "incredible" statement, would you really be saying "well, I don't see why we should talk about whether or not that's possible, there's no reason to think it's credible"? It's not outrageous to check whether politicians *can* do the things they say they are going to do even before they start to do them - regardless of if it's in the spirit of agreeing or disagreeing with the action.
Maybe it's because the "Grey Tribe" now seems to be (vaguely and superficially) right-coded?
Well, Curtis Yarvin seems to think that the bureaucracy will stop Trump/Musk from doing much of anything. And keep in mind that Yarvin would probably *like* radical change, so I don't think he's saying this as some form of wishful thinking. You can read Yarvin's full thoughts here: https://graymirror.substack.com/p/elon-versus-the-usg
You don't need to deport them, you need to get them to deport themselves. You do this by making the penalty for being caught in the country illegally a lot worse than the threat of being deported.
My question is about practical obstacles to doing this. For instance, what penalty could be used? If penalty is a fine, what is the penalty for not paying the fine? Seems unlikely that most will depart pronto, especially if penalty is a fine. Seems like not paying the fine won't make their situation any worse than it is already, and meanwhile they get to keep going to their under the table jobs, seeing local friends and family etc. Prison? For millions? Some kind of semi-confined housing? For millions? Also, problem of many businesses really relying on the labor of these people. Businesses would protest, push back and prob would not cooperate, plus if they do lose the workers various things managed by these businesses are going to be in short supply or more expensive, ripple effect etc. I'm not saying it's impossible, just that it seems pretty hard to figure out how to pull this off.
As I understand it, the primary thing preventing this is what the right would call 'The Deeps State', ie career bureaucrats who care about the country and their role in it and will continue to do their jobs properly and fairly no matter what orders they get from on high.
Obviously this is a biased source, but the book 'Toddler in Chief' collects literal hundreds of anecdotes and statements from people who worked directly under Trump during his first term, talking about how they would routinely distract and deceive him in order to ignore the insane things he wanted to do and do normal sensible things instead. Despite the book being on an obvious side, the quotes do not seem to be generally taken out of context or misrepresented, and many top officials form the first administration have gone on the record in longer interviews describing this state of affairs.
The main reason to be worried that a second term would be unlike the first is that the Trump organization and the increasingly-MAGAfied GOP has widely and loudly denounced this state of affairs, and vowed to have massive purges and loyalty tests to make sure that anyone in a new administration would do exactly what he tells them to every time. JD Vance was chosen largely on the strength of statements he's made to this effect, and Project 2025 is seen as basically a portfolio attachment to the resume of prominent right-wing actors who are pledging to do this and want a place in the new administration.
I think it's pretty credible that a second term would be very different and much worse/much more volatile because of this change. But obviously I also am taking a side here.
My understanding of the American political system is that presidents are quite heavily limited. So if you are talking about the LEGAL system of government Trump can't do that much to undermine democracy alone.
This limit means he needs either a) extensive party support, and republican house and senate) to legally enact change or b) HUGE amounts of popular support + coordination to execute some kind of coup. My understanding is that Trump has neither great dictatoral party support, or popular support, and he certainly seems to lack military support.
That all applies to him being a dictator. If you are scared of him deporting ILLEGAL immigrants. I don't know what to tell you. In a democracy, people can vote for policies you don't like - and the democratic thing is to accept that. You should make your case, argue, accept the winning position, and then if you lost, do a better campaign next time.
Who do you actually want to be president?
- You get to magic them into the white house
- They have to be an eligible and living person
- No picking Trump or Harris
- The magic only gets them there, it doesn't help them govern. (So otherwise impressive folk with no political experience should have some reason they won't pull a Herbert Hoover)
I'm considering casting my vote for a write-in candidate on the theory that one should try to vote for good things.
It would seem remiss to waste my magical powers on anyone except myself. (I am not actually eligible but I'm just gonna go ahead and magic myself up a Hawaiian birth certificate while I'm at it.)
Obviously, if we can magic solutions up, the ideal politics is:
"Clone Lee Kuan Yew and install him as dictator-for-life in all ~200 countries."
LKY took a country that started poorer than the Philippines, and in ~30-40 years of autocratic leadership, brought it to fully developed, First World status. And he did it with a lot of ethnic diversity, having zero natural resources, and with incipient civil war and external takeover looming most of that time. It's the least corrupt country in Asia, and the most beautiful, least polluted, and well run.
He's the only provably successful "enlightened tyrant" that we can both verify his accomplishments and actually get DNA for - so LKY for World Presidents!
I think it would be relatively hard to generalize the governance of a large city, to governance of a large country.
In Singapore you can plausibly have your hands on every lever of regulation and power, and a single good strategy can bring significant wealth to the whole city. The same is not true for the United States.
> having zero natural resources
I think in this context we have to consider Singapore's location as a pretty valuable natural resource, although his accomplishments are still pretty impressive.
Came here to say exactly this
Singapore reminded me strongly of the United States of the 1990s, including some of the social fissures that would later break apart here. I'd hesitate to say we should definitely replicate it until we see what happens in thirty years.
But arguably, Singapore has had major social divisions (Malay working and under class, Chinese middle and upper, Westerner upper) since the very beginning, and LKY was better than most at keeping everything aligned and progressing even with those divisions. And they were pretty serious, leading to riots and tense situations fairly frequently in his tenure.
Unless you're speaking of other social fissures? In which case, I'm interested in hearing about them.
For an example, social welfare as relates specifically to disabled people; the television programs while I was there were -extremely- careful to simultaneously insist that the way disabled people were treated wasn't okay (seriously, television there felt right out of the 90s in the US, complete with a Very Special Episode about vaping on a daily sitcom), and to showcase ways in which disabled people were being integrated into society - while also going to great lengths to make clear that No Government Money was going towards supporting disabled people.
You don't need to make it clear government money isn't being spent on an issue unless there are people agitating for it to be spent in this way. It seems there is something deeper going on there.
There were other little things. Another show, this one about prison inmates, implied, while very carefully not saying anything to this effect, that the drug punishment laws are too harsh. And a taxi driver claimed all the trees had GPS units installed in them so they could catch people who stole trees (???). A handwritten sign above a potted plant saying it was on camera and anybody who littered in the plant's pot would be prosecuted.
Little things, which in a society where I watched somebody leave an unlocked, signed-in Apple laptop sitting on a table on the side of the street (unthinkable!), suggested that social trust was beginning to fray.
On the whole, mind, I found it quite a nice little society; I'd recommend anybody visit just to see what a well (if, in my opinion, over) governed society looks like. But I'd hold off on assuming that it won't experience the same problems showing up elsewhere in the world, because it feels less like it has found a way to avoid them, so much as that it feels like they're still simmering under the surface.
Ah, gotcha, thanks for elucidating.
I see less portents in those than you do, I think - in terms of strictness and panopticon-against-crime, it's pretty much always been that way.
This is why they sell those (hilarious, IMO) shirts in Bugis:
https://www.etsy.com/listing/1564124012/singapore-a-fine-city-t-shirt-vintage
Arguably, the whole "panopticon" thing is WHY Singapore is so clean and well run and non-corrupt, especially compared to all the rest of SE Asia. I think it's actually a pretty good exhibit on the level of strictness and authoritarianism you need to have a clean and orderly society with a lot of ethnic diversity *and* a very high Gini index. Most countries so clean and crime free (Japan, Scandinavian countries) get by with a lot of ethnic homogeneity and low inequality, but Singapore is proof you CAN have it, it just takes a lot of effort and a panopticon society.
I agree on the overgoverned point, btw. But it's a tradeoff too - if they didn't overgovern so much, they'd be a lot more like Ho Chi Minh or Manila.
I think your disabled point and drug severity point is just regular politics, but agree there is a meaningful split in opinion.
But yes - anyone interested in what an incredibly well run (if overpoliced) society looks like, definitely check out Singapore!
I did specify 'eligible and living' but I guess if you start your cloning today this will be a great answer in 35 years.
Of prominent American political figures: probably Jared Polis. There are some people I can imagine doing better*, but they're bigger risks and I think someone with a good track record as a major state governor is safer.
*Sam Altman maybe? He seems unusually good at running effective organizations and also good at political maneuvering, and while he's less trustworthy and less worried about AI risk than I'd like for an AI guy he's still better about both than the average politician.
Troll answer: Along Levy (get us some real cost-effective train systems in place). Or that guy who built the entire Madrid metro in four years. Or Deng Xiaoping (use a time machine to get him during his exile period).
I agree w Alon Levy!
The capacity of the Spanish State to deliver infrastructure is under estimated and goes against a Northern European bias I used to have. I feel it’s better than modern Germany and far better than the U.K.
It hasn’t yet helped with economic growth, although these things matter in the long run.
Yes. Alon Levy has also noted that while their ability to build infrastructure is phenomenal, their actual operations (and maybe integration?) are weaker, leading to the Spanish AVE never really getting the kind of ridership the french TGV gets despite the more extensive network.
Sam Altman might be my strongest *anti*-pick—he is already extremely dangerous with the amount of power he has. He would be even more dangerous as president. Sam Altman has so many terrible qualities:
- is trying to build an AI that might kill everyone
- repeatedly lied to his board and then fired them when they tried to fire him
- repeatedly lied about how OpenAI's structure was going to keep it in check (first the independent board who got fired and replaced, then the Superalignment team that got disbanded, then the non-profit parent that got disbanded even though that was supposed to be legally impossible)
I agree that he's unusually good at running organizations effectively, but he runs them toward bad ends. That makes him an exceptionally *bad* candidate, not an exceptionally good one.
So my point is that level of cynical power grabbing politics is just the baseline I expect from elite level politicians (and if we picked a president who didn't have that he'd get played by politicians who do anyway), so it's not a minus there relative to the baseline.
(Plus, it would have the benefit of distracting him from OpenAI)
While we're at it, Paul Graham. He seems to be a competent manager that is more likely to tackle actual issues than be caught up in identity politics.
Saying Paul graham, the guy who blocked all Jews on his Twitter, is "unlikely to be caught up in identity politics" is quite a take.
I wasn't aware. Will have to update on that.
I've been hearing all kinds of "Elon is a Nazi" talk on Twitter, and not taken it seriously, but if he's installed a "block all Jews" button that you can click...
:-P
Iiuc there's a lot of custom mass block lists going around you can use for twitter (e.g. if you want to block all democrats/republicans/musicians/whatever). They're third party tools, not muskware.
Can you point to some evidence for that claim?
> Paul graham, the guy who blocked all Jews on his Twitter
What are you basing this on? Googling gets me one guy saying 'PG has blocked me and every Jew I know' and implying that Graham is blocking people solely on the basis that they're Jewish. But that guy's feed seems to be ~100% pro-Israel tweets, and it seems vastly more likely Graham is blocking people based on their stance on the war than based on their race or religion.
NSA Jake Sullivan
Great answer. Anything in particular put him over the top for you?
Un-polemic, willing to acknowledge and attempt to fix the other side's problems. Supports the correct foreign policy, understanding the immense role of US in world order, but still cautious and dovish, more so than neoliberals. Extremely smart, hard worker, MPhil international relations and JD (law), important for domestic and international aspects of Presidency. Already very experienced in the gritty details of the top echelons of government. A true political technocrat.
A profile for more reading if people are interested: https://www.newyorker.com/magazine/2023/10/16/trial-by-combat
Dean Phillips. The only credible challenger to Joe Biden in the Democratic primary. That shows real strength of character.
Our gracious host, of course, and I'm only 25% joking. Scott would hate having to be President, which would make him even better at the job.
If I had prep time to research someone, I'd go with someone else. If I had to answer in like 30 seconds, this is likely to be my answer.
Former speaker of the Texas statehouse Joe Straus, or former Texas Parks and Wildlife director Carter Smith (the latter an appointed position, not an elected one, but it is a political job in that one must go begging the Legislature all the time, and stopping the bubba libertarians trying to "sell all the parks" etc.).
Idiosyncratic and obscure but I have my reasons: chiefly because they represent a vanished world where normal people enter state politics and the state party machinery then elevates those people to a higher level, which machinery appears to have completely broken down in many states so that normal, intelligent people with some ambition need not apply.
On the premise that they need a decent amount of experience in politics/government and I'm just looking for competent, well-intentioned leadership, maybe Ron Wyden? Although he's on the older side as well.
Anthony Blinken? He seems like a very bright, very thoughtful, quite chill guy who knows a lot about a lot, prioritizes his family, and has a sense of how both the US government and the world work and don't work.
A key thing in his favor is that in four years as Secretary of State he has not been faulted much by either side, which seems like magic in itself. Since I know nothing about his thoughts on domestic policy, I get to infer them.
Been a while since we had a secretary of state become president. Google says the last was Buchanan.
Seems like such a great transition (though Governor and Joint Cheif also make sense).
I guess Hillary tried that route.
Wonder why it died out.
It seems like you're asking because you want to actually vote for someone who isn't Trump or Harris, and in that case I'd suggest that voting for one of the other candidates on the ballot is a better idea than voting for a write-in candidate. Given that the chance of the vote actually making a difference in the election is a rounding error either way, the expectation is that at least you might hear "x,000,000 people voted for the Libertarian Party nominees" in the news, but you probably won't hear "x0 people voted for Scott Alexander". Given that, Chase Oliver would be my top choice, followed by Jill Stein.
I'm aware that this is (close to a form of) strategic voting. I dislike that but it's a necessary evil in any electoral system, but especially the US system.
I often vote third party, partially to boost the legitimacy of third parties and signal dissatisfaction with the existing structure.
This time, none of them seem up to snuff.
I actually might vote for one of the better answers here (or Randy George, the Chief of Staff of the Army).
But regardless, its enjoyable and informative to think of what a better class of candidate might look like.
If I don't have time to go do a ton of research then off the top of my head I would say Penn Jillette. Unlike any career politician I know of, I trust him to be intellectually humble and seek out true expert counsel, and to be skeptical and careful about what to believe from his advisors. I also trust his ultimate values and goals as much as any other public figure I can immediately think of.
Brian Fitzpatrick seems to be the leftmost GOP representative ( https://progressivepunch.org/scores.htm?house=house ) and
Henry Cuellar seems to be the rightmost Dem representative
(unfortunately even they seem to be distressingly far apart)
One of them seems like they would be a reasonably sane centrist choice.
Along that vein, Kyrsten Sinema has fairly consistently ranked at or near “most moderate” in both the House and Senate. She went independent after falling out with the Dem caucus and is not currently running for re-election.
Pair her with Romney as an elder statesman and you’ve got a centrist ticket that has declared they are done with this shit in both directions. Has a certain charm.
Many Thanks! Yes, I'd be quite happy with that centrist ticket.
I considered Romney! Yglesias occasionally brings him up in interesting lights. Would that he could have saddled his horse in '16 instead of '12.
If Romney had run in 2016, he'd have just further split the pool of primary votes than went to Bush III, Rubio, Cruz, Kaisich, and I think I'm forgetting someone but not matter - that split gives Trump the win with or without Romney in play.
I'd say go with the zeitgeist and pick a sequel of some kind.
Alexander Hamilton has living descendants. Jimmy Carter could theoretically serve a mythical second term. There are yet more Kennedys you could pick, even if you don't want RFK Jr.
Since every movie and book and TV show is a sequel, why not bring it to the Presidency?
<mildSnark>
Speaking of sequels... Would Kamala Harris's proposal to "regulate" speech count as a sequel to https://en.wikipedia.org/wiki/Alien_and_Sedition_Acts#Sedition_Act ?
</mildSnark>
If it weren't for the apocalyptic AI threat looming over all of politics: Probably Robin Hanson, despite his rather eccentric views on many things. Having proper incentives makes a lot of things easier, and he's really good at figuring those things out. Markets are smart.
(Given that AI issues do actually far outweigh all other issues: Anyone willing to dump every scrap of their political capital into halting AI globally and indefinitely, plus has sufficient skills in politics/communications/PR/diplomacy/etc so that they'd have the necessary influence in the first place. But that's less fun to go into.)
If I was gonna do Hanson, I would probably just do Scott. But great public intellectual <> great president.
Greg Abbott
In the process of reading the Iliad right now. It’s amazing but I can’t help feeling there’s so much here I‘m missing that a more trained eye could show me. Barring just taking a class on it, anyone have recommendations for reading the Iliad and other classics and getting the most out of them?
I'd personally start with Dan Simmons' Ilium (and maybe the sequel), which is a very readable and high fidelity take on the Iliad embedded in another sci-fi story. After you have that high level view of the various motivations and characters as translated readably from a modern context storyteller, then go back to the original source, ideally with some ancillary texts others will recommend.
Uhm, this is a great book that I admire a lot. But I doubt very much that it helps with reading the original Iliad.
I would rather recommend the opposite: *After* reading the Iliad, consider reading Dan Simmon's Ilium, because that is a great Sci-Fi novel whose humor you can only appreciate if you know the Iliad well. Basically it's a fan-fiction to the Iliad, and just like HPMoR it loses a lot if you don't know the original story.
I actually did it this way though - tried to read the original Iliad, read through it but largely bounced off of it mentally, and then later read Ilium.
Ilium contextualized and grounded everything in a way that the original Iliad doesn't do at all, the Iliad was written from a really different culture and mindset, and depending on the translation, the archaic language can be a hangup too.
When I reread it after reading Ilium, it made a lot more sense to me. But that could definitely just be me, maybe it's bad advice generally.
Interesting. In the end, I don't know this way around, so apparently this works, too. :-)
The Emily Wilson translation published last year is snappy, expressed in iambic pentameter, and uses modern language, if the Gutenberg versions are too crusty.
Less high-fidelity I’m sure, but “Song of Achilles,” the story of the Trojan War from the perspective of Patroclus, is one of my favorite books of fiction. I thought it was fun, beautiful, touching, and meaningful.
Have also read that for school. Song of Achilles is great but the author’s foot fetish comes through wayyyyy too clearly.
I second this. SoA was fun and helped me better understand the timeline of the Iliad.
Luckily, much has been written on this. You could have a look at the recent 'Homer and his Iliad' by Robin Lane Fox.
I think the fastest way is probably to listen to a course that goes chapter by chapter.
I'm very frustrated (though I suppose it's only fitting) how much knowledge of epic literature is trapped away either in oral form or old dusty books. Not because that makes it off limits to me but because I feel like there is a real appetite for deep reading of books that reward that kind of thing.
Depends on what exactly you want to get. If it is about the story, then I can warmly recommend the retelling in "Gods and Heroes of Ancient Greece" by Gustav Schwab. It doesn't change the content, but in terms of language it meets halfway between the original and modern language. So you still get a flavor of the archaic language, but it is much more readable. It covers the Iliad along with many other ancient tales.
But of course, you will not get to see the original verses this way.
Joel Christensen's substack (all about the Iliad) is surpassingly good. Would link but linking in substack is as difficult as everything else about substack.
Find a translation that clicks for you, prose or poetry, and try to imagine what is happening in all its gravity. I know that this sounds trivial, "like duh, that's what reading is," but Homer narrates action and dialogue, not what happens in the characters' heads. You have to infer a lot about what the characters think and feel from by synthesizing what they say and do with your knowledge of their arc.
The scene in Book 6 with Hector, Andromache, and Astyanax on the wall moved me to near-tears when I looked through the characters' eyes instead of focusing on the description of the event. It's not suspenseful - you already know how the story ends - but try to feel what they must be feeling in that moment and you sense the crushing weight of it all: choosing between safety and duty; family and city; and the burden of leadership held by the shoulders of heroes who are more than mere men.
There’s free online courses on YouTube, often by world renowned professors teacher at Yale, Harvard or another impressive university.
Besides that I’d recommend to read through many of the Greek Classics. Iliad, Odyssey, 3 Tragedians, Aristophanes, then circle back to the Iliad with a different (maybe more difficult) translation. Maybe Herodotus, Thucydides, Xenophon’s Hellenica too.
More context on the Greek culture will give you a much stronger basis to understand the Iliad. Much information is assumed, many stories are told in tragedy that would have been basically common knowledge for anyone listening to the Iliad. It’s impossible to get full context, but I think it’s better to get it from other Classical Greeks than even the best written introduction of best taught course.
Your milage might vary - I find the The Great Courses match my desire for infotainment quite nicely when I'm curious about a particular academic topic but coming in pretty cold (for me at least, it's almost always about the right level of introduction).
I see that they have a course on the Iliad https://www.thegreatcourses.com/courses/iliad-of-homer. I have not listened to that particular course.
I have an annual plan which is less than $10 a month https://www.thegreatcoursesplus.com/customer/plans
(They have free trials which is plenty of time to knock out that course (looks like its 6 hours))
Also 1 credit on Audible https://www.amazon.com/The-Iliad-of-Homer-audiobook/dp/B00DTNW7IS/ref=sr_1_1?crid=F1ZJ7BN4L64M&dib=eyJ2IjoiMSJ9.cDZy63ec6TbMFy8Lf98--ZPWBBgs2bOZC64Gkpri2vYSziKLKFC_nJsPU87WEKKMuV1v1wer2H75dvKWBH_8CmAjZLYE9LdIJK9oaBVUVbROew_qP80rBpFVhiH-Prcl0eGJvVqYvwtF_yDAkkOuDznsIMV6N_ijGzl6K51Wn7c.j04dwVFnZuTAOq8-HiBGoKtQA4lIy1mCXDOd9QpaMoA&dib_tag=se&keywords=iliad+of+homer+great+courses&qid=1730133543&sprefix=iliad+of+homer+great+courses%2Caps%2C197&sr=8-1
I typically just listen to them when I'm in the car by myself, if i just listen while I'm going about my life, I find that I knock a lot of them out pretty easily.
Again, you might be looking for something different.
Try HomericEpicPodcast.substack.com, and The Homeric Epic Podcast. He's up to Book 9 and it's a banger for history nerds and use-of-language nerds and lots more.
Emily Wilson's translation of the Iliad has an excellent preface and translator's note that really helps bring out some details and themes.
This is what I used when reading the Iliad to my 4yo and he loved it.
Occasionally heavy handed (to the point of being incorrect) but overall excellent and probably as close as a translation gets to 'reading with an expert pointing out the good stuff'.
Unfortunately, Wilson's Odyssey is much worse than her Iliad.
Joen Christensen also has a series on notable passages you may enjoy:
https://joelchristensen.substack.com/p/managing-achilles
I’m using the Wilson translation actually. I know it’s taken a fair amount of flak on here but I think it’s actually a really good translation that makes the story quite accessible to modern readers, and I think most of the sections people point out as examples of why it’s a “woke” translation or whatever are just thin points in the translation that aren’t representative of the effort as a whole. I also agree; her notes and introduction are excellent. Sad to hear her translation of the Odyssey isn’t up to par.
She started a substack by the way which is an enjoyable look at the inner workings of translating.
Her Odyssey is ...fine. But the notes and introduction are less extensive and illuminating, the meter starts off stronger than the Iliad's average but lacks drive for long portions, and overall it just seemed flatter to me.
Which translation of the Odyssey would you recommend instead? I do strongly prefer metric verse translations.
Book recommendation: "The Ancient Greek Hero in 24 Hours"
Scott, please consider doing a “More than you wanted to know” post about the Jones Act.
Repeal of the Jones act is the most important issue of our lifetime and it’s not getting enough coverage.
Edit: See this Planet Money episode transcript as a summary of the issues:
https://www.npr.org/transcripts/554046425
Never heard of it. It appears to regulate some shipping in the US - why is it the most important issue of your lifetime?
I can’t say anything about its importance, but some implications are sketched in https://thezvi.wordpress.com/2022/05/05/repeal-the-foreign-dredge-act-of-1906/ (the main post is about something else, but there’s a section on the Jones Act).
It makes an enormous amount of things in the US more expensive than they should be in an effort to prop up various interest groups, like ship builders, that mostly no longer even exist.
It's like the debt ceiling, a pure deadweight loss law that no other country does. It's not really that big of a deal as the US isn't short on money.
I think Trump can easily carry Hawaii by pledging to repeal the Jones act or at least force Harris to the same pledge but he's not that sophisticated.
Maybe if he committed to that a few months ago. The existing ferry company would then donate to whoever supported the status quo. There isn't really much money in it for Trump for him want to do that.
Maybe it would work better for a Democrat appealing to Alaska over the issue, although I'm not sure how big of a deal the Jones Act is there.
Oh the other hand, maybe the winning strategy is to commit to repealing the act *right before* the election is held. With the right timing, the ship building companies wouldn't have time to significantly contribute to the opposing side, but Hawaiians would change their vote.
I would have expected the Jones Act to have a relatively smaller impact on Alaska because of the proximity of Canada.
Hell, he would even win Puerto Rico if he did that.
Don't be silly, Puerto Ricans are not allowed to vote!
I suggest this Planet Money episode as a starter (transcript below):
https://www.npr.org/transcripts/554046425
I would like to see you soon on the front lines in the fight against the evil yoke of the Jones act. Let's free the American spirit from its shackles. Thank you.
It sounds like an antiquated piece of legislation that's profiting some groups (US owned shipping companies) at the expense of others (Puerto Ricans and Hawaiians). Maybe it should be repealed.
I still don't see why it's particularly important. If it's repealed what's the net impact? A few million people save a few hundred dollars per annum? This doesn't outweigh the urgency of the climate emergency for me, for example.
It's a law that stops a certain market operating efficiently - why do you think it's evil? And the American spirit, whatever that is, doesn't really seem to be enslaved by shipping law. What's the moral basis for your argument here?
Well I may have been tad exaggerating to rally support for the issue.
But I guess the potency of repealing the Jones Act for me is that it is unambiguously bad. So everyone should be able to rally around its repeal. We should be able to come together as one Comment Section.
I mean perhaps it also relates to the climate emergency as if we repeal the Jones act, cargo ships will be more efficient and hence reduce CO2 emissions.
I am hoping once the Jones Act is repealed, other similarly unjust legislation will be on the chopping block. In that sense, it's the defining issue of our generation for me.
Again I hope to see you on the front lines in the struggle against the Jones Act. Thank you for your interest.
Thanks for explaining - I'm a little more bought-in to your argument now.
Unfortunately I'm neither a US citizen nor based in the US. Despite what Fox News might have you believe NEITHER side has offered to fly me in to vote.
>This doesn't outweigh the urgency of the climate emergency for me, for example.
While the jury's still out on wind power's effectiveness, the issues are not entirely unrelated: https://reason.com/2023/12/05/federal-shipping-regulations-sank-one-of-americas-biggest-offshore-wind-projects/
Hey, I work at a thinktank (Balsa Research) that's currently focused on repealing or significantly amending the Jones Act.
The Jones Act requires all goods shipped between U.S. ports to use vessels that are U.S.-crewed, -owned, and -registered. As of January 2023, the entire Jones Act fleet consists of just 56 tankers and 37 cargo ships (1), mostly serving Mainland <> Hawaii, Alaska, and Puerto Rico routes. Given the U.S.'s limited shipbuilding capacity, these numbers are unlikely to change significantly.
This single law has effectively eliminated domestic coastal shipping in the United States (again, there exists all of 37 domestic cargo ships! for the entire country!). While we do have significant barge traffic (2), especially on inland waterways, most domestic freight that could be moved by modern cargo ships is instead transported by rail or truck - methods that are OOMs more expensive and carbon-intensive.
It causes a bunch of other specific harms that I can get into (sluggish WTIV production that will cause us to fail our offshore wind targets; actual safety concerns for sailors; worsened response capacity in wake of environmental disasters such as hurricanes), but all of that pales in comparison to the fact that, again, it's an act that has rendered domestic coastal shipping by cargo ships effectively illegal in a country with two entire coasts.
Balsa Research will have a more comprehensive analysis of the Jones Act available by year's end - it will be announced on Zvi's blog :>
1. https://www.statista.com/statistics/646259/us-flag-oceangoing-privately-owned-jones-act-fleet-by-type/
2. A form of crappy boat that is technically not a vessel, and therefore not subject to The Jones Act. More discussion: https://digitalcommons.mainelaw.maine.edu/cgi/viewcontent.cgi?article=1422&context=oclj
Would it be reasonable to think that if domestic coastal shipping was rejuvenated, that some rail lines could be repurposed to passenger transport? That would be nice.
Isn't rail shipping more efficient than boat shipping though?
Wikipedia claims they're approximately equal: "On average, a gallon of fuel allows one ton of cargo to be shipped 180–240 mi (290–390 km) by truck (e.g. @ 6–8 mpg‑US (2.6–3.4 km/L) 30 ton load, 450 mi (720 km) by railway, and 514 mi (827 km) by barge. "
I wasn't able to find the source for their numbers, though
Thank you for your extremely important service. I will be trying to follow Balsa Research. I am glad I am not alone in my crusade against the corrupt and vile Jones Act.
Can you link to Zvi’s blog for me?
If there's one policy contribution the libertarians can get us to actually implement, it needs to be this.
100%
Maybe try without the hyperbole?
I think the Jones Act was a stupid idea that is causing a good deal of harm to no good end, but "most important issue of our lifetime" is either completely bonkers, or in need of an extraordinary justification that you haven't provided. And making the unsupported hyperbolic claim, makes me much less likely to take your position seriously or follow your links or whatever.
I am sorry that my enthusiasm for the repeal of the Jones Act has rubbed off on you the wrong way.
Nonetheless, I am glad you think it’s a stupid idea.
I do think it is a very serious issue but I understand if you are not as enthusiastic or animated as I am.
See you on the frontlines against the evil oppression of the Jones Act.
What kind of anthropic reasoning arguments are valid, which aren't, and why? And how sure are you/our best experts(?) about this? Are there good reading/watching materials online on the subject?
And for a more specific and concrete doubt, take the doomsday argument:
1. There have been ~100 billion people.
2. Suppose you are exactly the 100 billionth person to be born. Assuming 1, this is approximately the case.
3. You are way more likely to have been the 100 billionth person if there was only ever going to be 100 billion people in history, then if there was going to be 500 billion people in history.
4. By Bayes, this means it's more likely that you will be the last person to ever be born then that there will be 400 billion (or any number that matter) more people after you.
This *seems like* evidence that the end of Humanity is nigh. I've read (including from Scott) that this is counteracted by the principle that you are more likely to be born at all (in any order) in worlds where there will be more people throughout history -- which indicates that the end is not nigh.
Both the concept that "you are more likely to be chronologically late" and "you are more likely to exist in timelines where there are lots of people like you" seem to make sense, but I also have a vague but strong feeling that there is some sort of double-counting of arguments happening here. I'm unsure if we can use one to counter the other and end up with a uniform distribution, even if the probabilities seem work out. What is happening here?
3 is just flat nonsense. By that logic there's only one person, because think of the odds of being the first of 500 billion versus being the first of one. Therefore the human race will end 65 million years ago.
As stated it was somewhat garbled compared to other times I've heard the Doomsday Argument formulated. The more usual explanation is that if there are N people who ever live, a randomly selected person would more likely to be vaguely in the middle of the timeline than very close to the extreme ends.
The idea is to sample a set of sequentially numbered elements in order to estimate its total size. If you sampled thousands of elements, you could get a pretty good estimate of the size of the set since the mean of the serial numbers will be about half the size of the set. Unfortunately, we only have one independent random sample, so our error bars are going to be huge but that still gives us a point estimate of the size of the set.
So knowing you're the 100 billionth person to live gives a point estimate of 200 billion people living total in history. But the error bars at 95% confidence are stupendous, as you could still be within the middle 95% if 2.5% of humanity lived before you or if 97.5% of humanity lived before you. So the doomsday argument (if valid, and if I'm doing the math right) implies that there's 95% confidence that between 2.5 billion and 4 trillion people will live after you.
> if there are N people who ever live, a randomly selected person would more likely to be vaguely in the middle of the timeline than very close to the extreme ends.
I don't think that's the case. If we assume that the N people are uniformly distributed across time, then a randomly selected person isn't more likely to to be anywhere specific in that period of time, neither middle or end.
If the N people have a distribution that grows with time -- like the actual distribution of humans being born -- then any randomly selected person is more likely to be born towards the end of the time period. Hence the argument I made.
You would only get this "more likely to be vaguely in the middle" effect you describe if the distribution of births per year was something like a bell curve. Which it isn't.
To clarify, "vaguely" is doing almost all of the work. A randomly-chosen person would be 95% likely to be in the middle 95% of the data set, or 50% likely to be in the middle 50% of the data set.
So if your "random" person is the 100 billionth person, that suggests 95% confidence that there are between 102.5 billion and 4 trillion people who will ever live and 50% confidence that there are between 125 billion and 400 billion people who will ever live.
The argument falls down at multiple points. Firstly, everyone will only ever be able to see how many people there are in their past and present, so if you assume that's the median, this would tell every single person that humanity will end at roughly double its current size. From Adam and Eve to Pericles to Herbert Hoover, the equation is the same, and the numbers are always different.
Secondly, it assumes people pop into existence at random, when reproduction is entirely cause and effect. What are the odds you were born when you were? They're the odds of your parents having a child. The odds of two people willing and able to have children. having children, is ridiculously high. To play the median game is to argue that "you" can exist independent of the biological process that made you. Nobody born today could have been born five thousand years ago, because their parents weren't around five thousand years ago.
Thirdly, nothing says humanity isn't an infinite set. What's the median of infinity?
This whole line of reasoning strikes me as an argument of comfort. People aren't comfortable with giant numbers. They don't like thinking their time will be looked back upon with the same polite indifference we have for the Iron Age. So they create justifications for why they're actually much more important than that; civilization is peaking, right now, definitely. This cup in my hand holds 12 ounces, therefore there's only 24 ounces in the ocean.
I'll grant your second argument as being reason to doubt the doomsday argument. Serious consideration needs to be given to what the most valid reference class should be, and I haven't heard much discussion for that. You sound like you're proposing the individual as being the only sound reference class; I don't think I'd go that far, but I will provisionally accept that "all humans" is not necessarily a sound reference class for the doomsday argument.
The first argument is only an argument against taking the point estimate too seriously. A lot of people who talk about the doomsday argument take the point estimate way too seriously, and you are making a good argument against them, but I am not one of those people. Take all people who have existed or will exist and apply the argument to them, and if the correct value is finite, 95% of them should have the correct number within their 95% confidence interval.
It is conceptually possible for humanity to be infinite, yes. I have my doubts as to physical possibility, but I grant the conceptual possibility, and in that case the real value will be at the infinite end of the tail of the confidence distribution. We can't absolutely rule it out, any more than we can rule out the world ending next Tuesday due to false vacuum decay.
Bottom line is, I agree that taking the doomsday argument to conclude that humanity is likely to go extinct in the next century or so is silly. At most, I'd conclude that humanity is rather unlikely (but not vanishingly unlikely) to progress to be a Kardashev 3 civilization before we cease to be recognizably human. I'm inclined, based on triangulating with other priors, to expect that the actual number is likely to be towards the upper end of the confidence interval or perhaps a bit past it.
um where's 65 mya is from, is that sarcasm?
Maybe that was mammals. I don't care enough to find out.
If anyone is curious, anatomically modern humans are thought to have originated anywhere between 60,000 and 160,000 years ago.
No, step 3 is formally correct: P(I am person n | only n people) is indeed higher than P(I am person n given 4n people).
The error is in step 4: Bayes says that
P(k people | I am person n) = P( I am person n | k people) P(k people)/ P(I am person n)
Reearranging this, we get
P(n people | I am person n)/P(4n people | I am person n) = P( I am person n | n people)/P(I am person n | 4n people) * P(n people)/P(4n people)
and so the correct observation that P( I am person n | n people) > /P(I am person n | 4n people) implies that
P(n people | I am person n)/P(4n people | I am person n) > P(n people)/P(4n people)
However, in step 4 the OP wrongly omits the term on the right side, assuming it to be 1, and therefor arrives at the incorrect conclusion that
P(there will be n people) > P(there will be 4n people).
That same line of reasoning could have been used 99,999,999,999 times before, and it would have been wrong every single time. An argument with that kind of track record is entirely useless for anything, except maybe as an example for flawed reasoning.
Edit: I guess the most important takeaway is that "something something probabilities something Bayes therefore conclusion" will lead you astray. Bayes' rule is a mathematical theorem, and unless applied with mathematical rigor, it won't be more reliable than "my gut says X".
Thanks for putting into words something that's always in the back of my mind when reading these kind of arguments.
Indeed, the Doomsday argument (without taking into account the second argument, which is the Self Indication Assumption) would indicate that "the fewer people there are, the more likely". Maybe this can be salvaged if we consider that, yeah, it's a big Universe, most time groups of intelligent living beings appeared, they were few in number, and then died out. We just happen to be living in one of the few big groups.
But my point here isn't to defend any of the arguments I described in my original comment. My response to you is: ok, if that was wrong, where is the mistake in reasoning? The fact that this reasoning would have been wrong 99,999,999,999 times before doesn't tell us where the mistake is.
One possible answer I lean towards is "this argument is actually correct, but there are so many *other* factors coming from empirical observation that this argument's contribution to the probability distribution gets drown out. Is this something you believe?
Your example is wrong because I, a person reading your comment, am not in any sense being sampled from a random distribution of all people. Indeed, to derive 2 from 1, you assume this not to be the case. But then at step 3 you assume this is the case.
I would say most examples of anthropic reasoning one sees online are wrong, or at least under supported, because they nearly always handwave over the part where all the possibilities are equally likely.
I don't see how point 3 assumes anything aside from what is stated in points 1 and 2. Can you elaborate?
And about random sampling: well, yes, my comment is selecting for the people who are reading comment in this thread, not uniformly among everyone in history. But, in principle, anyone inn history could have just come up with the argument by themselves, so the argument -- not the post -- is still randomly selecting from Humanity..? Maybe? Maybe stone age humans wouldn't have had the culture necessary to come up with this argument? Does that change anything? What about neanderthals, do they figure in? So many questions.
You say (1) there have been about 100 billion people. I haven't checked, but I'm prepared to believe you.
Then you say (2) that I'm around the 100 billionth person to have existed, which assumes that, I, the reader exist about now and I'm not reading your comment from a colony in another galaxy a million years in the future.
Then you say (3) that I'm more likely to be the 100 billionth person if there were only a 100 billion people than if there were 500 billion people, which assumes that I, the reader, am equally likely to be each person who will ever exist.
It seems to me that some sort of dualism is underlying the argument, as if consciousness freely floated over matter, so that your consciousness could equally well have attached to any other human. But this is quite wrong: you could only ever have been Michael A. Katt. You have to actually perform a random sample, not just conceptualise it.
Damn, good point. I find that I can switch between a mode of thinking where your position is correct, where "of course I could only ever have been me", and a mode of thinking where you're wrong, where I truly could *hypothetically* have been anyone else. But these are both very informal, gut-feely modes of thinking, we should formalize what we're saying.
Classical probability theory, and the way that we often talk about probability in real life is frequentist. If I throw the dice 6,000 times I expect to see approximately 1,000 occurrences of each face, and as the number of throws goes to infinity, each face occurs 1/6 of the time. That's approximately what we mean when we say that the probability of any particular outcome in a throw of a dice is 1/6. What's key is that you've enumerated all possible outcomes and assigned a probability to each of them. This is black-and-white undergrad maths stuff.
We also say things like "Trump has a 55% chance of winning the election." You can't run the election infinitely many times, so this is a different type of statement. Essentially you're quantifying your uncertainty about an event - let's not get into the philosophy here. It's implicit here that Harris has a 45% chance of winning, and you've again got a satisfactory probability distribution.
What's the distribution in your question above? The space of all people who have ever existed, uniformly? Clearly not. What you're doing is basically reading the nearest car number plate on the street outside, saying 'There are 100 million cars in the US, what are the chances of me reading this one?' and assuming you've witnessed lottery-winning levels of unlikelihood.
Actually, if you pay attention, the way people use probability in practice is often incoherent from a mathematical perspective. And saying 'Bayesian' doesn't solve all the problems.
My goal when I explained the doomsday argument in with my comment wasn't to mean "and I believe this makes sense and is true", it was more to give an explicit example people could dismantle and point to the faulty part, is there is one. In this case, I understand you to to be pointing to this as the faulty part:
> What's the distribution in your question above? The space of all people who have ever existed, uniformly? Clearly not.
This isn't what the argument is doing. The argument assumes a priori that the distribution of people from the past up to the present is, well, the one we think happened (https://en.wikipedia.org/wiki/Estimates_of_historical_world_population#Deep_prehistory).
The distribution towards the future isn't assumed by the argument, it is actually the thing that the argument is trying to determine, based on both the previous distribution and the observation that the person making the argument was born at the time they were born.
> What you're doing is basically reading the nearest car number plate on the street outside, saying 'There are 100 million cars in the US, what are the chances of me reading this one?' and assuming you've witnessed lottery-winning levels of unlikelihood.
We are witnessing a 10^-8 probability event every time we look at a unique license plate from a pool of 100 million. Yes, this is a very very low probability. But it's not a surprising event, since every counterfactual event -- looking at any other license plate -- would have had the same low probability. An event isn't surprising when it's low probability, it's surprising when it's probability is small *compared to* other possible events.
What the argument is doing is analogous to looking at a "4321" license plate and thinking "if there are 1000 license plates, I just looked at a 1/1000 prob. event. If there are 10000 license plates, I just looked at a 1/10000 event. The former is 10 times more likely than the later."
I have a whole series of posts on anthropic reasoning starting from here:
https://www.lesswrong.com/posts/HQFpRWGbJxjHvTjnw/anthropical-motte-and-bailey-in-two-versions-of-sleeping
Specifically about doomsday argument:
https://www.lesswrong.com/posts/YgSKfAG2iY5Sxw7Xd/doomsday-argument-and-the-false-dilemma-of-anthropic
In general people seem to be horribly confused about anthropic reasoning, which originate from not properly understanding probability theory and its limitations. Common mistake goes like this:
1. Be not randomly sampled from a set.
2. Assume that you are randomly sampled from the set, anyway
3. Arrive to a wrong conclusion
or a bit more generally:
1. Participate in a probability experiment A
2. Assume that you participate in a probability experiment B, that is not A
3. Arrive to a wrong conclusion
Sadly, the confusion is so pervasive that almost noone is thinking about anthropics reasonably. Most mainstream anthropic theories are just arguing which crazy bullets to bite, instead of trying to find a systematic ways not to do it and arrive to correct conclusions.
Thank you for the recommended posts, will give it a read.
The two approaches you describe are the Self Sampling Assumption (SSA), where you consider yourself as randomly sampled from all possible observers (in this case, all people that will ever live) and the Self Indication Assumption (SIA), where instead you should weight across all possible observer-moments, adjusted by the probability of that moment (in this case that means that you'd weight worlds with more people heavily).
If you've heard of the Sleeping Beauty paradox, SSA results in a thirder position and SIA in a halfer position. (EDIT: Other way round, see correction below)
The Doomsday argument uses the SSA, and doesn't hold if you take the SIA (because the SIA weights timelines with lots of humans more highly). I think that's what Scott's referring to in your linked post.
However, the SIA has its own paradoxes. For example, if scientists had two theories of the universe, one in which the universe would have trillions of lives and one in which it would have hundreds of trillions, the SIA would imply we should prefer the latter just because it has more observers. (This is known as the Presumptuous Philosopher argument).
> If you've heard of the Sleeping Beauty paradox, SSA results in a thirder position and SIA in a halfer position.
The other way around. SIA leads to thirdism, while SSA leads to halfism. And you do not actually need to subscrive to SSA to be a halfer (double halfer) in Sleeping Beauty, but that's a different story.
Thanks! I wrote this before stepping away from my phone and then realised I must have got it backwards. Cheers for the correction!
I tend to think of the whole approach as cursed and probably hopeless. You're trying to best-guess distributions of possibilities that we cannot check even in principle, and the margins of error are so big that you can convince yourself of "reasonable-sounding" numbers, then apply some bayesian formulas, and get overwhelming probabilities about things that we have no actual information about. Count me out.
My take is that all of the above is true, it's just extremely weak evidence and you need to take that into account when updating your priors. It's easier to see this if you apply this logic to more human scale scenarios.
For example, since moving to a new city in August, I have purchased 5 pints of chocolate fudge brownie ice cream. Obviously it's more likely for the pint of ice cream currently in my freezer to be 1 out of 5 total pints that I've purchased than 1 of 500. So I probably won't end up buying very many more pints of ice cream (probably good for my health!)
Estimating my lifetime ice cream purchases from this information alone probably feels pretty sketchy. You would want substantially more information, and it would be easy for that information to change your best guess of how many pints of ice cream I'll end up buying. Similarly, a doomsday argument *just shouldn't change your mind all that much* if you have... basically any other priors about the longevity of the human race.
The error here is step 4.
It is absolutely true that the observation "I am the nth person" is more likely if there are going to be a total of n people than a total of 4n people.
But the implication of Bayes is that observing that should make you increase your estimate the relative likelihood of there being n as opposed to 4n people from your prior, not that it is more likely that there will be n than 4n people - if your prior on 4n people was significantly higher than your prior on n people - which it probably should be - then your posterior may well still be higher.
Currently the implied probability (to the nearest whole percentage point, mid-price) of the following people winning the US Presidential Election is, in the order Betfair/Polymarket/Metaculus/Manifold (numbers 4 weeks ago in brackets):
Trump: 60/66/54/56 (48/49/44/46)
Harris: 39/34/46/43 (51/50/57/53)
UK bookmakers' odds imply 60/40. I understand Nate Silver is at 53/47.
Besides the obvious shift in favour of Trump, it is notable that the gap between Polymarket and Metaculus on Trump's chance of winning has increased to 12pp, up from 5pp 4 weeks' ago. Previously, these were never more than 4pp apart (and not always in the same direction).
If one were to take at face-value the claim at https://www.astralcodexten.com/p/mantic-monday-12924 that Metaculus is more accurate that real money markets specifically at election predictions, it would follow that there is a profitable bet to be made for those who are able to bet on Polymarket.
That is, if Harris's true chance of winning is 46% and you buy 1,000 shares at 33.8c each, you lose $338 54% of the time and gain $662 46% of the time, so you make $122 (less transaction costs) in expectation. To be clear, the most likely outcome is still that you lose all your money.
If you can't bet on Polymarket, but you can bet on Betfair, you do less well, but (on the same assumptions) still have a bet with positive expected value.
You get a similar result if you take Nate Silver to be an accurate forecaster.
A note of caution is that on 4 March 2024, Metaculus had Biden as 45% likely to win while Polymarket had Biden at 32%. With the advantage of retrospect, Metaculus predictors were overconfident that Biden would be the nominee, while Polymarket predictors were correct is assigning a significant chance to the possibility he would drop out (although largely incorrect in predicting who would replace him as nominee).
One possible model would be that Polymarket was well calibrated up to roughly the end of August, but is losing accuracy as the election approaches because of an influx of "dumb money". If correct, one might expect this trend to continue, suggesting that it might be better to wait until even closer to the election before placing bets.
Right, I'm definitely sticking a tenner on Trump! Even if he loses, I will feel the smug glow of moral vindication when Harris keeps her promise to Build The Wall and everyone conveniently forgets this was Wicked And Evil because Trump proposed it.
https://www.hindustantimes.com/world-news/us-news/kamala-harris-tacitly-agrees-to-strengthening-trumps-wall-at-cnn-townhall-101729745809315.html
This is a common misperception. Trump only talks about the wall a lot, but its construction significantly predates him. The border wall goes back to the Clinton administration with the Illegal Immigration Reform and Immigrant Responsibility Act of 1996. During W’s administration, the Secure Fence Act authorized the construction of roughly 700 miles of fencing along the U.S.-Mexico border.
Under the Obama administration, much of this previously authorized fencing continued to be built. In fact Trump recently made a campaign stop at “his wall” to talk about how he was tough on immigration. The impressive section of wall in front of which Trump made his speech was constructed before Trump ever held office, under Obama’s administration.
There have been some very weird massive investments by single traders on polymarket recently. French guy put in $45 million on trump last week. https://x.com/Geiger_Capital/status/1849478811571585286 https://x.com/rajivatbarnard/status/1849573043170955276 Unknown person put in $3 million this week, including $274,300 worth of shares at 99.7%. https://x.com/punk9059/status/1849622187554050535 . So something funky is definitely going on.
While it should in theory be possible to arbitrage, the cut that the platforms take, and the third party risk of storing your money with them, is such that a lot of people would consider it not worth it.
I'm not sure what you mean by "funky", but I see no reason to think these bets are anything other than people with a lot of a money who expect Trump to win. The guy who bought 275,109 Trump Yes shares at 99.7c was a new account and I'm sure it was an error. He's since made a large number of trades in the same market at normal prices: his total position is 7,542,442 shares. If anything it's a warning about Polymarket's lack of protection for punters.
Some people can do the arbitrage. Polymarket states that they do not charge fees https://learn.polymarket.com/docs/guides/trading/fees, although their terms are not at all transparent and there likely are some costs involved. Betfair will charge 2% of profit in any given market (on their "basic" reward plan). The spread between the two sites is sufficiently high that you should end with a tidy profit, but the practical difficulties are significant. To do it properly you would need to hedge the currency risk. The sites also have slightly different resolution criteria which in theory could result in them having different outcomes (e.g. if AP, Fox and NBC all call the election for Harris, but Trump doesn't concede and ultimately Congress certifies Trump to have won, then it will count as a Harris win on Polymarket and a Trump win on Betfair). Nevertheless, I would expect the gap between the real money markets to close. I note that Betfair has moved to 62/37 and the bookies have moved to 62/38 since my first comment, which is the direction I would expect the correction to go given the extraordinary volumes on Polymarket.
ETA Bloomberg are wrong to say that the French guy bet $45m: he currently holds 47m *shares* across the 4 named accounts, which he bought for a total of about $26m. I'm guessing he held 45m shares when they checked.
Why would you assume that they *expect* Trump to win, as opposed to *wanting* Trump to win?
Prediction markets have not grown up to be the knowledge equivalent of stock markets, where serious people go to *make* money. Prediction markets are, for everyone but a handful of nerds, a place where people go to *spend* money. Rather like Las Vegas - loads of fun if you're into that sort of thing, and maybe good for your status or self-image if you don't lose *too* much, but if you're a Yankees fan who thinks the Dodgers are the 60/40 favorite in the World Series, you're still not going to bet on the Dodgers because that wouldn't be any fun.
And, like Las Vegas, that does make opportunities for smart, disciplined nerds to make a bit of money around the edges. But to extract useful signal from that, you'd need a way to distinguish smart-disciplined-nerd money from status-seeking-whale money. Do you have one?
In a reasonably well functioning prediction market, the disciplined nerds are going to jump on any pricing discrepancies caused by the dumb whales.
The best evidence that this is happening is that all prediction and betting markets around the world are currently in agreement that Trump's chance of winning is closer to 60% than 50%. If the markets were vastly inefficient and dominated by dumb money we'd see a wider range between different real-money markets. Only the fake-money markets are close to 50-50.
We don't yet have a well-functioning prediction market. We have well-*administered* prediction markets, which can be mostly counted on to e.g. pay out as called for by the propositions, but that merely a necessary and not a sufficient condition for a well-functioning market.
For that, you also need a large community of well-informed, well-capitalized traders with experience in the market at hand, and it's too young for that. There are a few disciplined nerds who will nibble happily and profitably at any price discrepancies, but not enough to arbitrage those discrepancies out of existence,
I don't think those things are as distinct as you suggest. Wanting Trump to win, they convince themselves that he will. Similarly, I don't think your Yankees fan *does* believe the Dodgers are 60/40 favourite. They can give you twenty reasons why the this is the Yankees' year, so it looks to them that the bookies are selling dollars for 40c.
I'm sorry to say this, but I think people do go to Vegas to make money. They usually fail, but that doesn't mean they're not trying.
What I am very tentatively suggesting is that the smart disciplined nerds congregate at Metaculus, so one can read off the true probabilities from there and bet accordingly. But also, you don't need a general rule if you are confident you've spotted the whale in this particular game, which it kinda seems like you are.
Looking at bets over $100k on Polymarket, there have been 13 in the past 24 hours, by 6 different users. Totals are $1.879m for Trump to win, $132k for him to lose, $325k for Harris to win. So it does look like whales are breaking for Trump.
I am tempted to do exactly that - bet a small amount, as in, something like monthly subscription to this blog, on Harris on Polymarket. Does anyone by any chance know what are the costs on Polymarket? I don't own any relevant crypto. I've asked Perplexity and it answered with this not-very-quantitative and not-very-encouraging lecture (Polymarket's own info is even less informative):
"Cost Structure on Polymarket
Trading Fees:
Winning Trades: Polymarket charges a 2% fee on net earnings from winning trades. This means that if you win a bet, 2% of your profit will be deducted as a fee15.
Losing Trades: There are no fees for losing trades, which makes it more appealing for users who may not always win their bets1.
Transaction Costs:
Polymarket operates on the Polygon blockchain, which generally has lower transaction fees compared to Ethereum. This helps keep costs down for users engaging in frequent trading13.
Users may incur gas fees when depositing or withdrawing funds, but these fees are paid directly to network miners and not to Polymarket itself2.
Deposit and Withdrawal Fees:
No Direct Fees: Polymarket does not charge fees for deposits or withdrawals, which is advantageous for users looking to manage their funds without incurring additional costs13.
However, when transferring USDC (the currency used on the platform), users may encounter gas fees associated with the Ethereum network if they are not using Polygon2.
Liquidity Provider Fees:
When trading, users might also pay small fees to liquidity providers who facilitate transactions on the platform. These fees are generally minimal and help maintain market liquidity34.
Market Prices:
The prices of shares on Polymarket reflect the probability of an event occurring and can fluctuate based on supply and demand dynamics within the marketplace45."
Then I found out that "gas fees" are apparently measured in a unit called gwei, and then I've exhausted my attention span
I would also be interested to know how this works out, i.e. if you currently have some fiat currency in a bank account, what is the total transaction cost of buying USDC, moving it to Polymarket, placing a bet, collecting your winnings, converting to fiat and redepositing in the original account?
At one point I thought I would try it to find out but I was unable to deposit USDC to Polymarket; when I contacted Polymarket they informed me that trading is not available to UK users (although there is still nothing in their Terms to indicate this).
By way of update, currently we have Polymarket at 65/35, with Betfair at 62/37, which matches my prediction that the gap between the two real money markets would narrow. I don't necessarily expect them to get closer than 2pp: my gut feeling is that the inefficiencies are sufficiently large that a difference of that magnitude could be sustained indefinitely. Meanwhile, Metaculus is at 52/48.
I note that over the past 24 hours, there have been 9 bets over $100k on Polymarket, by 8 different accounts, and all of them have been bought Trump Yes. It therefore seems to me that the proposition that whales are breaking for Trump still roughly holds, so I still weakly expect (ceteris paribus) the odds to shift in Trump's favour between now and the election.
Scott Aaronson, November 2016:
"In this particular case, of course, we know for a fact (!!) that Russian intelligence was engaging in cyber-warfare to influence the US election. We also know that Russia has both the will and the technological ability to tamper with foreign elections using vote-stealing malware—indeed, it nearly succeeded in doing so in Ukraine’s 2014 election. Finally, we know that Trump, despite losing the popular vote, surprised just about everyone by outperforming his polls in three crucial swing states—and that within those states, Trump did systematically better in counties that relied on electronic voting machines than in counties that used scanners and paper ballots."
Scott Aaronson, November 2020:
"These past few weeks, my comment queue has filled with missives, most of which I’ve declined to publish, about the giant conspiracy involving George Soros and Venezuela and dead people, which fabricated the overwhelmingly Democratic votes from overwhelmingly Democratic cities like Philadelphia and Milwaukee and Detroit (though for some reason, they weren’t quite as overwhelmingly Democratic as in other recent elections), while for some reason declining to help Democrats in downballot races. Always, these commenters confidently insist, I’m the Pravda-reading brainwashed dupe, I’m the unreasonable one, if I don’t accept this."
Unlike so many others, he does often sound like someone trying to take a principled position and to clarify the things he agrees with his opponents on (e.g. he does say in both posts that he's been long and consistently in favour of hand-recounting every election). And yet even HE is able to express almost completely opposite attitudes to similarly unsubstantiated "my side didn't really lose" conspiracy theories. Most people are a MILLION times even worse.
When is this kind of brazen hypicrisy going to stop? And more importantly, when are people going to realise that this kind of thinking is at the root of everything that's wrong with politics? Every single person who talks with disbelief and fear about the way Trump behaves needs to ask themselves if their reaction to stolen election claims was completely different in 2016 and 2020, and if it was to take a good look in the mirror and understand that everything they hate about Trump is a product of the society they've created.
Bret Devereaux just posted on how, in his professional opinion, Trumpism has become alarmingly fascist. His essay is here: https://acoup.blog/2024/10/25/new-acquisitions-1933-and-the-definition-of-fascism/
He ends it by mentioning this:
> By far the most common taxonomy of fascism you will see in most academic circles is that advanced by Umberto Eco in his essay “Ur-Fascism,” copies of which you may easily find online. Like most things academics like, it is complex and a bit fuzzy.
> As Eco notes, fascism as a set is somewhat like the series of “abc bcd cde def” in which all of the elements are clearly a family and yet in freely remixing core elements, it is hard to identify a single set of necessary components.
> instead of a single definition, Eco proposes a taxonomy of 14 points which collectively make up the ‘type group’ of fascism. An ideology doesn’t need to have every point in order to qualify as fascism, but the more criteria it satisfies, the more firmly it fits into the definition. Likewise, individual points can also be fuzzy or borderline – it is the cumulative fit that matters – after all, most ideologies have effectively 𝗻𝗼𝗻𝗲 [emphasis original to Devereaux] of these attributes, so an ideology that clearly has many and sort of has a few more actually fits the mold quite well.
So far so good. But these are the 14 points:
1. The cult of tradition, particularly a syncretic traditionalism that latches on to various iterations of an idealized past, even mutually incompatible ones.
2. The rejection of modernism, specifically, in Eco’s mind, a rejection of “the Enlightenment, the Age of Reason…seen as the beginning of modern depravity.”
3. The cult of action for action's sake.
4. Disagreement is treason.
5. Fear of difference.
6. Appeal to a frustrated middle class.
7. The obsession with a plot.
8. The deceptively strong/weak eternal opponent. This one is complicated, but fascism conjures an ‘enemy’ who is at once too strong (thus requiring the power of the fascist strongman to defeat and whose continued existence can justify continued mobilization and authoritarianism) and yet also degenerate and weak: thus the Nazi view that Jews were racially inferior, weak and cowardly, but also that they secretly controlled all of the world’s most powerful countries.
9. Life is permanent warfare, as Eco puts it, “there is no struggle for life, but rather, life is lived for struggle.”
10. Contempt for the weak.
11. The cult of heroism.
12. Machismo.
13. Selective Populism: fascism claims to speak for ‘the People,’ but in reality only some of the people and not through democratic, majority-rules systems; rather the Leader channels the Common Will which is taken as the Voice of the People, even when it contradicts the actual votes of the people.
14. Newspeak, which Eco identifies both in changing the meaning of words, often inverting them, but more broadly, I think Eco is getting at here what Orwell sums up in his famous line, “The party told you to reject the evidence of your eyes and ears. It was their final, most essential command.”
[He provides discussion of each point, at the link.] In conclusion:
> Fourteen out of fourteen; some are clearer and stronger fits than others, but every element is present to a significant degree. Keep in mind, this is the sort of taxonomy where a regime which, say, satisfied 10 or 12 out of the 14 would still be generally regarded as fascist and the ideology we might call ‘Trumpism’ 𝙛𝙞𝙩𝙨 𝙖𝙡𝙡 𝙛𝙤𝙪𝙧𝙩𝙚𝙚𝙣.
And this looks completely insane to me. Let's just stipulate that points 8/9/11/12 don't apply to the US Democratic party. The other ten clearly do. And I'm pretty sure that five of these points, #s 1, 3, 5, 6, and 13, apply to every political movement that has ever existed, modulo slight cultural translations. What happened to "even one of these would be alarming"?
1 (or 5, I think) would not apply to the Civil Rights, the European Socialists or Social-Democrats, not even the Modernists, the beginnings of the United States, or whatever the American political consensus was in the 1990s?
3 also seems to not apply that often (but my perspective may be twisted as a non-American). Saying “oh well, this seems important issue, surely we should study this more before doing anything rash” seems to have been uncontroversial for a long time.
This is why there are so many think tanks putting ideas out there and publishing studies rather than physical tanks vying for political control (obviously, the second part is a jest).
Note ACOUP’s elaboration on this: for Eco, the worship of action for action’s sake is coupled with a disdain for learning and thinking. It’s not “we’ve heard all sides and thought it through, now is the time to take action”. This is, in spirit, closer to “let’s get all these useless [sneer] thinkers back to doing real work”.
I also disagree about the universality of 6 and 13, but it’s harder for me to articulate precisely why.
6 is far from globally universal, but has been a common talking point in the US for the past 20 years or so.
Oh? What are some movements that don't or didn't promise to make people better off? The message "actually, people are doing fine" is the antithesis of a movement.
Are we talking about the same one?
> 6. Appeal to a frustrated middle class.
Not every place even *has* a middle class to speak of, so obviously this is not universal.
That's why I said "modulo slight cultural translations". If 94% of the population is subsistence farmers, then "appeal to a frustrated middle class" will be "appeal to a frustrated peasantry". There's nothing special about the "middle class" as defined by American usage.
> 1 [...] would not apply to [...] Civil Rights
Well, going over some speeches from https://www.goodgoodgood.co/articles/best-civil-rights-speeches :
JFK:
> One hundred years of delay have passed since President Lincoln freed the slaves, yet their heirs, their grandsons, are not fully free. They are not yet freed from the bonds of injustice. They are not yet freed from social and economic oppression. And this Nation, for all its hopes and all its boasts, will not be fully free until all its citizens are free.
Martin Luther King:
> On our part we must pay our profound respects to the white Americans who cherish their democratic traditions over the ugly customs and privileges of generations and come forth boldly to join hands with us.
Martin Luther King, same speech:
> Let us march on ballot boxes (Let us march. March) until brotherhood becomes more than a meaningless word in an opening prayer, but the order of the day on every legislative agenda.
> Let us march on ballot boxes (Yes) until all over Alabama God’s children will be able to walk the earth in decency and honor.
> There is nothing wrong with marching in this sense. (Yes, sir) The Bible tells us that the mighty men of Joshua merely walked about the walled city of Jericho (Yes) and the barriers to freedom came tumbling down. (Yes, sir)
Joseph Jackson:
> As Christians we are a part of our nation and a part of the struggle of America. America was brought into being to satisfy and to answer the human longing for freedom. There was the urge in man to be related to other men as men without a modifier or any kind of limitation or restriction. There was an awareness of a human kinship deeper than race, more profound than nationality, and more inclusive than any accepted religious creed. In addition to the quest for a new geographical spot there was a search for a new human relationship, a new freedom, and new opportunities. These basic urges inspired the early colonies to brave the dangers of a rough and unknown sea, and seek a land in which they could live as free men and aspire to the highest possible goals of life without the enslavement of the past or being the victims of the determinism of enforced circumstances. They wanted a chance to explore and to search out the meaning of life for themselves, and an opportunity to worship God according to the dictates of their conscience.
> This American venture is powerful but not perfect; ever growing but not grown; and still becoming, but is not yet complete. The kind hand of destiny and the benevolent providence of Almighty God have placed the American Negro along with other races and nationalities in this flowing stream of the nation’s life for which we are justly proud. As patriotic Americans we are devoted to our nation’s cause, and are wedded to its ideals and principles.
Malcolm X:
> When this country here was first being founded, there were thirteen colonies. The whites were colonized. They were fed up with this taxation without representation. So some of them stood up and said, “Liberty or death!” I went to a white school over here in Mason, Michigan. The white man made the mistake of letting me read his history books. He made the mistake of teaching me that Patrick Henry was a patriot, and George Washington… Wasn’t nothing non-violent about ol’ Pat, or George Washington. “Liberty or death” is was [sic; https://www.rev.com/blog/transcripts/the-ballot-or-the-bullet-speech-transcript-malcolm-x ] what brought about the freedom of whites in this country from the English.
> They didn’t care about the odds. Why, they faced the wrath of the entire British Empire. And in those days, they used to say that the British Empire was so vast and so powerful that the sun would never set on it. This is how big it was, yet these thirteen little scrawny states, tired of taxation without representation, tired of being exploited and oppressed and degraded, told that big British Empire, “Liberty or death.”
Moving on...
> 1 [...] would not apply to [...] even [...] the beginnings of the United States
Why do you think the more powerful half of our legislature is called the "Senate"?
Heck, I'm not even sure 8/9/11 don't apply to Democrats (oppression / capitalism are the eternal enemies; we're always fighting a war somewhere to export democracy abroad; the people who fight oppression / capitalism / whatever are our heroes). I can concede 12, but if something meets the other 13, I'm not likely to point at 12 and claim there's nothing to worry about.
My irritation otherwise parallels yours. My reaction was something like "looks like my ham sandwich is fascist". I concluded that I had better wait until I read a bunch of other Eco material before I form an opinion of him, because that alone was hackneyed.
Haven’t the recent foreign US wars (especially of the “to export democracy kind”) been instigated by Bush?
Obama refused to get involved in Syria even after Assad defied the stated “red line” about chemical weapons, if I remember correctly, and tried to appease Iran with the JCPOA (where Trump took a far more hawkish stance by denouncing it).
Deveraux explicitly states that 8 is a subtle point, and the weak/strong ambivalence does not apply as straightforwardly to capitalism or oppression.
Political movements usually *want* something, and have to struggle with people who oppose them. But 9 refers to the glorification of struggle in and of itself as the one purpose of life.
An anti-capitalist fights capitalism because life will be better after. Washington fought the British for American self-rule (or at least no taxation without representation, or so the myth goes). Johnson didn’t fight the Vietnam War because war is glorious and to make men out of the hippies, but for geopolitical reasons (the Southern ruler was an ally of the US, the ideas of containment and roll-back, the desire to protect the South Vietnamese population from communism…).
(Also, it seems wrong to conflate the Democratic party with the anti-capitalists? A serious anti-capitalist would hardly admit to any remotely significant economic difference between the Democrats, the Republicans, the Libertarians, and European-styled social-democrats. I’m extremely skeptical of any material influence that the “cultural anti-capitalists” that show up online could wield.)
11: I’d be surprised if Eco thought that all heroes (or cults) were alike. My guess is that there’s a very specific notion of heroism he has in mind, closer to the models of Hercules or El Cid (heroes that are great by war, conquest, or their raw strength), rather than a builder (in a wide sense), an artist, a thinker, or someone who spearheaded a great charitable effort.
Bush only started Afghanistan and Iraq with Democratic Congressional approval. Obama sent CIA operatives and air support into Syria. Clinton sent troops into Bosnia. Point being, it's not the case that Democrats never do this, and meanwhile, they're still our #1 exporter of rhetoric about fighting against oppression and racism and sexism and so forth.
Devereaux's explanation of 8 describes a duality of "too strong" but also "degenerate and weak", "often just a nebulous 'they'". That sounds like systemic racism to me.
I admit that "we're always fighting a war somewhere" (for its own sake) wasn't quite hitting me as a textbook example of "life is permanent warfare", but then you rephrased it as "glorification of struggle in and of itself as the one purpose of life" and I instantly thought of "revolution for the hell of it" and people who activists and look down on anyone who isn't as "sheeple" who "aren't paying attention".
I don't think it's fair to justify Washington's and Johnson's wars and then not look at the justifications for Bush's or Trump's continuation of them. We should either look at both, in which case none of these is fascist, or we ignore them all and point out their existence, in which case they all are. Which is again why I don't like this litmus test.
I don't think every Democrat is anti-capitalist, sure. But that's the side the anti-caps camp on. If we're looking for people who profess that eternal struggle, we're not going to start looking on the GOP side.
I'm sure Eco didn't see all hero cults the same, either, but if he's going to try to explain to us how to spot a fascism, I don't think he really meant "how to spot what meets Eco's personal peccadillos as fascist and naturally this will be an objective test". So I'm going to go by the picture I think he's trying to paint. I originally just trusted the comment above to tell me, but if I go to the original text: "The Ur-Fascist hero craves heroic death, advertised as the best reward for a heroic life. The Ur-Fascist hero is impatient to die. In his impatience, he more frequently sends other people to death." Are we to infer that Democrats don't revere people who will risk their lives to fight oppression / bigotry / capitalism / etc.? Earth First? Weather Underground? The PLO? Gabby Giffords? Heather Heyer? Antifa?
Sorry, I didn’t mean to sidetrack the discussion by bringing Bush up. I felt from your comment that you were essentially accusing the Democrats of being in favor of permanent warfare.
(Although, if I remember correctly, he’s the one who decided it, lied brazenly to get his way, and that did a lot of damage to the US credibility. But that’s irrelevant.)
I think that “permanent warfare” does not refer anyway to “state-on-state warfare”, or at least not necessarily. The Nazis’ designated “racial” enemies did not come from a particular state.
Part of me wants to jest that Umberto Eco was a European leftist, so the difference between Democrats and Republicans would have barely been perceptible to him.
But (one of) Bret Deveraux’s points is that Trumpism is not close to the Republicanism of Reagan, Bush, McCain, or Romney.
The dimension that I don’t find in your comments (but is crucial to the appreciation of phenomenon) is that of scale, or perhaps centrality. Concentration, a chemist would say.
I’m sure that in [sufficiently large group that you personally respect], there will be a lot of [bad things]. The issue is when the bad is prevalent – especially when [said bad thing] it’s part of the group’s identity.
Sure, a lot of political movements will have their extremists, or snub activists that will scoff at anyone that doesn’t embrace and spend their lives fighting for their pet cause.
(Although, again, an anti-capitalist who identifies as a Democrat is about as silly as an astrophysicist who believes in astrology.)
But what I think Eco is saying here is that for a fascist movement, this exaltation of life as a struggle does not come downstream of having extremists or people deeply committed to some aspect of the cause. It is a central tenet of the ideology (perhaps the word “vibe” would be more appropriate if the movement scorns intellectualism). Hence the necessity for an enemy – instead of, say, a challenge or a shortcoming of our societies to overcome.
I don’t understand how “systemic racism” fits as a permanent enemy in point 8, could you please elaborate?
Could we perhaps clarify what your final examples intend to prove? They seem to be a very loose mix without a strong common theme. In Eco’s definition, both the longing and the giving of death seem important.
This is why the two individuals that you cite seem, in my opinion irrelevant to the discussion – they were participating in pretty regular political processes until they were attacked (from a one-second skim of the Wikipedia pages).
The Weathermen seem to have avoided this vibe as well, although, few people would find much to admire in them.
The PLO is an entirely different kettle of fish, since it is literally a terrorist organization.
Antifa is admittedly quite a bit less clear-cut, but it’s also non-central.
If you're going to get in a parting shot, I'm going to have to return volley. :-) (As I understand it, GWB didn't lie brazenly about Saddam so much as tasked intelligence to gather evidence, and reported on what they found, which was going to be something anyway, and did so with the agreement of Congress and most Western nations who had their own intel. Bipartisan level view here is that neither account is agreed upon.)
But yeah, I can't in good conscience characterize the DP as pro-permawar. Pro-perma-protest, however, is certainly on the table. It's just not "machismo" in the Eco sense.
I don't see enough chemical concentration of bad in Trumpism to find Devereaux's claim convincing. Trumpism isn't "unusually bad" nearly as much as it's "made to look unusually bad by a group motivated to do so", and the evidence is hundreds of unfair exaggerations. Meanwhile, Devereaux isn't being scientific; if he were, he'd say, "so, wait a minute" and hold the Democrats up to those 14 points, as Michael Watts did above. And then he'd (I contend) be back to: either everyone's fascist or these points are too underspecified, as I've been arguing here.
I don't see why it should matter whether the life-struggle idea comes intrinsically or is downstream of extremists. First off, fascism itself doesn't arise from a vacuum; any group adopting it is responding to past events anyway. Secondly, we're not supposed to like that idea regardless. So arguing origins sounds like we're arguing whether this thing before us is "technically" fascism or not - as if federal funds will kick in or something if we can secure the credential.
"Systemic racism" is the permanent enemy in the sense that no matter how hard one tries, it'll be there. Its proponents present it as a grime that gets into everything, necessitating nothing short of a radical restructuring - or a permanent cadre of watchdogs, which the left is only too willing to provide in exchange for whatever resources and cultural commitment they determine. Fascism seeks the same sort of sinecure.
I get the sense you're unfamiliar with the history of the Weathermen. They made the news for bombing all sorts of institutions in the name of leftist causes. Bill Ayers' own girlfriend was killed while making one of them. He, and everyone else in that paragraph, are people who either risked their lives or risked others' lives for the cause, and are admired by the left more than by the right for it. This includes some Democrat voters.
"The Ur-Fascist hero is impatient to die. In his impatience, he more frequently sends other people to death."
> I concluded that I had better wait until I read a bunch of other Eco material before I form an opinion of him, because that alone was hackneyed.
I had a different concern; Devereaux explicitly notes that this framework is used by modern scholars who want to discuss fascism as a topic, and based on the contents of the framework, I can't see that being useful for any purpose other than forming an echo chamber.
Having a literal mind, I tend to see "fascism" as being defined by point 4, which is what's emphasized by the name "fascism".
I always saw "fascism" in terms of "fascia" - generally speaking, a casing around a bundle of threads or fibers or cables that holds them together and makes them stronger. Which, by itself, sounds fine and sensible. "Stronger, Together."
The side of it that worries people who are students of history is when someone like Mussolini steps up and claims "everything within the state, nothing outside the state", meaning not just "hey, we'll get farther if we work together" but also "anyone who doesn't is going to get smushed by my fascia and everyone inside it, hint hint".
I notice that practically no one who espouses some version of "Unity" or "Stronger, Together" doesn't seem eager to add "But If You Really Prefer to Go Your Own Way, That's Cool, Too". Well, except for libertarians.
Now, the more I think about that essay, the more I think it would have gotten farther with everyone if treated as "Here's a Rough Sketch of Warning Signs That Your Society *Might Be* Headed in the Wrong Direction", rather than the apparent treatment of "Run This Program on a Society and if it lights up Red, Drop Everything and Dismantle It". And only run it on whatever one didn't like already.
The only reason I think conservatives aren't as susceptible to this is merely that few conservatives read Eco.
As I understand it, the word is derived not from "fascia", a casing around a bundle, but from "fasces" (or the Italian reflex "fascio"), referring to the bundle itself. The idea is that while it's easy to pick up one stick and snap it in half, you can't do the same with the bundle; numbers and alignment lend it strength.
> I notice that practically no one who espouses some version of "Unity" or "Stronger, Together" doesn't seem eager to add "But If You Really Prefer to Go Your Own Way, That's Cool, Too". Well, except for libertarians.
Well, I don't think the libertarians are doing much to espouse unity. They're fine with you going your own way, but they don't have a plan for anything other than that.
The reason no one tries to send both of those messages is that they're inherently opposed.
> The only reason I think conservatives aren't as susceptible to this is merely that few conservatives read Eco.
In a sense, I agree. There's nothing unusual about the impulse to make up some criteria that describe everyone and argue that those criteria show that someone you hate is bad, while glossing over what they might show about anyone else.
Academic scholarship is very slanted, such that errors are going to be overwhelmingly pro-leftist, but that's just because academics are leftist.
This was a funny example of that slant: https://pubmed.ncbi.nlm.nih.gov/34383522/ . ("Authoritarianism has been the subject of scientific inquiry for nearly a century, yet the vast majority of authoritarianism research has focused on right-wing authoritarianism. In the present studies, we investigate the nature, structure, and nomological network of left-wing authoritarianism (LWA), a construct famously known as "the Loch Ness Monster" of political psychology.")
You apparently think that the 2016 and 2020 were the same in ways that Scott Aaronson does not. Perhaps you think that being concerned about Russia, which has sophisticated cyber attack capabilities and a demonstrated intent to interfere in the 2016 election in favor of Trump, is the same as proposing a wild conspiracy involving Soros (who is pro-democracy) doing the same thing. Perhaps you think Scott’s proposed response (performing a hand recount and accepting the result) is the same thing as the Trumpest response (trying to get the election results throw out). I say “perhaps” because you don’t really engage with Scott’s views at all.
You accuse Scott Aaronson of “hypocrisy,” an accusation that he is acting in ways inconsistent with his professed beliefs, indicating that he is being dishonest about those beliefs. Since you don’t identify any actions taken by Scott other than writing in his blog, the accusation is ambiguous. Sure, writing a blog entry is an action, but which blog entry constitutes the action demonstrating Scott’s true beliefs and which is the one where Scott is being dishonest about those beliefs?
To summarize:
1) You make a claim about the similarity between the 2016 and 2020 elections but don’t provide a clear statement of this claim or support it with evidence.
2) You claim that Scott’s views in 2020 are inconsistent with his 2016 views. This claim that is unsupported because it depends on the preceding claim.
3) You claim that this inconsistency shows bad faith on Scott’s part. Even if you could demonstrate an inconsistency, which you haven’t, that still wouldn’t establish bad faith, because people can honestly change their opinions over a four year period.
In short, I believe you have utterly failed to make your case.
I am saying that he is applying wildly different standards to extremely similar partisan claims of election rigging, that both lack any hard evidence and rely entirely on "this sure looks suspicious" vibes. In the case where the claim benefits the Democrats, he describes it in the most charitable and credulous way. In the case where it benefits the Republicans, he strawmans it and points out all its flaws. (I have barely ever heard Soros mentioned by 2020 election deniers--the main thing discussed is always the last minute pandemic-based voting changes.)
You can easily imagine a Republican version of Scott Aaronson (perhaps gay, and having been viciously bullied by his party but still slavishly loyal to them, to match the Democrat version) saying in 2016:
"This past week, I've gotten endless rants about the shadowy conspiracy involving Russia and the FBI, Facebook and the New York Times, all working in concert with a nebulous White Supremacist Patriarchical Agenda to hack machines and fabricate votes for Trump (but for some reason not fabricating enough votes to give him a popular vote win and greater legitimacy). They also confidently tell me that Trump is supported by all the most powerful institutions in our society, and in the same breath that he's "not normal" and "a deplorable" and an unprecedented threat to the security of society. Funny how that works."
And saying in 2020:
"We know for a fact (!!!) that Democrat officials have changed voting rules at the last minute and without proper authorization across the country. We also know that Democrats, despite falling far short of expectations across all down-ballot races, did systematically better in the presidential race, and that their vote was consistently higher in those states that adopted the voting rule changes."
And maybe you can see my point? There are maximally charitable and maximally uncharitable ways of presenting (in these cases bullshit) claims, and he chooses opposite approaches that concord with partisanship.
And I acknowledged that he is more consistent than most in his support for hand-recounts. It's not using the Russia claims as a reason for a hand-recount that's the problem, it's signal boosting bad epistemology that radicalises people and that was later used many times to justify various socially destructive actions, up to and including violent riots, just like Trump.
Your point about hypocrisy is confusing and I think greatly overcomplicating things. I'm using "hypocrisy" to mean applying inconsistent standards to the same thing (claims of election rigging that lack any direct evidence) depending on whether it benefits your side. That's it.
As for "he may have just changed his mind", I haven't seen any acknowledgement of that, and moreover that seems like a way to dismiss any charge of hypocrisy ever. Maybe the moralistic preacher just genuinely changed his mind about the importance of chastity just before he visited the brothel, and changed back as soon as he finished there! It's an isolated demand for rigor.
A few years ago, before Roe v Wade was overturned, I remember reading a Texas law banning abortion but somehow managing to escape judicial contest through some sort of loophole (was it banning the clinics, but explicitly not enforcing the ban, yet rewarding isolated citizens who did, so that no one had standing to contest the law?).
I remember thinking that the legal argument was a bigger deal than the content of the law itself, that it would trigger a flurry of similar laws (which each side would use to ban stuff that they disapproved of).
This doesn’t seem to have been the case, so I’m wondering what happened?
1) was that law struck down as unconstitutional (ie did contesting the law work)?
2) did it become a non-issue after Roe v Wade was overturned?
3) did the legal argument work, but US politicians were not actually willing to sacrifice the spirit of the rule of law to its letter?
4) were there no comparable “political gains” to be made with such laws?
Any insight would be appreciated.
Texas instituted a private bounty targeting abortion (SB 8) and then said they couldn't be sued to stop the law because *they* weren't the ones carrying out the law. Eventually this went to the Supreme Court who narrowly agreed. However the law could still be challenged by anyone who became involved in an actual case under it.
Then Dobbs came down and rendered the whole thing moot and as a result there was never a final finding that the scheme was unconstitutional. But when the Supreme Court was giving Texas a win on standing they also expressed a great deal a skepticism that Texas would win if they decided based on the merits. This probably explains why nobody rushed to pass more laws like it.
I'm sorry that I can't think of examples off-hand, but I feel like I've heard of several laws that were at least proposed since Dobbs that worked on a similar basis. If my memory is accurate, at least a few have been passed on both sides of the aisles.
The problem, as Jeff mentions, is that even though such a law would prevent someone from contesting the law early, it doesn't prevent contesting the law entirely. The original purpose in Texas was more to dissuade people from getting abortions, not that it would ultimately hold up in court. Previous laws were insufficient for this purpose because even lower courts could quickly and easily toss an anti-abortion law under RvW and Casey.
These newer laws that I'm thinking about were not trying to solve the same problem (keeping a law on the books longer) but instead about using governmental resources. By giving private citizens a cause of action, this new type of law uses the same legal technology, but to a different purpose - specifically to enforce laws that take too many resources for the government to enforce on their own. If a negative of a law is "how could we ever enforce it?" then this type of enforcement mechanism is a good fit, at least in theory. I did think of an example that's actually been in place for a number of years - whistleblower protection laws that pay the whistleblower in the event that the government gets a fine out of the organization.
Texas didn't do anything innovative, is the fundamental problem; this particular legal strategy was developed back in the 60s or 70s by, IIRC, Ralph Nader (or at least he used it heavily, if I recall the person correctly).
It has been used extensively by the Federal Government since then, being employed for things like environmentalism and disability rights.
Normally, a government ban on something is enforced by some government agency. If someone who would be subject to the ban thinks it would be unconstitutional, they can file a lawsuit against the agency, and a judge can issue an order to the agency not to enforce the ban.
Texas tried to get clever by passing a law, SB 8, that banned abortion but specifically prohibited any government agency from enforcing it: instead, it could be enforced by any member of the public filing a lawsuit against the abortion-facilitating people. The idea was this way people couldn’t preemptively challenge it in court: suing the government wouldn’t help, since the government wasn’t doing anything. As it happened, the Supreme Court decided that abortion wasn’t constitutionally protected not long afterwards, so Texas was able to just regulate abortion directly without needing to take advantage of this dodge.
Some politicians in Democratic states made some noise about passing their own version that would infringe rights that conservatives like, but as far as I know they never went anywhere much. (The most developed, as I recall, was a proposal for guns in California.)
For whatever it’s worth, while the idea is clever, I don’t think that it actually would have worked, and in my view the abortion-rights freakout was entirely unwarranted.
> For whatever it’s worth, while the idea is clever, I don’t think that it actually would have worked
Hm? It does work. It's working right now. The strategy is just a copy of the existing functioning of civil rights law.
Is anyone else increasingly interested in cybernetics as serious foundation to build a modern political philosophy?
The existing/traditional ideologies are too rigid and too slow to adapt. Yeah, I get the risks of over reliance on data, dehumanizing citizens, and rule by a technical elite. BUT don’t we already have those things happening. My current thought is the problem lies in the system not being built around real-time data and feedback, which turns its use into an exploitation lever rather than a tool for responsive government.
The ham-fisted swaying during election season highlights how poorly incumbents handle feedback loops—namely, how weak their predictive modeling really is.
That sounds like the way (soviet) communism was going in the 1970s. They didn't really have the computers for it back then though.
Salvador Allende of Chile was going to try it before he was killed. Might be some reading material there.
Thanks for the redirects. I recall something from All Watched Over By Machines of Loving Grace that talked about early Soviet plans.
If designed properly, my assumption is a feedback system is ideologically agnostic.
I probably got the info from that exact same source, lol.
Yeah, I suspect communists and libertarians both kind of want a machine to mediate human society. Personally I think that's a fair way off and it's just easier for humans to do it for now.
Very funny! Yeah, it’s totally a fair way off which is why it’s worth thinking about now. I want to get people thinking about the nuance that humans are always doing it, it’s just whether we acknowledge that at the outset. So many articles talk about algorithms and AI as these kind of other beings, but they aren’t. They are the product of human intervention.
Some people seem to believe that Allende's Chile was one spreadsheet app away from Singularity. I guess this is one of those things the future generations may want to run some simulations of. I am skeptical because e.g. Venezuela today can make as many spreadsheets as they want; your average desktop PC probably has more computing power than Allende could get in a decade; and yet we don't see any of the promised miracles. Most likely, the problems of socialist economies are caused by something other than the lack of spreadsheets.
Spreadsheets would probably work in some kind of RPG universe where the production methods are fixed, and you can always transform "2 units of leather + 1 unit of rope" into 1 pair of shoes. Then you can use some algorithm to find the optimum production. In real life, not all units of rope are created equal (using a lousy rope will make your produced shoes fall apart), and new inventions happen (things that do not yet have a place in your spreadsheets, but require materials that do).
FWIW, this prompts me to mention the book Complexity and the Art of Public Policy by Colander and Kupers. It harkens back to observation of Scottish philosopher Adam Ferguson in 1767 that there is a category of things that are the result of human action, but not of human design. A quote from the book:
“… we believe that this complexity pattern--- that nonlinear systems can self-organize—is an important pattern for all policy makers to adopt and keep in the back of their minds. Complex systems are continually in the process of organizing themselves, and the way that process works is at odds with attempts to control systems. At best complex systems can be influenced—not controlled.”
Excellent! Added that book to my reading list. Much appreciated.
Conceptually very powerful to be thoughtful of the difference of influence v control. This is exactly what is resonating about cybernetics and the idea that we take feedback loops seriously at the outset.
I read Phil Getz on the limit of batteries.
https://www.astralcodexten.com/p/notes-from-the-progress-studies-conference/comment/73937339?utm_source=substack&utm_medium=email
It had a lot of the anti renewable talking points that ads wrong or disingenuous - subsidies, the output of solar panels on cloudy days , and an argument that’s I didn’t get about solar being cheap because fossil fuel is available as a backup.
However Scott introduced it as a post about the limitations of batteries so let’s stick to that.
Phil said that the “ A lead-acid battery stores about 35 watt-hours per kg. So we'd need about 2,294 billion kg of lead for batteries. There are only about 95 billion kg of lead left on Earth that we know of … The situation with lithium is much worse”.
The problem there is that lead acid batteries aren’t used on the grid or are used very sparingly as an old technology (Chat GPT said not at all).
So even if Phil is right that there’s a similar or worse problem with lithium why even mention an old technology that is not being used. It’s like saying “the AI revolution is dead because there isn’t enough glass in the world to power vacuum tubes needed for GPUs, and silicon isn’t much better” - and then doing the workings out for vacuum tubes.
Afaict (not an expert) iron air batteries solve these issues (at the cost of being heavier and somewhat less efficient, which aren't blockers for grid storage), but most of what I can find on them is by pro-environmentalist groups who may be optimistic.
You're right - lead acid batteries are not the right tool for storing energy for the grid. I'm not entirely convinced by the maths (which is only sketched). But looking at energy densities for Li batteries, you get at least 100 times the energy storage as with lead-acid, seemingly with comparable global supplies of lithium as lead. It seems sodium ion batteries are in development, and obviously there's no issue with supply there.
The idea that you would want to store 24 hours energy supply with batteries is also a bit unrealistic. Our patterns of power use developed with a grid which supplies power at constant levels 24 hours a day. Going forward one would expect there to still be some conventional generation, both fossil fuel and nuclear, but with more variability in supply. Grids will need to be upgraded quite a bit, and we'll move increasingly in the direction of variable pricing to encourage consumers to use electricity when it's abundant and discourage use when it's not. Local power storage will play a part too: I'm looking at solar right now, with the understanding that I could be basically off-grid April to September.
There's an underlying point which is valid, but the argument used to get there doesn't hold up to scrutiny. We're not going to maintain existing usage patterns while building endless battery farms to move forward. But it's true that existing battery technology isn't up to what we will require.
The “cloudy day” argument was wrong anyway as solar panels do produce on cloudy days. However worrying about cloudiness isn’t really that important for large grids, there’s never going to be a day in the US which is full cloud cover across the country.
And while the panels can go anywhere, particularly if we are talking rooftops, the large solar plants will go in the places where cloud cover is low. Night is a different thing.
> Going forward one would expect there to still be some conventional generation, both fossil fuel and nuclear, but with more variability in supply.
Net zero, isn’t zero - there will be fossil fuels so it involves carbon extraction, which to most people isn’t clear. For that reason don’t worry about the doomster arguments that we will have runaway heating even if we get to net zero, because if net zero is possible, net negative is possible.
How bad would the transmission losses be if we went halfway round the Earth? It's always daytime somewhere. Is there any combination of costs where this might make sense?
>It's always daytime somewhere.
Agreed. It is an elegant idea (but I don't know the numbers - undersea massive power transmission sounds pricey...). Unfortunately, an additional problem is that transmission lines are intrinsically an extended, soft, hard-to-defend target, and the world has a lot of belligerent states and non-state actors...
It's always sunny somewhere, but does that mean you're building solar panels everywhere? The cost of energy from solar will definitely go up if you have to dig deep into the worst places to build solar panels just to have reliable production. I think the "unlikely but not so unlikely you can just ignore it scenario" is something like a week of unusually cloudy weather in the region that normally produces most of your solar power, leading to substantially less production. Solar still produces power on cloudy days but it's much less (google indicates 1/10 to 1/4 as much as usual). Maybe this sort of thing is much less likely than I'm giving it credit for as solar generation is more geographically dispersed, but I'm not sure "it's always sunny somewhere" is a sufficient answer.
It may be sunny everywhere, but it's not cost-feasible to have panels everywhere. So, yeah. Not only that, but the further you go from the equator, the less efficient those panels are going to be, unless you tilt them, which in turn means spacing them out of each other's shadows.
I feel like there *might* be a decent argument for putting arrays in the Sahara, presumably after paying Algeria, Libya, Mali, Chad, et al. a bargain price for land rights. I don't expect them to power the entire world, but they might put a big dent in the demand, and that might be enough.
I've long wondered what the practical hurdles actually are. I could imagine a few - transport, distance from supply hubs, distance from customers, political instability. But I don't have numbers for any of those.
> The cost of energy from solar will definitely go up if you have to dig deep into the worst places to build solar panels just to have reliable production.
We were specifically talking about the US in this sub thread, since the guy I was responding to was arguing against batteries in the US. I’m not a fan of Europe depending on the Sahara for its energy but wind power is a better bet for Europe anyway.
Solar in the US would generally, if positioned correctly, generate fairly reliable energy. During the day.
There is a company working on placing mirrors on satellites that will direct sunlight to solar plants during sunrise/sunset to extend their working time. So it could be "almost always sunny here in this one spot that we need it to be".
I would have to see some pretty serious math to suggest that it's possible to do this at a scale that would be meaningful without drastically increasing the cost of solar power.
Yeah no idea if it works or is economically feasible. Right now they've only done a proof of concept using a mirror on a balloon: https://www.reflectorbital.com
But there is a town located in a valley in switzerland (I think? norway?) that uses a mirror to reflect sun light over a mountain to give some relief from early sunsets in winter. No solar generation just vibes(?)
Why do you imagine energy *density* is the right metric for grid storage? The batteries are going to spend their entire service life sitting on the same concrete foundation; why does it matter what they *weigh*?
If a megawatt-hour of lead-acid battery storage weighs ten times as much as an MW-h of lithium-ion, but costs half as much, then lead-acid is the winning strategy.
I'm not actually sure if that's the case; the math gets tricky when we're talking about total life-cycle cost. And there are technologies other than lead-acid and lithium-ion to consider. But I see way too much hype about lithium-ion batteries with way too little explanation of why they're the right answer. Lithium-ion batteries usually *are* the right answer if you need to *carry* them somewhere, because of the energy density. But that doesn't apply here.
They were referring to whether it was possible to have sufficient battery storage with the resources viably available to us. If there’s too little lead (94/2.3k billion kg) then it doesn’t matter if it’s more economical per joule - we can’t do it. If lithium has lower total mass requirements due to energy density, then it’s possible to do at all (1/100 the required mass at the same viable material leads to 94 billion kg available and only needing 23).
>It seems sodium ion batteries are in development, and obviously there's no issue with supply there.
What is the oxidizer in this chemistry? I'm fond of sodium/sulfur, since _both_ elements are very abundant - but there are some problems with side reactions producing polysulfides. I'm not sure what the status of trying to avoid that is.
>we'll move increasingly in the direction of variable pricing to encourage consumers to use electricity when it's abundant and discourage use when it's not.
_Yetch_! That puts the burden of grid management on households instead of on large organizations which are far better able to absorb the management costs, both physical and informational.
I'm no expert on battery chemistry. I'm optimistic that smart people will figure it out.
As much power as you want when you want it is something we're used to because we don't see the costs it imposes. I'm suggesting that the grid management company needs to move from seeing its job as matching supply to whatever demand happens to exist, to shaping the demand to the supply.
It needn't be at the level of households. Although, e.g. my hot water is heated 2-4am, in a tank that's lagged to keep it warm for at least 24 hours. It wouldn't be a major inconvenience to me for the grid management to determine when this kicked in to better balance the load, for example. I'm not suggesting they tell me what time I can boil the kettle for tea.
Many Thanks!
>I'm suggesting that the grid management company needs to move from seeing its job as matching supply to whatever demand happens to exist, to shaping the demand to the supply.
This makes the power supply less useful to households. The degree of enshitification depends on how onerous the "shaping demand" is, but it _is_ enshitification.
I don't know about that. Ireland has a good number of power suppliers now, and each one offers multiple tariffs. It's up to the consumer to find the one that works best for them. E.g. I can choose something like a constant 30c per kWh charge or a 45c daytime but *free* power in the period 12am-4am. I'm told a good number fo people with solar take the later tariff, charge their batteries for free overnight (for the morning draw) and can reduce their power bill to a tiny fraction of mine. Is that enshitification, or just the free market?
Many Thanks! You have a good point.
Note, though, that
>I'm told a good number of people with solar take the later tariff, charge their batteries for free overnight
implies that those people are:
- paying the capital costs for solar
- paying the capital costs for their batteries
- maintaining their solar
- maintaining their batteries
- setting up the proper timing (seasonally dependent) for switching between charging and discharging their batteries, and switching back and forth between solar, charging from the grid, discharging into household consumption
- watching carefully if the tariffs change, reconfiguring the whole thing in response to tariff changes, and possibly being left with a "stranded asset" if the batteries stop being worthwhile.
Now, I could see an argument for distributed solar (though one of the comments in this post said that rooftop solar cost about 3X what a solar farm costs). Solar _has_ to occupy an area proportional to the power it supplies. In a built-up area with scarce land, you either put solar on rooftops or you transmit it from somewhere else and eat the resistive transmission losses.
For batteries, though, I don't think this makes sense. One can put the batteries in a cube, with a footprint scaling as power^(2/3), and you get all the economies of scale from having a _small_ number of people who are professionally skilled at the care and feeding of the battery cube, which can serve a huge number of customers. There are also economies of scale in constructing the batteries themselves. The electrodes just scale with power, but all of the packaging, power conditioning, monitoring, etc. has economies of scale.
> Yetch_! That puts the burden of grid management on households instead of on large organizations which are far better able to absorb the management costs, both physical and informational.
The only burden on households is to decide to charge the car, or do the washing, or even increase the heating or air conditioning when the electricity is cheaper. And no doubt there will be apps for that.
Many Thanks!
>The only burden on households is to decide to charge the car, or do the washing, or even increase the heating or air conditioning when the electricity is cheaper.
Those are not minor. A washing machine needs to be manually loaded and unloaded before and after its use. Constraining when it can be used constrains when the person using it has to be available for those tasks (there is _some_ flexibility if one pre-loads it - but leaving wet clothes sitting in the machine afterwards is a bad idea.)
>And no doubt there will be apps for that.
The app cannot load or unload the washer, nor ensure that the car is not needed for transportation at the time the power is cheap.
These burdens are indeed enshitification, as I wrote earlier.
Yes leaving wet clothes is a mistake but the idea is to load up before and the washer and dryer will start when on time. People will know when the cheap electricity is available a few days in advance. With solar that will be a lot of the day in summer, with wind it can be anytime day or night. There’s a lot of time to charge a car.
Nobody will be forced to do this, but many people will.
Many Thanks! To put it another way: If e.g. solar was the dominant energy source and electricity prices are set much higher at night, the situation approaches one of rolling blackouts, though not quite as severe. I'd rather have the utilities maintain battery farms, or use nuclear for baseline loads.
Even with traditional power generation we use power storage. Usually pumping water to high reservoirs then letting it flow through generators when more energy is needed.
That kind of storage isn't great at adapting quickly but allows you to even out peaks and valleys while a smaller amount of batteries can be used for unexpected high demand.
Any argument of the form "X wont happen because we will run out of Y natural resource" has proven wrong over and over again for the past 150 years. Oil, nitrogen, food, etc. Humanity always finds a way to overcome the limitation.
You've heard of the utility monster. Now get ready for... the Deontology Monster!
It crawled from the depths of hell, answering the call of a ritual which has entrusted upon it the summoner's moral code. It then proceeded to follow the code to the letter, but in a manner which wreaked havoc upon our world.
It put itself in situations where it could save many people, but only saved those who were destined to become murderers. It found people who could be aggravated into assaulting it and then killed them in self-defense. It taught the values of liberty and progress to members of fundamentalist households, selecting those most likely to get ostracized and harassed by their family. It initiated insurrections against evil tyrants and replaced them by benevolent rulers, but only ones whose successors were even worse than the previous dynasty.
Can a creature like this be called good in any meaningful sense? Is there any true deontologist, real or hypothetical, who can summon it safely? Find out this Halloween in - The Deontology Monster!
Could you be clearer about what specific deontological rules the monster is following in your examples? The only one that's clear is killing in self-defense; I don't understand the "rule" in any of the other cases. So evaluating this argument is rather difficult.
I admit this hangs together better than I thought. I want to ask if the creature is choosing on the basis of these consequences, or if it’s just really unlucky. If the former, lots of the moral rules I have seen described would say that otherwise-good acts done from evil motives lose some or all of their good character. But those rules are virtue-based. I’d be interested to see if any self-described deontologist writers thought people’s motives were relevant to moral rules.
I asked Claude Sonnet to riff on Unsong characters as various flavors of moral monsters inspired by your comment:
"The Comet King seems closest to a Deontology Monster, though with elements of the others. His core characteristic is his absolute commitment to duties and moral law, particularly the imperative to stop Hell's torture of souls. He embodies the terrifying logic of pure moral duty taken to its ultimate conclusion - he's willing to sacrifice his own happiness, his marriage to Robin, and potentially the welfare of living humans in service of what he sees as an absolute moral imperative to end Hell's torments.
What makes him particularly deontology-monster-like is that his actions follow from seemingly valid moral reasoning: If we have a duty to prevent torture and suffering where we can, and Hell represents infinite torture, then any finite sacrifice would be justified to stop it. His logic is cold but hard to refute within a deontological framework. Like a true Deontology Monster, he reveals how universal moral duties, taken absolutely seriously, can lead to seemingly monstrous conclusions.
Other characters that show "ethical monster" traits:
Uriel has elements of a Virtue Ethics Monster - his dedication to truth and order is so absolute that it becomes destructive. His virtues of precision, dedication to cosmic order, and commitment to mathematical truth are scaled up to angelic levels where they become almost incomprehensible and potentially catastrophic from a human perspective.
Thamiel could be seen as a twisted Utility Monster - he argues that he maximizes some form of "utility" by ensuring sinners receive their just punishment, though this is more of a perversion of utilitarian logic rather than a pure utility monster case.
The most interesting case might be Metatron, who in some ways combines aspects of all three monsters. His divine nature and connection to pure abstract good creates new moral duties (Deontology Monster), manifests perfect virtues that become alien to human understanding (Virtue Ethics Monster), and generates overwhelming divine utility that could theoretically outweigh human concerns (Utility Monster).
The broader theme in Unsong seems to be how transcendent beings - whether they're the Comet King, angels, or God himself - can embody moral principles taken to such extreme levels that they become almost incomprehensible or horrifying from a human perspective while still being internally consistent. This maps well onto the "monster" framework where ethical principles that work at human scales break down when pushed to absolute extremes."
The fact that any deontological rulebook that is smaller than the universe will eventually recommend something bad is something the deontologists presumably already know. This creature isn't a counterpart to a utility monster, it's a counterpart to the alignment problem.
There is a counterpart to utility monsters in deontological systems that actually get used: god. "What if there is a creature that enjoys so much that the greatest good becomes whatever it wants?" -> "What if something wrote all the rules, such that the right thing to do became whatever it wants?"
Ha ha ha, you think you can foil The Deontology Monster?
Matthew 10:
"21 Brother will deliver brother over to death, and the father his child, and children will rise against parents and have them put to death, 22 and you will be hated by all for my name's sake. But the one who endures to the end will be saved.
34 “Do not think that I have come to bring peace to the earth. I have not come to bring peace, but a sword. 35 For I have come to set a man against his father, and a daughter against her mother, and a daughter-in-law against her mother-in-law. 36 And a person's enemies will be those of his own household. 37 Whoever loves father or mother more than me is not worthy of me, and whoever loves son or daughter more than me is not worthy of me. 38 And whoever does not take his cross and follow me is not worthy of me. 39 Whoever finds his life will lose it, and whoever loses his life for my sake will find it."
Luke 12:
"49 “I came to cast fire on the earth, and would that it were already kindled! 50 I have a baptism to be baptized with, and how great is my distress until it is accomplished! 51 Do you think that I have come to give peace on earth? No, I tell you, but rather division. 52 For from now on in one house there will be five divided, three against two and two against three. 53 They will be divided, father against son and son against father, mother against daughter and daughter against mother, mother-in-law against her daughter-in-law and daughter-in-law against mother-in-law.”
Revelation 19:
"11 Then I saw heaven opened, and behold, a white horse! The one sitting on it is called Faithful and True, and in righteousness he judges and makes war. 12 His eyes are like a flame of fire, and on his head are many diadems, and he has a name written that no one knows but himself. 13 He is clothed in a robe dipped in blood, and the name by which he is called is The Word of God. 14 And the armies of heaven, arrayed in fine linen, white and pure, were following him on white horses. 15 From his mouth comes a sharp sword with which to strike down the nations, and he will rule them with a rod of iron. He will tread the winepress of the fury of the wrath of God the Almighty. 16 On his robe and on his thigh he has a name written, King of kings and Lord of lords.
17 Then I saw an angel standing in the sun, and with a loud voice he called to all the birds that fly directly overhead, “Come, gather for the great supper of God, 18 to eat the flesh of kings, the flesh of captains, the flesh of mighty men, the flesh of horses and their riders, and the flesh of all men, both free and slave, both small and great.”
Beware the wrath of the Lamb!
I'm not sure that really is analogous. The utility monster is supposed to demonstrate the perverseness of maximizing utility. The deontology monster presumably should demonstrate the perverseness of maximizing rule-following.
So it seems like a deontology monster would be more like, "it nobly goes out of its way to save people from would-be murderers, following the principle of stopping murderers being good, but only when it knows those people will themselves become would-be murderers that it can save more victims from. It is lethally effective in self-defense, but only when it knows its attacker is part of a large clan which will attempt to avenge that death on the deontology monster and 'forcing' it to defend itself more. It is diligent in working for reform and revolution against evil regimes, but only assisting those movements with poorly chosen policies or leaders which would eventually turn out to be even more evil and in need of reform/revolution." It carefully follows all its deontological rules like saving victims or killing only in self-defense or fighting oppressive regimes, but ensuring that it will follow those rules as much as possible in the future.
I think a deontologist would answer something like 'a perfect deontology with a perfect set of rules would not be exploitable by such a monster. Any human construction of the rules *is* exploitable, but *any* human-enumerated moral system is exploitable in some way for much the same reasons.'
The most famous form of deontology is Kantism, and Kant's fundamental principle is famously that "nothing in the world—indeed nothing even beyond the world—can possibly be conceived which could be called good without qualification except a good will." The Deontology Monster clearly doesn't have a good will, by your construction, because its will is to increase murder, assault, ostracism, harassment, and dictatorship.
> It put itself in situations where it could save many people, but only saved those who were destined to become murderers
If it can see the future, then the creature's threshold for murder suddenly expands massively. No ordinary person can say, "If I do X, then Y will incidentally murder Z, and therefore doing X is murdering Z" because we treat individuals as independent agents who can make whatever decisions they like, not as objects that can be triggered to act as easily as nerves control muscles and muscles control weapons. Additionally, for ordinary people, intention doesn't always match outcomes. If the monster has absolute prescience, then all actions leading to deaths become acts of murder. So no, it's not Good. In fact, I think it could be argued that basically any entity capable of seeing the future is (under deontology) extremely Evil.
How would you operationalize "Trump will be an authoritarian" or "Trump will undermine democracy" in a way suitable for a prediction market question?
There are way too many confounding factors to make these useful metrics.
+1 for this. Everything else seems difficult to define, like project 2025 - given the very obvious ideological slant of almost all institutions, far-reaching personal changes by themselves are quite reasonable and whether it's authoritarian depends entirely on whether the replacements actually are Trump/republican loyalists or just simply not democrat loyalists.
Is there a difference between Republican and Not Democrat? Are there any True Neutrals left in those circles?
That seems like way too narrow a metric. As a ludicrous example, if he ruled by fiat, suspended civil liberties, but appointed Don Jr to succeed him, that wouldn't count.
I'm really not sure this is operationalized enough for a bet to resolve cleanly. Suppose Trump spends 2027 going around saying "I want another term, the American people want me for another term, but the haters and losers say I'm disqualified on a technicality. I guess we'll see what happens." Then he never attempts to file his candidacy or take any official action related to running. I don't think this is terribly far fetched, and I don't see whether this counts as attempting to retain power.
>Conditional on him winning this election, will he attempt to retain power in 2028?
As an index of Trump's intentions, this is kind-of tricky given his age. Maybe, as a prediction market it should be conditional on both Trump winning the election _and_ on Trump being in reasonably good heath (more coherent than Biden is today) in 2028...
>That’s the best one I can think of. It’s quite likely he will appoint MAGA types to cabinet positions, move the federal bureaucracy significantly right, and centralise executive power.
Sigh. Given the power of the Federal bureaucracy, I'd rather see it moved right by having the heads of (cabinet level?) departments elected by the voters. Yeah, this would take a constitutional amendment, and we don't seem to do that any more... :-(
> 1000 women on trail for having an abortion over 4 years
Didn’t Trump say that he wasn’t going to do anything federal about abortion.
> Number of federal officials replaced that match the list in project 2025
Trump has said he had nothing to do with project 2025.
> 3rd term
He’s also denied that.
Why is the list in project 2025 especially authoritarian? Most of what I've skimmed through in the project 2025 doc is pretty vanilla republican stuff (some good, some bad). I'm sure there's some worse stuff I missed (it's like a thousand pages and I only skimmed), but the median point in it doesn't seem particularly authoritarian.
Having read the entire plan, for me it's the unitary-executive stuff that is most troubling: all federal agencies being placed under the POTUS's direct personal control, specifically including eliminating the guardrails which limit a president's ability to hire/fire the top officials in the Dept. of Justice. That would result in a POTUS being able to (just one example of many) order the head of the Census Bureau to have the census come out a certain way and just keep firing them until he found the one willing to do it. Combined with the recent SCOTUS ruling on presidential immunities, the POTUS would be literally "above the law" in a way that the US has never seen before and the Framers explicitly wanted to prevent.
Second most troubling for me, though I could be persuaded to move it up actually, is that as a secular person who does not think government should be carrying out any particular religion's worldview or precepts, Project2025 repeatedly laughs in my face. "Biblicly based" comes up a lot.
If you are strongly pro-choice, a whole long section of Project2025 is a complete nightmare. It tries to deny that it's taking steps towards a national ban but no sensible person would believe that (and the anti-abortion groups sure don't).
Internet neutrality would be gone, with tech companies that in any way allow the transmission of pornography [to be defined in a biblically-based manner] being summarily shut down on a "prove your innocence" basis.
Those are the lowlights for me personally, YMMV of course. For one counter example, while I disagree with the plan's approach to immigrants and immigration that's mostly within the bounds of policy disagreement. Though creating a new national superagency with plenary prosecutor+judge powers at all borders, not subject to legislative control, does reinforce my negative reaction to the unitary-executive stuff.
My Project2025 summary to my wife was, "These guys basically want us to have a Christian Bible-enforcing king", and not a figurehead constitutional monarch like the UK has. Whether that counts as fascist is unclear to me (political science definitions of that word are pretty squishy), but it certainly is authoritarian.
>Your best bet is to just choose some organization that rates nations for how democratic they are and bet on whether they'll pronounce him a dictator within 4 years. Not necessarily reflective of reality, but it's probably better than nothing.
Difficulty: All the organizations I'm aware of that do that are based in the US or benefit significantly from US money, making it very difficult for them to declare the US a dictatorship even if its true.
If anything, the opposite is the case. https://kottke.org/21/01/usa-downgraded-from-democracy-to-anocracy-part-democracy-and-part-dictatorship
If the US becomes a dictatorship then you'll know about it because all these orgs will suddenly start rating it a perfect democracy.
V-Dem is the most credible organisation I know that assesses authoritarianism.
This is probably the best option. Esp since it has a variety of different indices (Deliberative Democracy Index; Egalitarian Democracy Index; Electoral Democracy Index; Liberal Democracy Index; Participatory Democracy Index) and also a large number of subcomponents.
My only concern is that the indices dropped quite a bit in the first Trump term, so there might be some bias. The drop seemed to be a bit more than merited, at least at the eyeball level. Eg Deliberative Democracy Index was the highest ever in 2015, but in 2019 was the lowest since 1971. (But at least it didnt behave like that silly Polity score did) Perhaps averaging a number of similar indexes would result in something with less volatility? Or perhaps just choosing a low enough score as the threshold would be sufficient
So how do you rate the Democracy of "we're not going to give people the choice to choose who runs, we're telling them this is the candidate, and in some states we try to have a non-aligned candidate removed from the ballot, but trust us, we're the defenders of democracy"?
A one-party government is generally not considered very democratic, but in some American states that's the rule (Republicans shouldn't even bother running, the choice comes down to Democrat A or Democrat B). Trying to get candidates taken off the ballot because you fear people may exercise their choice to vote for them instead of Your Guy/Gal isn't very democratic, either.
If we're going to do "Is the USA sliding into becoming a non-democracy?" then apply it to Tweedledum as well as Tweedledee.
Instead of waging the culture war at every opportunity, why don't you go to the V-Dem website and look at their methodology? That might answer your question as to whether that would effect the ratings of a given country. I would start at pp 94-95 of their codebook (https://www.v-dem.net/static/website/img/refs/codebookv111.pdf) re the sub-component "Candidate selection-national/local". Assuming, that is, that you are actually interested in assisting Scott in his inquiry.
gd, popping up with "Is X the most evil person?" is partisan, whatever way you slice it. Maybe X is, but Y doesn't seem to be 100% spotless either, and if you're really worried about the possibility, then you should consider all relevant persons, not just "I don't like X so I'm saying X is the most evil".
How do I know this random website you recommend is neutral and not partisan?
EDIT: Okay, had a look at the code book. University of Gothenburg, so that gains points from me for being less likely to be partisan on American grounds.
Let's have a look at one of those measures, then:
"3.1.2.5 Elections multiparty (C) (v2elmulpar)
Project Manager(s): Staffan I. Lindberg
Additional versions: *_osp, *_ord, *_codelow, *_codehigh, *_sd, *_mean, *_nr
Question: Was this national election multiparty?
Responses:
0: No. No-party or single-party and there is no meaningful competition (includes situations where a few parties are legal but they are all de facto controlled by the dominant party).
1: Not really. No-party or single-party (defined as above) but multiple candidates from the same party and/or independents contest legislative seats or the presidency.
2: Constrained. At least one real opposition party is allowed to contest but competition is highly constrained — legally or informally.
3: Almost. Elections are multiparty in principle but either one main opposition party is prevented (de jure or de facto) from contesting, or conditions such as civil unrest (excluding natural disasters) prevent competition in a portion of the territory.
4: Yes. Elections are multiparty, even though a few marginal parties may not be permitted to contest (e.g. far-right/left extremist parties, anti-democratic religious or ethnic parties).
Scale: Ordinal, converted to interval by the measurement model.
Data release: 1-11.
Cross-coder aggregation: Bayesian item response theory measurement model (see V-Dem Methodology).
Date specific: Election-specific dates (v2eltype).
Citation: Pemstein et al. (2021, V-Dem Working Paper Series 2021:21); V-Dem Codebook (see suggested citation at the top of this document).
Years: 1789-2020"
Seems like "2 - Constrained" describes efforts to get other party candidates off the ballot, do you agree? Now remind me, who tried doing that?
https://www.politico.com/news/2024/03/04/states-cant-remove-trump-from-ballot-supreme-court-says-00144673
https://apnews.com/article/wisconsin-jill-stein-harris-trump-lawsuit-405e8bae8ff9becfa81a1360708d59a0
>Now remind me, who tried doing that?
Again, you seem to want to wage the culture war, in a thread about how to help Scott operationalize certain phenomena for the purposes of developing a prediction market question. Not cool.
Please try to engage in good faith. One of the huge strengths of ACX is the loose moderation coupled with a community that actually tries to communicate, and by defecting on the communication norms, in the long term you mostly just push towards more restrictive moderation.
>Republicans shouldn't even bother running, the choice comes down to Democrat A or Democrat B
This is a silly argument - it's not like anyone bans Republicans from running, those states are just far enough left that Republicans aren't popular. There are also states that are far enough right that Democrats don't really have a chance. There's no conspiracy here, it's just that the 50 states have a wide enough range of positions that you can't have a nation-wide party that aligns with all of them at once.
By your logic, Vermin Supreme is the most oppressed candidate of all time, because he's never had a realistic chance of winning, ever.
For faith in democracy, fewer votes per capita cast for the two leading candidates at the 2028 election than the 2024 one?
That isn't necessarily a bad sign. Two boring candidates with centrist policies and sensible advisors campaigning for the votes of a sleepy electorate is a better outcome than Nazis vs Communists with record turnout.
I agree that my suggestion isn’t perfect. As I saw it, if people feel democracy is threatened, they might be more likely to vote if they think it’s a necessary to disrupt whatever plans Trump might enact (e.g. to soften the impact of him changing eligibility rules to favour his own party). I think most of the people engaged enough to do that would be voting against him already anyway, but I’m not sure.
With that said, I think I’d probably see both parties choosing centrists as being more likely if both parties agreed Trump had gone too far. It resembles the fact Biden campaigned on being centrist, and the fact that Australia’s main party leaders after the Rudd-Gillard-Rudd-Abbott-Turnbull fracases have been more boring than the ones before that period. If Trump is unpopular for non-authoritarian reasons, I’d expect the Democrats to move left by a median voter argument, since their base will feel vindicated that the other side is bad.
Many Thanks! I'm not sure if I'm reading the sign of your metric correctly.
>As I saw it, if people feel democracy is threatened, they might be more likely to vote if they think it’s a necessary to disrupt whatever plans Trump might enact
So are you saying more turnout is an indicator of people feeling more threatened (perhaps re democracy, perhaps re other parameters)? That sounds plausible, and is analogous to the Nazis vs Communists situation in pre-WWII Germany.
( FWIW, I, personally, see both Trump (with Jan 6th) and Harris (with her proposal to "regulate" [censor] speech as being threats to a democracy with civil liberties. Regrettably, they can't both lose (barring some really improbable events). )
Not a perfect metric, but "will he be denounced as authoritarian by (specific, named) people currently on his side (e.g. the republican house speaker or "at least X members of his starting cabinet ") is an interesting one.
(Advantages: more flexible about catching directionally vague things. Disadvantages: can have both false negatives (he could go authoritarian and keep their loyalty) and negatives (he could lose popularity and get called out despite not being authoritarian).
I had an interesting back-and-forth with chat gpt about this, by asking it to build a score card for evalutating democracy/authoritarianism, then feeding it examples of eroded democracies and finally about which Trump's proposals align with stuff that happened in those eroded democracies.
My main takeaway for the most telling signs:
- Politicize the DOJ and the FBI
- Schedule F proposal to make it easy to replace civil servants
- Changing electoral laws unilaterally by invoking the so-called "2020 rigged election"
- Revoke public broadcast licenses
- Change media regulation to weaken the press
- Prosecute political opponents
- Military intervention in protests
Wanna bump this, it’s a pretty solid list of what authoritarianism could look like.
Yea it is.
The media pose an interesting question. Except for Fox, the media seems very close to acting as a Democratic Party organ. Amongst other problems, it is a miserably bad watchdog when watching Democratic politicians.
Any suggestions for balancing this? Require that the GOP/Dem ratio in the reporter pool be within X% of the ratio in the general population? Something else?
I think the curtailing of media metric applies mostly to "official" or "established" media, and its leaning is irrelevant, the idea is to show that an authoritarian regime doesn't accept criticism and oppositon. In other platforms, like social media, it's a tricky metric. Example: a lot of non-authoritarian regimes have some limits for what can be said in such platforms, at least more limits than in the US.
Many Thanks!
>I think the curtailing of media metric applies mostly to "official" or "established" media, and _its leaning is irrelevant_, the idea is to show that an authoritarian regime doesn't accept criticism and oppositon.
[emphasis added]
I disagree strongly. When a media channel leans strongly towards the party with power (as the New York Times does with the Biden/Harris administration), one can't _tell_ whether the government would accept criticism, because the media isn't delivering criticism. Now, right at the moment, the existence and toleration of Fox _is_ an indication of the current government's acceptance of criticism, but that network is largely alone in this.
>In other platforms, like social media, it's a tricky metric. Example: a lot of non-authoritarian regimes have some limits for what can be said in such platforms, at least more limits than in the US.
I agree that First Amendment protections (which I support) are stronger in the US than in most other nations. "Non-authoritarian" is a very fuzzy category. Great Britain has imprisoned people for politically incorrect comments on social media. I view that as evidence that they are sliding into authoritarianism.
Most consolidated democracies have hate speech laws, actually the US is a notable exception; Norway for example ranks 9.8 in the democracy index and has hate speech laws, whereas the US ranks 7.8 and doesn't.
Many Thanks!
>Most consolidated democracies have hate speech laws, actually the US is a notable exception
And I _STRONGLY_ want it to remain an exception, with First Amendment protection for free speech.
I can't speak for Norway, but in the UK, "hate speech" laws have been used to attack people who e.g. merely note that transpeople retain their original chromosomes, or who merely call to "reject white guilt". I consider those uses of "hate speech" laws to be grotesque and sliding into authoritarianism.
One of the amusing things has been seeing the New York Times denounced as a right-wing organ and maybe, who knows, harbouring Fascist sympathies itself for being insufficiently denunciatory of Trump and even covering the news in an even-handed manner.
How the turns table, indeed.
Have you seen the comments on Jeff Bezos' column explaining why he's directed the WaPo to stop endorsing candidates?
Many Thanks! That is bizarre... I get the NYT daily summary, and it is glaringly biased towards the Democratic Party. They pretty much phrased their campaign coverage from the point of view of "Let's look at this as an optimization problem in how can Trump be defeated" - as if that goal were a given.
He will invoke the insurrection act within 90 days of inauguration
He will buy a horse and then make it a senator.
Even worse, this will raise the average IQ in the senate. (Or at least the senate will average more horse sense.)
He'll regret it when the horse votes nay on all his proposals.
Whether Trump has fired 10-20% of federal bureaucrats by 2028.
It's what Curtis Yarvin wants him to do, which is a decent metric for how authoritarian someone is: (https://graymirror.substack.com/p/elon-versus-the-usg). I think it's also something Trump wants to do quite badly and it's what Project 2025 is about, which means it might actually happen. And I think it fits what came out of your Dictator Book Club; a hallmark of authoritarianism is destroying other power bases until only yours is left, whether that's a somewhat democratic Erdogan in Turkey wiping out the Kemalist military or Xi Jinping in authoritarian China using anti-corruption to wipe out rivals.
I'm not sure if this fits your frame though. I think a Trump who can fire significant amounts of the federal bureaucracy is pretty unavoidably more authoritarian but about half the country would see that as restoring democracy, not undermining it.
Also, more than happy to defer to someone more knowledgeable on the specific percent of the federal bureaucracy to fire or whether specific groups, like senior managers, are more important.
Julius Caesar, basically, although you'll find variations of this concept of mob democracy->tyrants/kings through Greek & Roman political thought.
The core concept is pretty much always the same. There's an entrenched, unpopular oligarchy. The general population is too uncoordinated to create/elect their own oligarchy so they give all their authority to a single individual, usually a successful general. That general then destroys the entrenched oligarchy but then has no rivals, there are no "check and balances" from other institutions because, well, those were all rotten and the people empowered a general to destroy them. Theoretically a truly great leader like Cincinnatus would then relinquish power and restore an oligarchic balance of power but in practice when guys like Sulla and Diocletian tried it, it failed.
Hey, man, Caesar isn't increasing centralized power like Diocletian here, building a huge bureaucracy, he's centralizing power like "Hey, we need to stop all these provincial governors and generals from murdering each other in brutal civil wars" a la Marius and Cinna and Sulla and Pompey and Caesar himself, plus all the Latin revolts. Constant, internal warfare between chief members of the government is, like, the primary feature of this period. The strong oligarchic government of the Gracchi period is in full disarray at this point.
If you don't understand this theory, awesome. If you do and you're just not convinced by it, cool, I'm not your guy.
It's because your model is lacking a little bit of nuance. Reducing government scope and size tends to mean reducing government scope and size *for the parts of the Government you think should be reduced*.
This is so sort of obvious that it's hard to see, and I feel like I'm insulting you by writing it but I don't mean it that way. The parts of whichever government processes that Trump and his associates like are unlikely to be reduced. But the parts they dislike are. This creates an unbalanced, adjusted power structure. Autocratic tendencies increase because though there may be abstract reduction in the size or scope or scale of "government", the individual component that represent a certain factional bias is likely to remain or even increase.
You can avoid this with a principled generalised stance of "reduce everything equally", but I think you will agree with me when I say that this is unlikely. I do not think that Trump (or Musk or whoever) have a principelled libertarian stance when they say they wish to cut the government excess. I think they have regulation they'd prefer not to have to deal with and so want to get rid of it.
To extend that example, consider something as relatively pedestrian as "Food inspection services". A relatively large component of government spending goes to managing food safety. You can therefore reduce a good amount of Government spending by reducing, dismissing or getting rid of those departments. But this is where we get into advanced political interactions that are hard to model out fully. As some utter toy numbers, not to be taken seriously, if there are 10000 Food Inspectors in the country and you cut headcount by 80% you are left with 2000. This sounds fantastic on paper if your goal is solely to "decrease the scale, scope and size of government".
If we start with 10000 Federal Agents and we end there as well but now we're running a 10000 Fed Agent / 2000 Food Inspector mix on the spectrum, the outcome of this will be total compliance with the mandate of "reduce government" but it will lead to a distinctly more autocratic process for the functioning of the government *that remains*. It might also, depending on size and scope and scale, have effects that are longer lasting. There could be some optimal amount of food safety inspectors per 1000 people, and by shifting the balance around, you then start having listeria outbreaks, e. coli, whatever. The government keeps trucking along, we have the same amount of guns / tanks / border patrol / telecommunications / whatever but the agency in charge of food inspection is now significantly compromised.
Again, toy model, but it illustrates the point. Then you have to write this out larger, to the fact that government is multipolar and very large. If you reduce IRS headcount and slash IRS funding you will get more tax fraud. This is unavoidable. There is a certain amount of tax fraud occuring every year, you need an unknown value of X IRS employees to detect and prosecute it, and those agents are now no longer available. We might get effeciency gains elswhere. I am positive we will. But there remains the basic fact that government mandates and taskforces and insitutitons need people to function.
Now we go back four steps.
You are a billionaire, and the Government ability to regulate food safety, the tax code, traffic regulations and let's throw in something like agricultural, whatever, is now significantly reduced because of massive budget reductions, downsizing and firing.
The private individual that is a billionaire remains a billionaire. A lot of people may benefit. Perhaps there is benefit to slashing the above departments. However, in a legal state operating by a lawful system, if the agencies involvedin regulation those law processes are knocked out of action, backlogged or unable to carry out a function, individual rule of individual free will and capability starts reigning supreme. We've veered over here to libertarianism. Which is a wonderful idea, and I support it fully, but it does not start here from first principles and a blank slate. The deck is already stacked - the McDonalds board of Directors no longer need fear surprise food inspections so they start slashing food safety requirements for their large corporation because the regulations that dictate what is or is not allowed are now not being updated and a corporation reasonably takes the steps the law mandates that it should, and here we are: the gradual push of individual incentive have now resulted in a world in which there is a somewhat increased chance your cheeseburger is going to make you sick, you have little recourse because you are not a billionaire, and we have transcended through "freedom from overzealous regulation" into "the individual free choice of individuals, outside of communal oversight, but mediated by the unofficial powerbrokers and movers and shakers that have accrued the means to shift the social balance".
There will then be push back and reorganisation and such-like. This is where it becomes useful to remember that quite often "regulation" or "zealous government agency", absolutely not an unthinking unalloyed good, still represents in the abstract and quite often in the specific the communally pooled ressources of the polity. Everyone gets together, pays XXP% in tax, and YYC% goes to Food Safety Inspection because the larger collective and commune decides that's a better idea for individual freedom. A million 1 buck donations is 1 million bucks, which offsets and checks the 1 million I inherited from my grandparents. That's how you get society.
There are politically reasonable theories that the natural state of man is freedom and that full freedom from meddling beaucratic middle managers will result in the exhultation of the human spirit.
There are also fairly decent counter-arguments that the historic trend, for all of human history, has been towards beaucratic middle managers, processes and laws being required to dictate, at times, the way people act to together to ensure some basic, underlying communal freedom. The shift away from feudalism is a shift towards a legal state where the various organs can meditate the power of strongmen.
So yes, decentralized power is by its nature more difficult to collect, and that's a fine truism. But the power here does not start decentralized. It's already been collected and pooled and adjusted. The playing board is not randomized. You can risk loosening the wolves by reducing the watchdogs. History - which you cite - is rife with examples of tension, horror and decay increasing because of the reduction of power in a centralized authority simply because without compliance from some central force, power becomes what people can project. A German Robber Baron rules the road by right of sword. A small town sheriff that ducks federal corruption investigation has pooled power already. Large multinational corporations that produce many a wonderful and fine good occassionally include ingredients in their food products that are deleterious to human health above certain values, and unless we want the "free market" to sort out the process of what food producer to risk buying from, we do need some basic rules.
Try not to forget here that if we veer in the other direction - towards freedom, reduction in government, lack of regulation, Do What You Can For There Are Less Laws, the friction that happens instead is the econonomic friction of the inefficiency processes of people dying. The free market will probably eventually sort out the issue of McDonalds Cheeseburgers making all their customers sick. Eventually, they'd have no more customers. This isn't a got'cha, it's just the basic logic of how markets work, but the path towards "have no customers" is paved with the millions of half poisoned burgers they, in this scenario, have sold on the way, half by intent, half by accident. That's not freedom. That's just endemic risk taking in every venture and a constant appeal to the hope that enough other people have suffered for the kinks to have been worked out. It doesn't even have to be malicious. Federal food safety standards as they are currently applied is ane evolving body of doctrine that comes from the hundreds of years that people have been killed by eating things. But you want to reduce the size of government. So you slash 80% of the FDA. And the IRS. And the DMV. And sure, Education too, why not.
So what's left, in this admittedly cooky example? Your personal millions, the soldiers you can buy with it and severely reduced pathways for various sorts of friction to be mediated by the consensus mechanism we call "society".
It works great if you already have all that power that never, ever starts as loosely distributed as the logic requires. And now some guy in Nevada is raising a militia and enforcing [the thing you don't like] on a large area. And we could deal with that, but we slashed 80% of the budget for all government services and now here we are.
In freedom, we found the old chains. History shows that centralized powers and processes tend to reduce the fractious friction of a far flung realm, least you invite in the horror that is individual warchiefs, bandleaders and strong men running the local area as how they see fit.
It won't be mad max, of course. I'm not talking "The apocalypse" here. Could just be the freedom of the free market to store your burgers at suboptimal temperatures because Food and Safety has 3 guys to inspect the entire country, and they'll never know, and if you get fined, you can just not pay and it'll never get enforced and . . . on and on.
It's not a simple linear relationship between "cut government" --> " avoid fascism". Rule by corporate forces and strong men is Fascism too, in the classical sense.
That's fair and well reasoned. I just happen to disagree quite a bit with your assessment though. You're conflating larger scale bits for smaller scale bits. Old man Benito in Italy broke up some monopolies and I'm with you there, but we call it Italian Fascism - or classical fascism! - because it's the classical model. It has a few more particularities than your dictionary definition. And it does in fact precisely lead us to a rule of corporate forces and strong men subsuming the larger will into autocratic, entrenched power interests.
They take a while to get there, I grant. There's a lot of "crushing external power blocks" on the way out there, because ideally you'd want the whole region under your control as a good autocrat and that's the thing I'm pointing out is the core of the issue. The struggle of a thousand sub-polities for scraps of power they can carve out because they have the possibility to do so looks a lot like anarchism and anarcho-tyranny but it might as well be marching bands of people in uniform and the rule of autocratic, despostic strongmen with particular preferences!
But that's all politial theory, and we could be here for hours slinging terms, so I'll call that I think we've probably just got different feels for what constitue the components of a fascist movement and leave it there.
But I will sidestep into your defintion, and raise a paralle point. You say that the left uses it so much that it glosses over what it actually means and boils the whole word down to "Icky" as some kind of unthinking reference to Bad Thing. Yeah, I'm with you that the leftist throw the word around a lot. I think the rightist do too. I also happen to think - and I'll doubletap this one to note it's just an opinion - that it is sometimes an accurate assessmet. By any particular definition, surely ". . . centralized autocracy or dictatorship, suppression of dissenting beliefs, and subordination of individual interests to the common good. . . " is the exact and compete political description of a platform that chants "lock them up", threatens to prosecute political opponents and fairly reliably threatens to arrest journalists, along with a few other parlances of the hour like kicking out the foreign immigrant hordes and promising that the nation comes first before all other things.
Is modern scale American Fascism going to look exactly like 20th Century Italy or Spain, or that other place, Georgia, that's the one? No. Of course not. But do I (as a random sample of a random observer) find it meaningfully accuate to describe stated policies of some political parties as "Fascistic"? Yes. Which is presumably why I do it, and I'm at least somewhat careful about the term. Maybe some hard core ravening leftists aren't (quite a lot of them aren't), but from my perspective the term gets thrown about because it fits.
So lets dance back to the larger point, and bear with me here, because it's a bit of a subtle one:
You might reasonably say that "nothing you/I described really points to an increasing centralized authority and subsuming the individual interest to the common good" in the examples above.
I say I disagree, because the process described is a general example of how such a thing happens over time. The people who would end up wanting to squish your interests for the "common good" or centralize authority are the people who step into the vacuum left by, in this hypothethical example, the government that we slashed by some double digit high end percentage. Or it's the natural process of what the government is, after it's been carried out. If, upon taking power, I replace civil servants with those bound to be me by personal loyalty, persecute my enemies, slash funding and components of large organs I disagree with, assume more direct power not balanced out by large organs I disagree with, and push whichever *personal* policies I might prefer, unopposed by a large civic substrate, well, there we are, "Fascistic" or "Authoritarian" are correct terms to use. This holds across all levels, from the random sheriff in some town pushing his weight around to presidents and pretenders elsewhere.
But the coy magic (And this is probably the roof of my real disagreement here, in a sort of generally opposed manner) is that I find your insistence on dictionary definitions of large political tendencies somewhat tendentious? Someone describes a system as "having some measure of ..." and the rejoinder then becomes a got'cha, ah ha, it does not fit the precise definition we are using in this case. We cannot use fascism because the definition I am using it to mean is not the definition you're using and that makes you the slightly snarker prankster.
There's *judo* going on here, and I think you'll know it enough to see it: You can decry "the leftists" for over using the term authoritarian or fascist, which rather than have a discussion about the political theories of the 20th century as it pertains to modern elections, instead becomes a way to turn the very invocation of a concept into a signifier that the people claiming something are to be disregarded for being somewhat hystronic. I think that's a bit too pat? Surely you'll have to at least *somewhat* agree that if I were to wander off now and spend 30 minutes collecting a bunch of excerpts of opinions expressed by leading political figures (democratic or republican, I shan't mind, though I have priors about which set is easier to find them from) then some of those excerpts will certainly have a *tinge* of the old Fascism. A bit of the Authoritarian creeping in. A gentle veneer of "an ultra nationalist form of government characterized by a centralized autocracy or dictatorship, suppression of dissenting beliefs, and subordination of individual interests to the common good".
Is it Fascist enough to institute an Italian style minimal wage and plans to enforce America-wide syndicate corporatism? Maybe not. Is it fascist enough to suggest we criminalize unions, strikes and labour movements? Possibly. Are we sometimes veering into a few little uncomfortable touches of moral purity projections, discussions about enforced gender roles, concerns about the corruption of the body politick at the hands of foreigners? Well, you tell me which group talks more about mass deportations, militarize camps and the need to squish individual elements for the need of the greater good. Well. Whether that's national stage or the governor of a state or the plutocrats of a small town, it's more or less the same fascism, and if you don't have enforcement of one style of government there will be alternatives and now we've gone all the way back around. You start slicing too far into the substate of "government" as we have it, and out the other end comes a lot of authoritarianism, fascism, tension and turmoil (And probably anarcho-tyranny too, and some communism communes and libertarian free holds and so on)
It's a complicated field, semantics and political development and societal opinions.
But I think you do a serious disservice to the notions of taking accusations of authoritarianism, or, heavens forfend, fascism, seriously simply because they're *also* sometimes mildly hysteric exaggerations of something where "Icky bad" would have sufficed. Sometimes they're true descriptions of real processes. But we hardly need the social aikido of subtly implying even the terms themselves have no applicability because they're washed out from overuse. Don't forget either - if things need to be leashed to the state or a movements idea of what's right, that's a lot easier if there's no governing overall structure that makes such compulsions harder. We're back to the fascist militia communes of upper Ohio marching in goosestep to enforce their will because somewher along the way, the processes we've spent two-hundred-years building to avoid all that happening got turn down.
You have to think at different scales for these things - otherwise the dictionary definition slide simply allows someone to gather power, crush popular dissent and centralize all power around their own set of preferred policies, but we can't call it XXXXXism because they're calling themselves National Socialists.
Anyway this all got a bit meandering, so I think we're done here.
Only if he replaces them or just if he fires them?
Firing a good 50% of Federal bureaucrats seems like a good week one project for DOGE.
You could predict any of these happening before the 2028 presidential elections.
a. Whether various Trump enemies go to prison.
b. Whether various Trump-hostile media outlets will be closed down sometime before the next presidential election.
c. Whether homosexuality or transgenderism will be treated as either a crime or as grounds for involuntary psychiatric treatment.
Maybe split it into pieces, like one could be 'political appointees in DOJ increase from x to y'? Another one could be DoJ drops cases against trump and starts cases on prominent democrats, could have w one about interference in midterm elections, constitutional crisis, etc.
Could you write a question related to any kind of constitutional crisis?
There's a clear SC ruling that gets intentionally circumvented or ignored, for instance. Similarly, if there's a refusal to leave office at the end of the term.
Looking back over the past eight years, was Biden worse for democracy than Trump? It's a tricky question. Trump was worse in some ways, Biden was worse in others. And the ways that Trump was worse were the ways you might predict him to be worse (e.g. attacking the press) while the ways that Biden were worse were the ways you might predict him to be worse (e.g. by co-opting the press).
If you ask the above question then you're going to get a list of things that Trump is likely to do. What you want to do is also ask the question (independently, to a distinct but otherwise-identical audience) "How would you operationalise 'Harris will be an authoritarian' or 'Harris will undermine democracy' in a way suitable for a prediction market question?"
And then what you want to do is to apply the Harris standards to Trump and the Trump standards to Harris.
>while the ways that Biden were worse were the ways you might predict him to be worse (e.g. by co-opting the press).
And using corporate intermediaries to censor voices in social media...
>And then what you want to do is to apply the Harris standards to Trump and the Trump standards to Harris.
Agreed!
One I would copy from None of the Above's list in https://www.astralcodexten.com/p/open-thread-353/comment/74431255 is
>a. Whether various Trump enemies go to prison.
Not an answer but here's the best article I've found about the process of undermining democracy -- how it's done, what the stages are. It's by a Turkish journalist, and is based on the course of Erdogan's rise to dictator. https://www.politico.com/news/magazine/2024/01/28/a-trump-dictatorship-is-possible-but-not-in-four-years-00137949
> in a way suitable for a prediction market question?
A meta comment - that Scott, who's no dummy and is into prediction markets, is asking for advice on this, is a sign of one the problems of prediction markets. Beyond asking who's going to win the World Series, coming up with a good prediction market question with good resolution criteria can be a Hard Problem, like writing good survey questions.
Thanks to Scott for the platform.
Two new posts this week:
Criticisms of a particularly bad NYT report on climate change (and that’s saying something!), “Climate Snark from the NYT” https://thomaslhutcheson.substack.com/p/climate-snark-from-the-nyt
And
Partial boost and partial criticism of a Scott Sumner observation on “Recession and Real Shocks,” https://thomaslhutcheson.substack.com/p/recession-and-real-shocks
Comments welcomed.
> Thanks to Scott for the platform.
Uhhh, Scott told you to dial it down:
https://www.astralcodexten.com/p/open-thread-352/comment/73463235
and
https://www.astralcodexten.com/p/open-thread-352/comment/73463225
How do you all time your covid vaccines, those who take it? It obviously isn't effective for a whole year.
This year I will just get it in November together with the flu shot for convenience.
Last year there was a point in time when the vaccine against a new variant came out. (September or so in Europe? I don't quite remember.) I haven't followed it this year, but I assume that this is now going to happen once or twice a year. The best time to get vaccinated is right after the switch, because then you have the smallest lag between the currently active variant and the variants that is the basis of your vaccine.
It's not effective after some months though. Can't figure out what that is.
I looked into this pretty carefully, and came away with the impression that resistance to becoming infected fades very fast, but protection from severe illness fades quite slowly.. There is still some protection even a year afterwards. Covid hospitalization rates have dropped hugely, and the main reason isn't that there's not much covid around, but that almost everyone has had the virus, had the vax, or both. That should tell you something about how long-lasting the protection of exposure to covid is, whether it occurs by infection or by injection.
Thanks. I'm not concerned about a severe infection but what about long covid? Increased risks of heart attack etc if you catch covid multiple times? That's why I want to avoid it if I can.
Last week, echoing Scott's popular "Beware The Man of One Study" essay, I posted a discussion of new data questioning whether light drinking has any health benefits.
Thanks to a few poorly designed studies, plus a 60 Minutes special from 1991, we've inherited a meme that light drinking (especially if it's red wine) improves cardiovascular health and increases longevity. The new data suggests a different and more subtle narrative.
https://statisfied.substack.com/p/how-much-alcohol-should-you-drink
I read https://www.nytimes.com/2024/10/23/science/puberty-blockers-olson-kennedy.html (https://archive.ph/plv81) and this seems like some serious scientific malpractice to me.
EDIT: while this still looks like scientific malpractice, there are some problems with the NYT article. See comments below.
As I understand it:
A multimillion dollar federally funded project researching the effects of puberty blockers on children's mental health didn't find any positive effect and the lead researcher is deciding that she therefore is sitting on the data because she doesn't want people to use her study as evidence that puberty blockers shouldn't be used. She did say she will publish the data at some point.
She's claiming that the 95 participants in her study just had really good mental health to begin with and so they didn't have a lot of room to improve. This is contradicted by her earlier report that 'one quarter of the group reported depression symptoms and significant anxiety, and one quarter reported ever having thoughts of suicide.' When asked about this contradiction she said she is 'still analyzing the full data set.'
It is possible that the puberty blockers prevented a decline in mental health. If, as it seems, there was no control group, it's impossible to say. People who got HRT after no puberty blockers are doing worse than the puberty blockers group.
What worries me most is that Olson-Kennedy asks us to trust her clinical experience instead of the data she doesn't want to publish. If her self reported clinical experience was good enough evidence, then the NIH wouldn't have funded her study.
I am surprised that she is allowed to wait to publish, especially since her study was funded by the NIH. I would have assumed that more follow through would be required. Perhaps it is but the New York Times article didn't make that clear?
No control group means it was a bad study, but the worst of all is the propaganda around "would you prefer a live daughter or a dead son?" to persuade parents into taking the puberty blocker then transition bus. If the researcher is wedded to the idea "dead cis kid or live trans kid", then her beliefs mean she must repress her own work lest the narrative be questioned and all the kids not on puberty blockers will then kill themselves.
Except that lots of studies with a control group have been done, and find a benefit of treatment.
Can you cite some? For puberty blockers for children? The Cass Report couldn’t seem to find any quality ones.
Another commenter links to a summary, which has links to several such studies:
https://www.erininthemorning.com/p/fact-check-new-york-times-publishes
This seems to be a good one with large N size and a proper control group:
>https://www.sciencedirect.com/science/article/abs/pii/S1054139X23005608
The article you picked out seems to be the only study linked to in that substack post that actually deals with studying the effects of puberty blockers for children. I'll take a look at it, but one study does not "lots of studies" make. The only other studies she cites are about kids who have not yet taken puberty blockers, how puberty blockers affect height, how effective puberty blockers are, and the study Chen did on the effectiveness of hormone therapy where they abandoned the hypothesis and method of analysis from their pre-registered version, which is indicative of p-hacking and cherry picking (in the pre-registered version they were going to study 8 different variables tracking mental health, in the final study 6 of those variables are gone and three brand new ones are there. Seems like a open and shut case of cherry picking when the study didn't give you the data you were hoping for).
But thanks for the one study though, I'll be sure to read it thoroughly. Unless you have more I recommend you stop telling people that there are "lots of studies with a control group" that "find a benefit of treatment" with puberty blockers.
https://www.erininthemorning.com/p/fact-check-new-york-times-publishes this may also be a worthwhile read
Thanks for sharing this. It definitely gave additional perspective on the New York Times article and raised lots of questions.
The Substack writer Erin Reed who criticized the New York Times article is covering this from the perspective of a trans rights activist, and there seems to be bias here.
However, Reed does raise some good questions:
> Twenty-eight studies have been published from the Trans Youth Care (TYC) Network study, including from patients on puberty blockers.
So while researcher Dr. Olson-Kennedy has said there is some data still being withheld, some data has been released, and the NYT article doesn’t make this clear. It seems like the papers which have been published are the ones which show results supporting the use of hormones in treating children with gender dysphoria.
> What Ghorayshi fails to mention is that among those who didn’t receive blockers, suicidal ideation was much higher—66%, more than twice as high.
So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers. This would’ve been good for the New York Times article to mention.
I am still wondering why there was no control group. One important comparison we need to see is between people who got puberty blockers and people who didn’t use hormonal intervention.
Ben Ryan claims that Erin mischaracterizes the studies:
https://x.com/benryanwriter/status/1850343387679137899
Also, in the next tweet down:
"inexplicably, this study did not find that the various types of laws restricting trans youth were tied to considering suicide, only attempting suicide. [...] Also, the increased reported suicide attempts were not tied to the first or third year after the passage of these laws, only the second year after the passage."
> So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers. This would’ve been good for the New York Times article to mention.
Unless the study initially listed suicidal ideation as an endpoint, this is likely just p-hacking.
Just going to link a classic:
https://slatestarcodex.com/2014/12/13/debunked-and-well-refuted/
P-hacking exists, but if you don't have a specific analysis showing that it happened in a given case, just looking at any study that disagrees with your intuition and saying 'probably p-hacking' is not adding to the conversation.
If you have a more specific analysis to justify this claim I would be happy to hear it, maybe my accusation is unfair, but it's impossible to tell from your one sentence post.
This isn't "a study that disagrees with my intuition" though, it's simply not a study.
That's _if_ the study didn't initially list suicidal ideation as an endpoint, as I said. I'm confused about exactly which study is being referred to by this sentence.
True that it's not referring to a study directly; it's comparing the results of the unpublished study in the article, to data about a natural control group from another study by the same group:
>A total of 95 youth were enrolled in the GnRHa cohort. Mean age was 11.22 years (SD=1.46), and the majority were white (52.6%) and designated male at birth (51.6%). Elevated depression symptoms were endorsed by 28.6% of GnRHa cohort youth, and 22.1% endorsed clinically significant anxiety. About a quarter (23.6%) endorsed lifetime suicidal ideation, with 7.9% reporting a past suicide attempt. A total of 316 youth were enrolled in the GAH cohort. Mean age was 16.0 years (SD=1.88), and the majority were white (62%) and designated female at birth (64.9%). Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt. Life satisfaction was lower amongst both cohorts compared to population-based norms.
So it's not p hacking because there's no t-test.
It's just looking at a natural control group and saying 'with puberty blockers depression/anxiety/ideation were around 28/22/24%, without puberty blockers rates were around 51/57/66%.
We can't do a t-test on those sets of numbers because we are just getting summary statistics rather than the data set, BUT given that it's a 30-40% difference on ALL measures, it seems likely to be significant and unlikely to be p-hacking.
>>> What Ghorayshi fails to mention is that among those who didn’t receive blockers, suicidal ideation was much higher—66%, more than twice as high.
>So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers.
No, that's not true at all. Those numbers are looking at rates of suicidal ideation (and other things) *before* treatment, and that study doesn't report any results *after* treatment. So the study says that "before" HRT occurred suicidal ideation was 66% for the group that was going to get HRT but no blockers. Which tells us nothing whatsoever about whether people who get puberty blockers do better than people who do not. The study doesn't even claim to tell us that.
I think maybe you are confused? Or else I am.
People who are about to get HRT at 18, but didn't get any treatment before that, *are* people where puberty blockers were potentially indicated at 12 but who didn't get them.
That is the natural control group for people who *did* get puberty puberty blockers at 12, and are now 18 (or w/e ages are used in the studies).
There’s an argument for that, but it seems like it was being interpreted as “we gave this group puberty bickers and their suicidal ideation is 22%, and we didn’t give this group puberty blockers and theirs is 66%”. I think the OP at least may be viewing it that way, and they wrote “So in the study, people who got puberty blockers are doing better than people who did HRT and not puberty blockers”, which simply isn’t true. It’s possible that the puberty blocker group will end up doing better than the HRT alone group, but that’s the very data that she decided not to publish!
Ok, so both of those options are wrong, let me see if I can pose it simply. Here is the relevant text fro the study:
>A total of 95 youth were enrolled in the GnRHa cohort. Mean age was 11.22 years (SD=1.46), and the majority were white (52.6%) and designated male at birth (51.6%). Elevated depression symptoms were endorsed by 28.6% of GnRHa cohort youth, and 22.1% endorsed clinically significant anxiety. About a quarter (23.6%) endorsed lifetime suicidal ideation, with 7.9% reporting a past suicide attempt. A total of 316 youth were enrolled in the GAH cohort. Mean age was 16.0 years (SD=1.88), and the majority were white (62%) and designated female at birth (64.9%). Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt. Life satisfaction was lower amongst both cohorts compared to population-based norms.
The experimental measure we are interested is self-identified trans kids who got puberty blockers from 12-16 vs self-identified trans kids who did not get puberty blockers from 12-16.
This study indicates that self-identified trans kids who did not get puberty blockers from 12-16 (ie people who are seeking HRT at 16 as their first medical intervention) have very elevated levels of depression, anxiety, and suicidality.
The researcher in question claims that there was no before->after effect of puberty blockers because levels of those traits were about the same before and after treatment, meaning at 16 their levels of depression and anxiety would be 30-40% lower than the no-treatment control group of 16 year olds.
(and note, the 19%/26% for the puberty blocker group is within 5% of the population average for people that age, AFAICT. They're basically 'normal' on these measures)
What happens before->after HRT is completely irrelevant to the study in quesiton, which is only about puberty blockers. We're only citing this study on HRT because the 'before HRT' cohort is a naturally-occurring control group for the 'after puberty blockers' cohort.
I think the reason there's no control group is that this isn't actually a researcher, it's a doctor at a clinic. Basically, this 'researcher' couldn't get a control group of people who wanted but did not get treatment, because their only contact is with people who got treatment at their clinic.
Grant money is available in many cases to physicians who 'do research' by providing care normally and measuring some things while they do it so we can get metrics on those treatments. But the primary motivation of the doctor is generally to subsidize their clinic and provide more care to patients.
Arguably the grant-giving agency should not have approved this grant without some way of getting a control group to compare to. Although arguably they had no way to know that the study would flounder on ceiling effects, because they had no way to know that most kids coming in for puberty blockers at that age are not yet doing very poorly on emotional health assessments.
It could easily have turned out that those kids are already struggling, in which case we would have had a perfectly good study about whether or not the treatment improved their metrics from there.
Basically this just feels like a case where someone tried to get grant money to support what they were already doing, it was a more-or-less reasonable grant to make, unfortunately it turned out that the experimental design didn't work due to ceiling effects that might not have been predictable ahead of time, and so it was file-drawered as bad methodology leading to a null result.
Not great, but extremely normal. The only notable thing being that it touched on a culture-war issue, so a bunch of people decided they could profit by applying isolated demands for rigor to it.
I don't think that "y'all should have a control group" is an isolated demand for rigour, it sounds like a routine demand for rigour.
That's not what the article or OP's post is about, though.
The accusation is 'you should have published your null result, the fact that you didn't publish your null result is evidence of politics-driven malfeasance'
And that's an isolated demand for rigor, because null results are rarely ever published, in any field.
The thing is that the doctor in question explicitly told the New York Times that she doesn't want to publish the null result so that it isn't 'weaponized'. That's the evidence of politics-driven malfeasance, I'm not just inferring that she didn't publish the data because of politics.
My post is about several things, including pointing out that there should have been a control and that she's withholding her data for political reasons and that she appeals to her 'clinical experience' which obviously isn't good enough given that people saw fit to fund her study.
Yes, but most researchers that find nothing don't then do an interview with a newspaper about "the study found no effect but I won't publish - not because it found no effect, but because I'm afraid of political backlash".
If these studies are being used to prop up "this is medically necessary treatment, without it children will kill themselves, do you want a dead son or a live daughter?", then a finding of "intervention did not make a difference" *is* an important result.
If it's "intervention didn't make a difference here, but other intervention did make a difference there", that's also an important result. "Suicidality is not correlated with gender dysphoria, because the patients presented with a complex of psychological ailments" is also important.
But saying "I'm not going to publish a result because I'm clinging to my article of faith and I'm afraid the heretics would use it to discredit my religion" isn't science.
> 'one quarter of the group reported depression symptoms and significant anxiety, and one quarter reported ever having thoughts of suicide.'
A few things to note here:
1. ">Data from the recently released Teen National Health Interview Survey from July 2021 to December 2022, reveals that 21% of adolescents ages 12-17 report experiencing symptoms of anxiety in the past two weeks, and 17% said they had symptoms of depression."
The kids coming in before treatment were very close to the population average on depression and anxiety measures already. So if after treatment they were still the same as teh general population, you'd have ceiling effect and a null result. You could only get a result here if puberty blockers made them *significantly better* than the general population for their age, which is asking a lot.
As for the metric of 'has ever had suicidal thoughts,' obviously it's impossible for any treatment to improve that metric (unless it involves time travel).
2. The same linked study gives rates for 16 year olds who never got puberty blockers and are now pursuing HRT:
>Elevated depression symptoms were endorsed by 51.3% of the GAH cohort, and 57.3% endorsed clinically significant anxiety. Two-thirds (66.6%) endorsed lifetime suicidal ideation, with 24.6% reporting a past suicide attempt.
Basically, this is the correct control group for the study at hand - self-identified trans kids who got puberty blockers from 12-16 vs self-identified trans kids who did not get puberty blockers from 12-16.
If it's true that the kids in this study had the same mental health ratings before and after puberty blockers, then this indicates a huge improvement relative to this control group.
Overall: the thing is, this is a negative result; there was no effect of treatment, nothing happened. Negative results are rarely published; this gets called the 'file-drawer-problem' and there are reasons to dislike it (though I think they're over-stated because of how we design our studies), but it is 100% common and near-universal, not some crazy thing that only this person has done.
The only surprising thing about not publishing a negative result is that anyone noticed and cared about this one, and the culture-war reasons for anyone caring in this case are pretty obvious.
Add to this that there are already plenty of studies like this with actual control groups that show a benefit for treatment, and the story feels pretty clear-cut. One study produced no results, it was a culture war issue where the non-result would obviously be misrepresented by culture-war outlets eager to make hay, many other better studies on the topic are already available in the literature, so they decided to just move on.
In a perfect world where everyone pre-registered results this type of thing wouldn't happen. But in our world it happens all the time, and focusing on it for culture war reasons is an isolated demand for rigor.
And, really: This is NY Times writing a hit piece on someone related to the culture war. Given what we've seen from them here in this community, do you expect them to be fair and objective?
Yeah I agree the NY Times article is not great. Erin's 'fact check' was also full of twisty words and culture war tactics. I've said above why I don't think this was just a routine 'let's not publish this null result'.
I just wanted to respond to:
> If it's true that the kids in this study had the same mental health ratings before and after puberty blockers, then this indicates a huge improvement relative to this control group.
This could be entirely right. However, if we imagine that only the 50% most gender dysphoric group of 11-16 year olds who do not receive puberty blockers go on to seek HRT, and more extreme gender dysphoria is correlated with higher rates of suicidal ideation and suicide attempts, then this could explain the higher rates in the 'control' group.
So we don't *know* that these groups can be compared, unless we know that nearly all the kids who received puberty blockers were highly likely to seek HRT later in life without receiving those blockers.
It's currently fashionable on Substack to post your life story as a series of bullet points. E.g…
At 20, I dropped out of university.
At 24, my boyfriend dumped me.
At 27, I was treated for depression.
They are overwhelmingly about failure and/or very depressing. How come no one writes happy life stories? Is it shame? Embarrassment? Respect for the sad people? I just wrote my story and it was mostly about happiness and success. My wife chastised me for showing off.
Why is it acceptable to post about sadness and failure but not happiness and success?
I think its because when a stranger does better than you, you will feel worse, and when a stranger does worse than you, you feel a bit better about yourself. Therefore, writing about happy life stories increases bad feelings in the readers and consequentally, it's discouraged. In my opinion context matters here, its much easier to be happy about a friend's success than a stranger's, so imo one should tell the happy stories to one's friends and not to strangers on a public platform.
I think you are right about not making readers feel worse about themselves. Now that you mention it, my wife gets upset when a friend does well if she is not also doing well.
I am always very reluctant to share stories of success with friends. I can tell sad stories to them though. I have two blogs now and about 90% of my followers here are strangers but it is the reverse over on WordPress. I'm very reluctant to tell person stories of misery here while I tell them all the time on my WordPress blog.
It's possible to tell many stories from the same events. A few years ago I made the explicit choice to focus on the generative, dynamic, "what would I be pleased to see someone writing in an obituary" presentation, rather than a "woe be me all these things went wrong and my enemies were out to get me and I should be pitied" framing, and wrote a life summary like that. Using the first kind of framing made a big difference to how I felt about myself. Others seem to find motivation in the second kind of story, using anger at injustice to fuel activism.
I think we in N America, Europe, Au/NZ have allowed incentives to creep in to prefer the second kind of story in public. I think this is damaging. Both kinds are valuable. This extends to the stories we tell of our collective memory. The Progress Studies conference seems to be an explicit attempt to use the first kind of story to talk about our history, whereas degrowth, imminent climate catastrophe, prepping for the apocalypse, and regressing to an imagined version of 1950s Ireland all seem to be examples of the second.
Framing is important. I prefer the framing of sad stories as 'I survived' or 'I learned something'.
I think it's bad too for young people to hear sad stories from their elders. If they read about striving and success, I expect they are more likely to aim for it. If their expectation is that the future is all failure and depression, they might start life with that expectation for themselves.
So how can we get zoomers to pay attention to biographies of people who were not defeated by adversity? Does this material now have to become podcasts or videos? Do we need to petition Kate Beaton to bring back Hark! A Vagrant, pretty please with Jane Austen/Sojourner Truth sprinkles on top?
The Art of Manliness had a post a couple of years ago about how young men used to expect a time of adventure in their twenties and gives the examples of Earnest Hemingway, Steve McQueen, Sean Connery etc going off to join the navy or a Gold Rush or to fight in a civil war.
Since the 1990s, the expectation is that young people will go to college, get a job and buy a house. Those adventures probably had more than their share of sadness — but they were still adventures. Perhaps younger people no long have that mix of good and bad to look on. Now it’s all boring and bad.
https://www.artofmanliness.com/character/the-least-interesting-generation/
It's possible (as Tolstoy implied) that unhappiness simply makes better stories.
The yellow press thrives on stories of gory transgressions. True crime is wildly popular. I understand that American daytime talk shows used to do brisk business with convoluted family drama. Even people who are quite successful tend to dwell in their memoirs on the challenges, not the intervals of smooth sailing.
I expect there is a lot of truth in that. I'd guess though that most people want to know about the challenges that were overcome leading to success.
From The Hobbit:
"Now it is a strange thing, but things that are good to have and days that are good to spend are soon told about, and not much to listen to; while things that are uncomfortable, palpitating, and even gruesome, may make a good tale, and take a deal of telling anyway."
I wonder if that trope comes from self-help books? Those I read back in the day did like to punctuate the text with little anecdotes about "Bob was a successful businessman. But at age 40, Bob's marriage failed. By 45, he had lost his business. At 50, he was a homeless wino sleeping in the gutter. But by following the techniques here, by 55 Bob had cleaned up, stopped drinking, started a new and even more successful business, and got married to a former Miss Cornpone twenty years younger than him!"
Just that today people prefer to leave out the "And at 30 I turned it all around" part? Or maybe they're not 30 yet, so there's still time for the turning it all around.
Let's hope they figure out how to turn it around!
I wonder if there is an expectation that life will be miserable anyway — because of all the misery stories that they've read online.
Where on Substack is this happening?
I don't know if there is a specific place for these but I am seeing maybe a couple per day — mostly from women substackers.
Here's mine, FWIW:
https://raggedclown.substack.com/p/my-life-so-far
In my experience, it's been easier to relate to failure than success. Plus, usually, they allow for a better future.
When I worked in a bookstore, a woman gave me a Russian currency note for no particular reason, after telling me she had just returned from there.
Another time, when I was driving a bus, an 8-year-old boy gave me a stamp with a fingerprint on it. 50 years on, I keep them in a glass box on my dresser.
It has a 10 on it, and a profile of a dude with a goatee. Lenin? She gave it to me in the early 1990s.
This sounds like the first few lines of an intriguing short story.
Describing eight years of disease wouldn't match it.
A narrative of success is about cause and effect, and needs to be written in full paragraphs. Because I did X, I had the opportunity to do Y, which gave me the opportunity to do Z.
A narrative of failure feels like a series of disconnected bullet points. This happened. Then this happened. Then this happened.
People who see their lives as a disconnected set of bullet points that just happen to them are likely to be failures.
> How come no one writes happy life stories?
Sob stories get more engagement.
If the US is considered to have budget deficit problems- how does Italy manage? How has it not collapsed? US debt to GDP is 124%, but Italian debt to GDP is 135%, and Italy is obviously not the world's largest economy or in possession of the world's reserve currency. 10 year bond rates for the US are in the low 4's, for Italy they appear to be in the mid-3's. I do understand that interest rates are lower in Europe at the moment, but these rates are not indicative of a bond market that's very troubled by what's going on in Italy. If huge deficits are such a gigantic problem- why haven't the bond vigilantes come for Italy?
I don't know much about economics, but:
Japanese debt to GDP is 263%.
10 year bond rates for Japan are about 1%.
Interesting comparison. Japan's economy hasn't exactly been good though, since the 90s. After an amazing 10-15 years, the Japanese economy pretty much just stagnated. Long term the outlook is grim, as the population ages and fewer people are born. Fewer people will be supporting more elderly, with either more debt or a significant reduction in benefits. More debt when the ratio is already that bad may not be possible.
Italy is bound to the Euro, so it has a lot fewer options than the United States in dealing with the debt.
In theory the US *can’t* default. All our debt is denominated in USD, so there’s always the option of printing your way back to almost-collapse with high, but not catastrophically high inflation. Italy, as part of the EU, particularly with German central bank philosophy being the guiding principle, can not. They have the advantage of northern neighbors potentially bailing them out, but considering most of Europe has its own economic woes, it’s uncertain how far the fiscally responsible north would go to bail out the improvident south.
In truth, no one knows what debt to GDP is unsustainable. Modern Monetary Theory will tell you it only depends on the interest payments, not the level of debt. Austrian Economists will be unhappy with almost any level of government debt that isn’t easily paid back on demand. It’s undeniable that 124% is less sustainable than 100% though, so whatever the real limit is, a higher debt to GDP is closer to that limit.
I agree with everything you've written- my point is, why isn't Italy's bond yield higher? Why isn't the market demanding more of a premium to lend to an even more indebted and fiscally weaker country than the US? Your 1st two paragraphs are reasons for a higher yield
That's a more complicated question with no single answer.
It's a combination of lower economic growth expectations in Italy, currency control preventing free movement of money to other markets, investment requirements (particularly for institutional investors/banks), and ECB policy.
Personally I'd point to the ECB policy as the primary factor. Italy's debt is supported by the ECB and their interest rates are kept artificially low, as the central bank buys many of the bonds that the government puts up for sale. The bond markets of national currencies are anything but free, and the intuitions of markets that would dictate higher vs. lower prices aren't always at play when it comes to national debts, especially those in capital-rich countries like Italy.
If economic growth is predicted to be basically flat in Italy (as it's been since 2008), a 3% bond is more attractive to purchasers when compared to the alternative in the domestic market. In the US, with consistent GDP growth, bonds have to compete with an investment in the market, which drives up bond prices compared to Italy.
At the end of the day, Italian banks and individuals prefer to purchase Italian securities, as there's lower regulatory and political risk. International markets are hardly efficient, and it's a better comparison with Germany (within the Eurozone) which has lower bond yields and is seen as a more secure borrower.
It would take some serious analysis by someone much smarter and more dedicated than I to sort through all the positive and negative influences on bond yields accurately. All us laymen can do is identify at all the different influences, and see what they sum up to, not what they specifically are.
Different countries are hard to compare on such matters. For example, Japan has a much higher debt, but that is not consider problematic by many economists since almost all its debt is held by domestic residents.
Italy's interest payments as a % of the GDP are quite manageable and in fact follow a long-term downwards trend https://data.worldbank.org/indicator/GC.XPN.INTP.ZS?locations=IT
Also, most of the debt is held by Italian companies and individuals. It's their savings, pension funds, etc.
Nobody explains these matters better than Cullen Roche, see, e.g., https://disciplinefunds.com/2024/10/22/we-need-to-have-a-talk-about-bond-vigilantes/
Edit: for an in-depth treatment, see his paper on modern monetary systems: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1905625
Another thing that makes this all the more mysterious is the flatness of Italy's GDP curve over the past decade and a half, while the US has seen considerable growth. One would think that this would mean that the US has considerably greater potential to outgrow its debt.
This feels like a fat man pointing at an even fatter man and saying "If obesity is so unhealthy then how come that guy is not dead?"
Things can be bad and yet medium-term survivable.
Two answers.
1. The US is not considered to have especially bad deficit problems by economic experts. This is a much bigger issue politically than it is among technocrats.
2. The US dollar is the default store of value around the globe, and it gives us a uniquely privileged position in terms of borrowing and loaning money. Very few countries can get away with the type of financial shenanigans we can.
The question was about Italy, not the US
The US has a long history of alternate elector slates showing up due to disagreements over the minutia of State election laws. I believe the most recent was from Hawaii in 1960 (Kennedy v Nixon), but as far back as 1876 a second set of Electoral Votes from a State made it as far as Congress.
"Cars have a long history of occasional brake issues, so from a certain perspective this is just normal car problems we ought not make a big deal about."
- Some guy who you just caught cutting the brakes on your car, 2020
A very interesting recent interview with Mike Levin about expanding our definitional boundaries of concepts like "intelligence". https://m.youtube.com/watch?v=6w5xr8BYV8M
One indirect observation that he made (in the context of cancer cells going rogue) was that "selfishness" can be described of as an individual simply having an overly-narrow self-concept. Whereas, if you think of "you" as being a part of a larger collective whole, then being "selfish" means doing things which benefit that collective whole, which superficially looks like altruism.
My immediate reaction to this is to dismiss it as trivial word games. I get what he's saying, there are intellectual blind spots if we keep insisting that AI is "not intelligent", but simply changing the words will not fix this blind spot, because most people still find the distinction between e.g. ChatGPT and a human brain meaningful.
Moreover I default to being suspicious of arguments that boil down to "we can fix this problem simply by changing the language we use".
He definitely does more than just play word games.
I first came across him because of research he did showing that by altering the electro-chemical attractor states of tissues, you can radically change the functioning of an organism, in ways that were previously only thought possible by genetic engineering.
https://m.youtube.com/watch?v=RjD1aLm4Thg
I saw an inspiring art exhibit about the planned but unbuilt projects of Frank Lloyd Wright. Suppose I am a billionaire and I want to build those houses at long last. Given that Wright has been dead for over 60 years and the blueprints are available on the internet, could I just do it, or would I be violating some kind of copyright and/or need to get permission from his estate?
It depends. See chart here https://guides.library.cornell.edu/copyright/publicdomain
This is largely pointless whining, but I *am* curious about the answer. Why is the US healthcare billing system so fragmented? Two months ago, my son broke his elbow (he's fine, the surgeon did a great job, and five-year-olds heal amazingly). The treatment involved a lot of moving parts, but people seemed reasonably able to talk to each other, and all the procedures ended up in a single electronic chart. But, I've gotten at least half a dozen, if not more, separate bills (which are not in 1:1 correspondence with the insurance claims). Why isn't there some attempt to aggregate the bills? As far as I can tell, everything I paid was genuine, but I can't help thinking that it wouldn't be hard at all to set up a random payment processing website and fleece me for $100 or so (which admittedly isn't the hospital's concern, but it would be nice if I didn't have to worry about it.)
Any time you receive a medical bill you don't recognize/expect, call the hospital to verify it.
Fraudulent bills are, indeed, quite common.
And if we annoy hospitals enough that they stop sending bills which look fraudulent (holy shit I got a bill for an anesthesiologist once that looked LESS LEGIT than any scam I have ever received, the only way it could have been worse was if it were handwritten in crayon), maybe this problem will get a little bit better.
Seriously. Annoy your medical provider. Make their lazy accounting processes cost them customer service hours, and maybe they'll get their shit together.
The problem isn't that I don't "expect" the bill, I was just kind of hoping that, just as I didn't need to care about the anesthesiologist being logistically different from the surgeon or the radiologist or the emergency room or the ambulance *during* treatment, I also wouldn't need to care about it at payment time. For some reason they all insist on sending me separate smaller bills, on top of the large one sent by the hospital as a whole. (I don't remember this being as much an issue when I gave birth, although that may have genuinely involved fewer distinct groups of doctors.)
I was seriously pissed off when I received a steady series of bills, because I deliberately chose not to involve insurance after looking at the bill at the hospital, and deciding the personal cost of the additional hassle of dealing with insurance wasn't worth it. I was definitely not expecting those bills.