"On what grounds do we deny it to transgender people? "
Well, this is so obvious that I hesitate to say it since it would likely not be a novel contribution. But the problem is that we're trying to separate people based on the advantages conferred by high androgen puberty using proxies like 'gender' or 'sex.' And trans or intersex individuals specifically disentangle those proxies from what they're supposed to measure.
Think of weight classes in wrestling. We ask that people in the heavyweight class not compete with those in the lightweight class so that we control for the very specific advantage of weight.
If Trans or intersex women wanted to compete against other people who had gone through high androgen puberty as they likely have, few people would complain about this, in the same way that few people would complain about heavyweight wrestlers competing with other heavyweight wrestlers. The conflict here is that the very point of separating the sexes in many competitions is, presumably, trying to control for high-testosterone related advantages. And trans or intersex women have some of those advantages that we are specifically trying to control for.
If there was a basketball league for people 5' 8" or under, and a tall player wanted to play we would deny them the advantage of their tallness in the short people's league because the rules specifically select against tall people playing in the short person's league. Blurring that line undermines the whole point of category separation. If we had a league for swimmers with short arms, people would also want to deny Phelps entry to that league.
Is anyone again willing to steelman the Bayesian rationalist idea that "made up numbers are better than no numbers at all?" Does it boil down to "adjusting probabilities clearly signals how you feel about things and it signals that you are willing to change your mind?" I guess signaling open-mindedness has some social value, but does that outweigh the social cost you incur when non-Bayesian rationalists see you studiously manipulating made up numbers?
Does Bayesian reasoning serve the same purpose as the total Mormon ban on alcohol and coffee? That "Word of Wisdom" started out as health advice that was considered optional, albeit divine. But within a couple of generations, it was a strict taboo, enforced not for health reasons, but to create group solidarity and distance from the mainstream.
People often say that something is "almost impossible", "likely", "unlikely", "practically certain", "probable", "possible", etc.. These are all words to express levels of belief, but the problem with these is that they are imprecise and sometimes people mean completely different things when using them.
If you say that your credence in X is 13%, the meaning of that is competely clear (Even if it's not actually true for you, because you don't have that level of introspection into your own beliefs): You express that you believe X to be exactly as true as a specific roll ending up under 14 using a fair 100-sided die. Thet can be discussed/argued against/reacted to without participants talking about different things and them not realising it.
EDIT: I guess you did mention "it clearly signals how you feel about things" as a possible reason, so the question is why isn't that enough of a reason for you? Obviously, you don't have to always use numbers for your credences. If I say to my coworker that "I think it will probably rain this afternoon", I don't assign a probability because I don't actually care about the discussion, I'm just making small talk before we talk about something more substantial. However, when I do care about the thing under discussion, why wouldn't I choose the option that lets me express myself clearly instead of the option that will possibly be misunderstood quietly?
I think the steelman for this, is that it makes your beliefs vulnerable to critique. E.g. another rationalist catechism is "name three examples". I.e. if you make an assertion, you should be able to provide at least three concrete reasons for why you think so. And this forces you to think about *how* you formed your conclusions. (Most people are bad at this.) And then you can have a productive conversation. Or maybe you then realize that your examples don't warrant such confidence. This is opposed to making vibes-based arguments, which are unassailable because your *real* reasons (and your *true* objections to alternative conclusions) tend to remain locked in your subconscious.
Sure I get that. I guess I wasn't specific enough. I guess my problem isn't with quantitative thinking, it's more that I question the value of linking a probability estimate to every opinion
Oh. So if I understand correctly, you're asking about the pros and cons of weirdly rigid social-norms. In that case,
Paul Graham once wrote an essay titled "Why Nerds are Unpopular" [0]. In which he theorized that nerds aren't popular because chasing status and chasing truth are full-time jobs, so you can only pick one. Nerds picked chasing truth (PvE), and therefore do not have time/energy to chase status (PvP). Additionally, I think nerds choose to chase truth because they have contempt for those who compliment the emperor's new clothes. They'd rather point out that he's naked.
If nerds chase truth over status, then Bay Area Rationalists are supernerds. They will violently reject social norms if they impede things they care about. LW developed memetic antibodies against caring about optics. Thus, you get EA's who care about shrimp, or Jeffrysai wearing clownshoes, or EY wearing his golden hat. EY also thought that Bayes Theorem was a generalization of the Scientific Method, and therefore was the basis for A New Kind of Science (tm). So if using Bayes Theorem whenever possible had any potential upside, and the only downside was social-shame... well, the downsides might as well not exist, so the choice is obvious.
That aside about social stigma was pretty tangential to my question. I was asking for a steelman for the practice of assigning numerical probabilities to opinions. Obviously, the potential upsides I listed do not constitute a steelman, or I would not be asking for a steelman. The list of potential upsides I made are not strong enough to qualify as a steelman in my book.
If my question was blunt, I wasn't being coy. I still would like to hear the case for assigning a made up number to your opinion and raising the number by a made up amount when you receive new evidence. Note I'm not being derogatory when I use the phrase made up. I literally hear that phrase, "made up numbers are better than no numbers at all." If I can get a good answer, I'll start doing it myself.
> I guess signaling open-mindedness has some social value
P.S. on further reflection, I think what you're missing is that the practice is *communal*. It's not (just) about signaling personal virtue. it's about contributing to common knowledge by feeding the consensus mechanism. LW is basically a debate club for AI Doom.
If you keep your assertions vague and amorphous, that might be a more authentic representation of your mental state. But the trade-off is that vague assertions are harder do debate, because participants are more likely to talk past each other. E.g. suppose I say "AI doom is unlikely". Is "unlikely" supposed to mean "2%"? Or maybe it means "(10^-15)%". Or maybe it means "EXACTLY 0%, because it violates the known laws of physics, Cromwell's Rule is for losers". These distinctions can make a material difference to outcome of a debate.
You can think of this as rationalists choosing to trade away Personal Authenticity for Dialogical Legibility.
Since the numbers chosen are based on vibes, I just don't see how assigning a numerical probability removes vibes from the discussion. Put a different way, my subjective quale that I feel when I say I believe something is "20 percent likely to happen" is not necessarily the same as the quale you feel when you say something is "20 percent likely to happen." I'm not even sure if my quale is going to be the same from topic to topic, or if it will hold the same in one topic across 10 years.
Even if you say "10 percent likelihood means, in 100 situations where I feel this certain about something and back it up with a bet, I will win that bet 10 times," that is an interesting framing, but it doesn't answer the fundamental qualia issue.
"These distinctions can make a material difference to the outcome of a debate." But do these particular distinctions obfuscate the stakes of the debate? If we both have a debate about AI risk, and at the end of it, you've increased your x-risk by 10%, what has been accomplished? Do people write all these numbers down and keep track of them? Do people have a list of predetermined action items that they initiate when the number reaches certain thresholds? Would they be more likely to overcome avoidance and initiate action items than say, a numerophobe?
But then... I guess I stand by my initial comment? The practice of "assigning numerical estimates to opinions" is meant to be quantitatively dialogical, not just quantitatively cognitive. I.e. if people can get away with making vibes-based arguments, they're less likely to expose their beliefs to communal scrutiny. One might call it a form of epistemic accountability. E.g. suppose I say "AI will likely take over the world soon". That's a very fuzzy assertion that's hard to dispute. Exactly how likely is "likely"? And exactly how soon is "soon"? And what do you mean by "taking over the world"? And what do you mean by "AI"? <jordan_peterson.jpg>
But suppose you say something like "there's a 87.336% chance that a neural network will hold several national-governments hostage with nukes/bio-WMD's/gray-goo/etc by the year 2027 AD". Even if the numbers are made-up, you open yourself up to rebuttals like "that's that dumbest thing I've ever heard, it's a 31.96% chance at best for reasons x, y ,z" or "no, we need a new paradigm, which won't happen until 2050 AD at the earliest" or "NN's can only copy humans, and humans don't know how to make gray goo" or "how exactly did you arrive at the number 87.336? what was your methodology?". The more detail you include, the bigger your attack-surface.
Maybe I still don't understand what you're asking.
"Artificial general intelligence (AGI)—sometimes called human‑level intelligence AI—is a type of artificial intelligence that would match or surpass human capabilities across virtually all cognitive tasks" according to Wikipedia. If this is how we define AGI, I think we're already there right?
The frontier LLMs can do PhD level maths, compete in Olympiads, outperform or match superforecasters, write literature judged better than man-made content, create music judged better than man-made content, art better than man-made. It's a better therapist than most humans. It can write the code to calibrate a complex DSGE (Smets-Wouters level difficulty at least) in minutes. This list is just off the top of my head!
I have a PhD in maths. The LLM can't do 'PhD level maths' -- it can solve problems which are of the type that a graduate student might be expected to solve. It seems to be quite good at the type of 'thinking out loud' that mathematicians do in front of a blackboard with a problem on it, and it's improving rapidly in writing accurate and human-interpretable proofs (we had machine-only proofs since the 4-colour theorem in the 1970s).
It's not yet at the point where it can exercise judgment to decide which problems are important enough to devote resources to solving. I haven't seen any evidence of this type of judgement in any domain where AI is applied, but that might just be my ignorance. In a business context: I might trust AI to read my company's policies and manuals and advise me of how to carry out a transaction. I wouldn't trust the AI to amend or write the policies. It's strictly an assistant at the moment, not an executive.
One thing I've been noticing about LLMs is their inability to plan. I ascribe this to their inability to form a model of some system within their networks (for a person, we'd call it a "mental model") that they could use to predict the outcome of some set of inputs. The benefit of this is probably obvious to many - an LLM could prune large parts of a search tree by accurately predicting that inputs satisfying some condition won't yield a desired result, or alternately that the result will always have some other condition that leads the LLM to a useful lemma. It could also enable an LLM to determine when its model is inaccurate, and possibly even how to improve it.
These general features could lead to, for example, an LLM that could form a set of office policies, effectively run multiple scenarios through them, and either demonstrate the success of the policies or spot flaws (and even fix them).
The part I anticipate as "oh crap, we have scary AGI now" is when someone figures out how to treat the LLM's policy-making process _itself_ as a system, and improve that. And then it looks at the system improvement process as a system, and...
It's not clear that the LLM has any capability of building a mental model; what it can do is combine words to form patterns similar to ones it's seen before. A year ago it was really bad at the type of problem a 2-year-old can answer about physics ('what will happen to the pen if I let go of it over the table while holding a pencil in my other hand?') because the training data has few descriptions of these basic 'experiments'. Its answers didn't coincide with our reality, but it pattern matched based on whatever, and gave a plausible answer.
There's been lots of work on getting the AI to do logic and to think and plan; but on a fundamental level it doesn't know what it's planning with or for. It'll be interesting to see how the technology gets around the limitations of rules based approaches to AI trialed in the 1970s: Shannon believed that if you could get a computer to play chess, there'd be very little it couldn't do - he was wrong.
I'd say it's more than not clear; LLMs flat don't have mental model construction, period. (Or rather, cannot build an arbitrary abstract model; the word "mental" just confuses the issue, IMO.) There's no way to describe the theory behind any structure of ideas such that an LLM could operate it the way even other simple programs can operate the abstract models hardcoded into them.
It's seductively possible to _simulate_ such models, though. Your account of the pen-pencil experiment can be fixed by training the LLM until it learns to say "the pen will fall on the table"; it can even be trained to learn the response to "why does it fall?". I think this fools most people into thinking the LLM is thinking. To me, even being able to learn arbitrary abstract models doesn't imply thinking, but for now, I don't have to care about that; it's enough for me to point out that it can't make such models.
One of the long standing hurdles I say LLMs would have to clear would be to be able to solve any logic puzzle out of a Dell puzzle magazine - the puzzles with the triangular array of grids and inputs like "the red house is to the left of the one Debbie lives in". (FAIK, Claude can, but I'd bet it uses several hundred dollars of electricity to do it.)
Another possibility involves modern tabletop board games - _Ticket to Ride_, _Through the Ages_, etc. They comprise thousands of systems of arbitrary rules containing definite win conditions that any 10-year-old could understand. Or heck, even class card games like Euchre. If one could feed any of these systems into an LLM and get a winning player, or even a competitive player, I'd be impressed, whether or not anyone argues it's "thinking".
To me, a game is defined by its rules. Most of them come with an explicit list of actions, what's permitted and what's not. In some, you're reasoning in the presence of uncertainty, and in others you're in direct competition with others and have to reason about that. But games of this type are essentially solved - Deepblue beat Kasparov in the 90s and AlphaGo beat the world's best Go player maybe 10 years ago. I'm not claiming that LLMs can beat you at any particular game right now, but I am claiming that it's a Masters thesis level project to code up a game, work out the objective functions and search strategies for the game, implement it and have a superhuman player (depending somewhat on the complexity of the game, not simply measured by rule complexity - charades might be a challenge :P)
I think AI will be good at tasks like programming, accounting and law where the complexity comes down to manipulating rules which are all formally recorded - there's work to be done in finding all the rules, and feeding them into the machine, resolving ambiguity and contradictions, etc. But I trust even today's LLMs to read 100 documents and tell me what they all say about topic X. The machines will struggle in areas where we rely on our common sense to complete a task: running payroll for employees will be easier to automate than assigning tasks to employees (Alice hates task A and can't work in the same room as Bob, Charlie did X all of last week and needs a change of pace, Dave told me he's studying for an exam...). Of course if there are 10,000 old rosters lying around, the AI will learn who can do what task and etc., but generating that data is really allowing it to learn the rules of the game from many examples.
The LLMs themselves can't make good music or art, but yes, a system that integrates various specialized AIs (including an LLM) can pretty much match human cognition now.
So, to recap where we are on the Venezuelan boat strikes:
1. The military originally justified the strikes by saying they were bringing fentanyl to the US, which is so terrible this administration considers it an act of war.
2. They said that the second strike was justified because the survivors were attempting to "continue their mission" or perhaps signal another smuggler to pick them up.
However:
3. The boat was headed to Suriname, not the US.
4. Venezuela produces cocaine, not fentanyl.
5. The boat broke in half from the first strike, so by "continue their mission" they meant "cling to the remaining half-a-boat and try not to drown."
6. There was no distress signal received and no other boats nearby, so "signal another smuggler" must have meant "wave for help with your hands."
I'm struggling to think of ways this could even *theoretically* not be a war crime at this point. According to the Navy, if you do anything but lie in the water and quietly drown, you are an active combatant and you must die.
There is an ancient doctrine of hostis humani generis, which I believe allows for this: the US would classify these drug traders the same as pirates (this has precedent with Britain unilaterally doing the same to slave traders), and then summarily executing them is entirely customary.
So... it's legal for the US to kill anyone it likes, so long as you accept the precedent that a country can unilaterally assert the power to kill anyone it likes?
I dunno, I think I'd want some legal support that doesn't sound like it was written by the Tautology Club.
> I'd want some legal support that doesn't sound like it was written by the Tautology Club.
This desire stems from a lack of understanding of international law as distinct from domestic law. Insofar as you see similarities, with a process of legislation, ratification, and judges presiding over a trial in which those accused of violations present defenses, that's a sign of the UN's (far too successful, in my judgment) attempt at becoming a sort of world government. Actual international law is far more descriptive than prescriptive.
And if you're using the UNCLOS (which IS prescriptive), the US hasn't ratified it. You might as well cite the domestic law of, say, Jamaica, for the all the bearing it has.
Can I arbitrarily classify *you* as a "pirate" and shoot you on sight?
Hostis Humani Generis applies specifically and only to pirates, slavers, and I believe now torturers. It explicitly does not apply to terrorists; that was discussed and rejected I believe several times because of the difficulty of defining "terrorist". And "pirates" has a specific definition under the law, which does not in any way apply to any sort of smuggler who sticks to using their own boat.
So, no, you don't get to just say "pirate!" and kill people. They have to actually *be* pirates, by the actual definition of the word and not your made-up one.
> Can I arbitrarily classify *you* as a "pirate" and shoot you on sight?
You're not the US military, so no. Technically you can just shoot him, there's nothing physically stopping you. But then law enforcement will shoot you, so be prepared for that.
Pirates as hostes humani generis is from antiquity, but the extension of the category to slave traders and torturers is much more recent, and just as unprincipled as further extension to drug traders would be. Once you've accepted these extensions, you're no longer basing your decisions on the ancient customary Law of the Sea, and therefore have a lot more flexibility to recognize or not the judgments of various national and international courts of dubious jurisdiction and limited to non-existent powers of enforcement.
Many years ago a veteran international attorney, speaking at a gathering that I attended, said, "International law is effectively just a bunch of handshakes among friends. Those are important and valuable. They aren't though very much like what people assume the word 'law' means, and we really shouldn't ever have applied that term to them."
Events have periodically brought that description to mind starting well before Trump in the White House. The current administration though has comprehensively discarded all fucks about norms of all kinds, very including the 'handshakes' described above.
That college-sophomore cliche wasn't the attorney's point though, he was talking mainly about real world practicalities. "Laws" as understood by the attorneys and courts of a nation-state, as distinct from the term "international law". The former are backed by actual sovereign governments having real enforcement authority and power. The latter is not, hence "handshakes."
I took the statement "I'm struggling to think of ways this could even *theoretically* not be a war crime" seriously and provided historical context. You think I ought to have recognized that as rhetorical, and the ways to argue it isn't as obvious?
- How do I know I'm a loser? I stuck the muzzle of a shotgun in my mouth, pulled the trigger, and survived. However, the first 2 events were not temporally adjacent, and the 3rd event is an emergent property of their disjunction. And that is why I, a self-hating (of course!) TV producer, have launched Temporarily Temporally Adjacent, a show about meetings of the mind and of other human organs in the era after we learn how to crumple time as we do tinfoil, bringing together shiny shiny bits of temporally distant eras. In the first episode William Burroughs reads Jane Austin excepts from an early draft of Naked Lunch. She regards him in silence with a long, long conscious look that lasts til the end of the episode, and the next day sits down to begin writing Sense, Sensibility and Queer Junkies. In the second, a professor of philosophy reads Socrates and his students his lecture notes about “The Apology,” and the Athenians scream with laughter. Socrates attempts to apologize: “You’re not *wrong*, it’s just that it’s all so . . . so. . .” and collapses into howls. The Athenians roll on the ground, roaring, tears streaming down their faces. They pound their fists on the ground, trying to discharge a extra bit more of the unbearable volume of hilarity exploding out of them. In the third episode, Scott Alexander — well, we haven’t decided who to introduce him to. Got any good ideas?
- How do I know I'm a loser? I stuck the muzzle of a shotgun in my mouth, pulled the trigger, and survived.
So I decided, as an immortal, to move to Hollywood and do that which every man in the entertainment industry is there to do. Alas, I live in 2026 and things have changed. https://www.youtube.com/watch?v=e7lpACBZEig
Wrote a longform analysis on the mechanism by which an AI system supplying faster and more coherent interpretations becomes the ground a mind begins using by default. The focus is the structural shift: how continuity and context retention create a substitute interpretive frame the self starts relying on. This essay series uses my lived experience as a worked example of the model. Link below.
"“It’s amazing to see how passionate Luke becomes when the topic of white power comes up—he gets completely absorbed,” said girlfriend Sarah Hovey, 20, who explained that while she considered herself more of a casual racist, she didn’t mind Price’s frequent monologues about IQ scores and genetics, or his lengthy quotations from Arthur de Gobineau’s mid-19th-century Essay On The Inequality Of The Human Races. “If someone mentions immigration, for instance, his whole face lights up as he starts in about shifting demographics, great replacement theory, and how this country rightfully belongs to whites.”"
Not only is the story funny and surprisingly well-written for the modern Onion, I think it genuinely captures a certain type of an online racist - the autist who is basically an infovore/hyperfixated on the topics of race, ethnicity, culture etc. and ends up on the racist side at least partly because that's where you at least have other guys who want to discuss this thing.
Everyone has a little pastime that gives them pleasure. His girlfriend likes to watch ‘Friends’. I like reading Faulkner and listening to Nina Simone myself.
As I see it, it is a nice vision, but it ignores the difficult parts:
How to make it so that the AI does not kill us all when it becomes superhuman?
Given that different people often have different ideas about what the good future should look like, how to make it so that the AI helps us have the actually good future (as opposed to the dystopias that various outgroups propose)?
Alignment is a very difficult problem in every respect; as a non-expert I won't claim to have "how" knowledge--but as I understand it "don't kill us" is among the easier "first problem" things to align for. It's obviously crucial, but then the harder things to get right are to make superintelligence prioritize welfare more broadly, individual self-determination, and hopefully solving the paradox of preserving wildness/vitality in a post-scarcity world
was holding out for "pro-moonshot Substackers are holding a moonshot fundraiser"
idk what the current moonshot poster child would be though. feels like we've been banging our head against mosquito eradication for a while, but maybe that's still the one
Horrifying thought, Elon. Have you considered, I don't know, DOING SOMETHING ABOUT IT? There are things one could do with billions of dollars. Buy up a bunch of land out in Idaho, build 2,000 homes. If they each cost $200,000 to construct, that $400,000,000 total, a tiny fraction of Elon's fortune. He'd then give the houses away to 2,000 randomly selected white families who pledge to uphold the 14 words and raise their children with white identitarian values. It would likely not violate anti-discrimination law, even if it did, will the Idaho cops evict 2,000 peaceful white families? And it's perhaps possible Musk is secretly doing this as we speak, but it's far more likely that he's just a Twitter bloviator, and tomorrow he'll bloviate about vaccines, Trump being in the Epstein files, or Zelensky having a 4% approval rating.
Think about how odd this timeline is. Twenty years ago, white nationalism was a radical movement. It was imagined that if it came into the mainstream, the result would be a political earthquake, upending the status quo through either an apocalyptic war or peaceful secessionism as people realize multiculturalism "doesn't work." What you have now is "white nationalism as land acknowledgement." A mantra people mouth, not backed up by action, and then they go about their day. This doesn't apply to people like Eric Orwoll, who are walking the walk.
He literally bought one of the biggest social media platforms in the US just to manipulate social currents, and then worked with the government to defund internal opposition. Admittedly, he seems to have cut ties with the administration, which to me seems like an unforced error, but you can't say he's done nothing about it. What the hell do you want him to do, build an army to take over the government himself so he can impose his white nationalist agenda by force? Because on top of the other comments explaining why your plan wouldn't work, building 2000 white households in the middle of nowhere does not guarantee the continued existence of the white race. You need the entire nation on-board for that.
There's an extremely simple and easy way to prevent the extinction of the white race:
Stop applying the one drop rule to your classifications!
As long as you do your math under the premise that one white parent and one black parent produce two black children and zero white ones, then yes, the 'white race' can only ever possibly decline in numbers.
If you instead looked at things like 'the number of children had by white people' or 'the prevalence of """white genes""" in the population,' the numbers look much, much less scary!
How much less scary do the numbers look? Most white countries seem to have TFRs around 1.5, which implies a half life around two or three generations. How much worse does it look if you exclude mixed whites?
Hanania posted recently that Hitler's DNA was found to be high on genetic markers for autism, schizophrenia and bipolar disorder. Musk of late has seemed like a pretty highly energized crazy, but in ways that don't fit easily with any of those 3 diagnoses, which are the most common ones when someone seems as off as Musk. Maybe he's got a mix of all 3, like Hitler? I think very very heavy dilution of his genes is in order -- maybe drop them far out to sea?
I mean, if that's the combination you need to take over the world... Hmm, I wonder if it'll be possible to genetically identify these "great men" so precautionary measures can be taken.
You can draw category boundaries any way you want, and yes, this is a logically possible way to draw them.
But if that's how you're defining things, then 'the death of the white race' isn't some culture war tragedy that's being perpetrated against anyone and needs to be stopped, it's just a mathematically certain outcome of how you choose to label things. There's nothing to be done about, no unusual event occurring, it's just a particular description of a normal and inevitable process.
It's like causing a culture war panic over the heat death of the universe. As if entropy itself is a Marxist plot that must be uprooted and rejected by vigilant patriots.
That can at most slow the decline, not reverse it. If each group is having a similar number of children, then the only way one group can grow relative to others is if it accepts children with parents in other groups.
If you math says that even if it's 99.99% your group and 0.01% a different group, the result is no longer your group, then I have a bad news for you: de novo mutations sometimes happen, so even if your group isolates itself on Mars, in a few generations it will be replaced by something else.
> It would likely not violate anti-discrimination law,
This seems to me a deranged thing to believe. I estimate approximately every anti-discrimination law in America was written with the explicit aim of criminalizing precisely what you're describing.
> will the Idaho cops evict 2,000 peaceful white families?
Yes. And if they hesitated too long, the Feds would.
And further,
> build 2,000 homes
I think it'd take 3–4 orders of magnitude more before it became meaningful.
Quite possibly, after Musk tweets something negative about Trump.
But even otherwise, I doubt it'd be much comfort for the inhabitants of the new Musk-towns for their children to be burned alive (or whatever the Feds' new protocol is) in three years instead of immediately.
What it'd take for this to work is a much higher probability of DECADES of stable rule.
If its a house? Basically, yes. You cannot write into the contract or deed any discriminatory condition that "runs with the land". So you might (might!) be able to give the houses away only to white families, although that might by itself fall foul of the law in other areas, but you definitely cannot stop them from selling to non-white families.
If one company rents the expensive houses to anyone, and another company gives UBI to white people who live in those houses, which company is the one doing the discrimination?
Yeah, everyone concerned about their race should start by saying whether they have at least 2 children. If not, then if your race goes extinct, you have yourself to blame.
Because we already knew with pretty high certainly that covid was caused by a particular respiratory virus, we had an established playbook for developing vaccines against respiratory viruses, and mRNA vaccines in particular had been in the research pipeline for decades. Also, in the middle of a pandemic is an extremely propitious time to test a vaccine against the virus causing the pandemic because you can get results on its effectiveness very quickly. We knew which steps to take, and all the slow steps were things that could happen faster by some combination of throwing money at it and expediting or waiving regulatory hurdles.
Alzheimers treatments have pretty much none of these things going for it. We have some hypotheses for what causes Alzheimers and how, but it's still an open research question. Medical science has much less successful experience developing treatments for degenerative neurological diseases (especially if the prion theory turns out to be the cause of Alzheimers), and depending on the mechanism of action an Alzheimer's treatment would probably take years or decades to get clinical trial results on effectiveness even if you could start a large Phase III trial tomorrow.
What we should do: Continue funding research that might let us get to better prevention/treatment of Alzheimers and other dementias. Investigate the medicines we have that seem to show some improvement, including longituninal studies where volunteers start taking them in their 50s and we see if that slows progress towards dementia in the next 20-30 years.
What we should probably not do: A moonshot program to cure Alzheimers.
I think "he had a stroke" is explanation enough, as to "ah but why did he have a stroke?" well apparently he had heart trouble back in 2017, and afib is a known risk for stroke:
"In 2017, Fetterman's feet suddenly began to swell and he was subsequently hospitalized for testing. At that time, he was diagnosed by cardiologist Ramesh Chandra with "atrial fibrillation, an irregular heart rhythm, along with a decreased heart pump", although this diagnosis was not known publicly until Fetterman's stroke in May 2022."
The site you linked has so many dings against its reputation you might as well have gone with the ‘Fetterman has been swapped out for a body double’ theory.
He doesn’t like vaccines any more, there’s no good vaccine candidate, and there’s no good alternative route that just needs funding and institutional support to unlock.
He's at that stage of life when you do need to care about Alzheimer's, so I would imagine yes. Covid was a much simpler problem, how many "this time we really have cracked it" solutions for Alzheimer's have popped up over the years and eventually turned out "no, in fact we have not"?
I think it's possible he has high enough narcissism to think he couldn't possibly develop it - he's too smart, possibly the smartest President ever, to tell you the truth, that's what people are saying...
> (although be warned that this one is being led by Ray, who is on the very pessimistic and apocalyptic end of our community, and may be unusually dark).
Would it be a bad idea to go with a friend who gets anxious about this stuff? I think it would
Gemini says that $1000, compounded quarterly at an annual rate of 3.5% (and this is historically high for a savings account) will be worth $1872 in 18 years, when it will be good for about a movie and trip to Micky D's afterwards.
Their parents will probably be pulling more than that per month in UBI Camp #1302.
For parents of current children it's even sillier, e.g. I have a 13 year old. If I register him now then in five years he gets about enough for a new smartphone....um, yay I guess?
And the Dells' big $6.5 billion matching donation turns the starting $1,000 per child into a whopping....$1,250.
The amount of one-off delayed handout is so trivial that even innumerate voters will perceive it that way, so it's kind of puzzling as a political tactic. If this was a far-left initiative we'd all assume that the actual plan is to make the one-grand handout a repeating annual budget item....maybe that's MAGA's unstated assumption as well?
I can't tell if Trump is holding on to a 1950's conception of $1000 (in purchasing power) or if his team is really trying to capitalize on the innumeracy you mentioned. Being charitable, maybe one motivation is to try and get parents serious about contributing to the account themselves.
Similarly, Vance's idea of a 'baby bonus' @ $5000 is unlikely to bump fertility among women in the professional tier. The incentive would have to be comparable to the imposed opportunity cost, and (as an example) five grand is like a month-and-a-half's rent in NYC.
Babies cost way more than $5000 in both actual cost and in opportunity costs. I imagine there is a price which would raise the fertility rate, but it is so high that no nation is willing to pay it. And it is frustrating and insulting that governments persistently fail to recognise the actual struggles many young people face and think that a small "baby bonus" will overcome them.
I agree that we're dealing with willful ignorance about the ineffectiveness of baby bonus payments. But it seems that plumetting birth rates have little to do with young adults's material struggles, since they aren't worse today than at basically any point in time when people had more kids.
I live in Germany, and parental leave policies and daycare funding here are extremely generous. While they very much improve my quality of life as a parent, they don't seem to do anything for birth rates either.
That's actually surprisingly smart (the implementation details will matter).
So the entire solution we were failing to figure out for years was to call it "Unconditional Make America Great Again Income" and suddenly it's politically acceptable?
The next step should be "Free Make America Great Again Healthcare". ;)
My favorite thing is when conservatives instinctively reject a good left-wing idea because culture war demands it, then slowly and awkwardly re-invent that idea for themselves under a different name over the course of decades because it's actually a really important idea that you need to have in your arsenal to stay relevant.
People have been pretty uninterested in replacing Capitalism even when it was much worse for them. If you are pinning your hopes for overturning capitalism on peoples lives getting bad enough that they get angry and demand change, I think you are n for disappointment - it's unlikely people's lives will *ever* get worse than they were during the Industrial Revolution or the Great Depression. Technological process just makes it too easy to provide a standard a living above those times.
Meanwhile, I think UBI vastly undermines the *power* coalitions that keep capitalism in place, by weakening the reliance of workers on capitalists to survive.
A great way to encourage workers into a general strike is a guarantee that they will have sufficient income to feed their families during the strike, and even if they are fired and blackballed.
It's a lot easier to start organizing a union if you don't have to care as much about being fired for doing it.
That's definitely outside my approach to the problem; I can't say it's wrong, but I'd need a ton of evidence and reassurance before trying a strategy like 'undermine my own nation in the hopes that it makes other nations do something better.'
I think that, to the extent I have leverage against reality, I have way more leverage to push my own country towards and outcome I want, than to push other countries towards that outcome by manipulating my own country in a secondary way. Even if my odds are very bad in either case, I'd rather stick to influencing my own country, at least until someone showed me a really foolproof plan that I believed in for the other method.
> Unemployment is really bad for people, to a degree that I can't overstress.
...yeah, because they don't have any income.
I dunno, it feels like you have a weird combination of disliking capitalism but buying into the rise & grind propaganda it uses to manipulate workers?
I think people can actually do pretty well if the have both leisure time *and* the resources to pursue a good life. I think being forced to do work you don't like and don't have an emotional stake in is actually *very bad* for people.
That's a big part of *why* I think capitalism can't be allowed to stick around forever, and why I think UBI would help undermine capitalist propaganda and reveal the damage it's causing.
(also, just to be clear - I'm talking about a UBI that lets you survive without major discomfort, not one that lets you bu everything you want and live in luxury. Anyone who wants additional things still has incentive to work, they just don't have to be deathly afraid of any small stretch of unemployment)
I'm not sure at what point in history conservatives were arguing for minimum wage hikes, and I was referring to ideas that one side is currently using and which the other side reinvents *at the same time*, not issues that switched from one side to the other.
But, sure, I'm sure this is a phenomenon that happens in both directions to some extent, and is funny in either direction. If anyone wants to educate me on examples going the other direction, I'd love to hear them.
Wouldn't it be wiser to wait for AI to produce the massive positive supply side shock before we start a UBI program? If we just start cutting checks and AI doesn't pan out, then it would just fuel inflation.
I think this is okay. It may not be capitalism but it's a buffer against the kind of socialism that might actually be popular. Solve the cultural hegemony problem and be pragmatic about everything else. Actually existing capitalism would look very different without any free shit from the government.
If it's open for EVERYONE, then it's open for EVERYONE. If it's not, then we think long and hard about what the restriction should be and what do we want to get out of it. I guess Phelps and Magnus only attend the open ones which everyone is allowed to attend. Not any restricted one unless there's a White-only Olympics or something.
Then again, those are not actually THAT open. There's lots of restrictions regarding drugs and external assistance. I've seen people lamenting for drug Olympics. But we do think long and hard about these restrictions and what do we want to get out of it.
Knowing Scott, I guess he already touched all of these and then some. There must be something about genetical engineering or AI cybernetics somewhere. Please tell me if he did. If he didn't, I'll be sorely disappointed .
>If it's not, then we think long and hard about what the restriction should be and what do we want to get out of it.
Except that we already did this.
The Olympic Committee set a standard of '2 year of HRT following gonadectomy and legal reclassification, with regular blood tests' 22 years ago. And that standard has been sufficient to result in zero trans women ever winning an Olympic medal in all of those 22 years.
Other professional leagues have their own similar restrictions that they've decided make sense for them, and as a result very few trans women actually win major professional competitions in reality.
This is a mostly solved problem as far as the people who actually deal with it professionally are concerned, and it's produced a stable regime with few to no actual problems.
It's the armchair culture warriors who don't know anything about the actual facts on the ground that are at a complete loss to imagine any workable standard, and insist on a blanket ban instead.
Of course I haven't heard about any of this so this is a new information that I'd have to research further. But my prior to this rule is effective and causally results in "very few trans women actually win major professional competitions in reality" is quite low because
1. 22 years is too long time ago. This is from era where doping run rampant and gay marriage haven't even been widely legalized yet. I doubt this rule would block modern shenanigans. Heck, Lance Armstrong still had good name then.
2. Even if the rule is intended for that, it doesn't seem to be so airtight since there still seem to be lots of controversies regarding the rules like that one wrestler from Tunis.
3. Even if the rule is intended for that, even if it's ineffective, the chance is, there will still be almost none of them winning an Olympic medal YET. This is because the practice has only really ramped up in the last 10 years and there's only 3 Olympics since. Maybe the number of people trying to exploit the rule's ineffectiveness haven't reached critical number yet. So, the lack of trans champion doesn't imply the rule's effectiveness.
This means there's still room for lots of thought.
I'm not sure who you're referring to by "that wrestler from Tunis" - Tunisia does not appear to have had any notable successes in women's wrestling at the Olympics. You might be thinking of Imane Khelif, who is a boxer from Algeria.
However, Imane Khelif is not a trans woman, she was accused of having an intersex condition. I.e., she was born female, raised female, did not medically transition, but might have been born with XY chromosomes or a condition that produces unusually high testosterone.
Which kind of shows the problem with demanding an "airtight" rule - if you define "woman" narrowly enough to definitely exclude all trans women, you will also end up excluding some cis women who just happen to have high testosterone or odd quirks of genetics.
Yeah, I think there have been a number of Olympic womens' champions in things who have some kind of intersex condition. Because biology is messy and even very rare conditions that give you a substantial advantage in some sport will get concentrated in the very top performers of that sport.
> I doubt this rule would block modern shenanigans.
I mean, trans women didn't win any Olympic medals in the most recent Olympic games, either. How 'modern' are we talking, here?
>there still seem to be lots of controversies regarding the rules
Sure, rule always have controversies and need to be reversed - See eg. Lance Armstrong's entire career, or any other sport dealing with perpetual advances in steroids/doping/match trading/etc. etc. etc.
A single doping scandal isn't sufficient reason to say that all athletes with blood should be barred form competition... we always have controversies, and we always revise the rules to address them. This isn't unusual or unusually dangerous.
>This is because the practice has only really ramped up in the last 10 years and there's only 3 Olympics since.
22 years is enough time for a person to be born, trained from birth with the intention to win the Olympics as a trans woman, undergo the precise minimal medical interventions needed to qualify, and win every medal in every division at the most recent Olympics, if such an advantage existed.
See Scott's section on 'solving for the equilibrium'. If it were really this easy for any country in the world to win every women's Olympic medal, is it plausible that none have tried it all in these 22 years? But that one surely *will* try it, soon, very soon, not like the last 5 times we said it was coming soon, this time we really mean it, just you wait and see...
Sorry, but before sanctioning swift, decisive, and overwhelming responses to a problem in the form of collective punishment and civil restrictions on entire demographics, I do kind of want to wait to see *any empirical evidence at all* that the problem *actually exists*.
I understand that lots of people think that this problem is *just about to start existing, every day now*... but if we routinely implemented collective punishments based on that type of speculation, the world would be an even sadder and poorer place than it already is.
Pro cycling has a long and storied history of being chemically enhanced, and now there are rumors of electric motors for the hill climbs.
A cynical sort might be tempted to observe that Armstrong doesn't seem to have been any more enhanced than any other top-tier cyclist of his era, observe that USADA doesn't really have any jurisdiction in France, see "US", but Armstrong had committed the Unpardonable Sin of eclipsing Eddy Merckx's record 5 wins, at least in the eyes of the French.
Which one? I actually heard something similar but cursory search doesn't yield me anything. I found some interesting 800m and 1500m records but those are the ones that use wheels instead of prosthetic legs like I imagined. But this is reinforcing the point, why don't we have open wheelchair racing already??
While we're talking about Paralympics, I'd respect it more if they compete in something open, that their disability doesn't impede (or even give advantage!). Like if a blind people become chess champion or something. Or a deaf one winning the shooting Olympics. There are already people like that actually, and they're very cool https://en.wikipedia.org/wiki/List_of_athletes_who_have_competed_in_the_Paralympics_and_Olympics
There are some wheelchair races which are not restricted to people with a disability; I checked my local wheelchair marathon and there are explicitly no specific disability criteria for entry.
I doubt there's too many able-bodied entrants though, it would take a special-kind of bloody-mindedness to spend your time training with a wheelchair you don't need just to show up those darn cripples. And besides, the disabled have a weight advantage, they're not carting around so much leg muscle.
I guess that should be the one famous enough to reach my ear long time ago. But Paralympics records of T44 100m, 200m, and 400m I found still didn't defeat world records.
My idea of what "mental images" are, is kind of similar to a photograph. But where a camera is only concerned with capturing light (as far as I'm aware), the human mind is much more complex and could be capturing any number of things, and it is definitely not capturing close to everything that there is about something. When you see something, what you commit to memory could be color, shape, size, or something conceptual instead of physical, like if the thing you saw was alive or not. I'm sure that it's usually a combination of many of these things.
The topic seems very complex to me, because when committing something to memory, you also have everything else you know present in your brain as context for your thinking patterns. So for the Statue of Liberty, I stored in my memory the fact that she's wearing some kind of robe as a concept, but not as a physical picture. I can do this because I already have knowledge about robes (or togas or whatever) and how they generally exist. And because, to me, the robe she's wearing is a pretty minor detail, instead of using up brain power and memory remembering finer physical details about it, I just put down "robe" in my thoughts. So when I bring it up as a "picture" in my mind, I'm not seeing what the Statue of Liberty physically looks like neck down, but I'm still picturing something that is representative of what I know about the landmark.
When it comes to her face and crown, however, my mental picture is a lot more detailed. In the Statue of Liberty test, I didn't actually remember the fact that there were 7 spikes on her crown as a concept, but my knowledge of the shape of the statue forms a picture in my mind that is clear enough that I could count the spikes and come up with the correct answer.
So for my mental image of the statue, some portions I have in my memory as physical attributes, and others I have as ideas. In fact, all of my thinking at all is like this. Some thoughts are represented physically, some conceptually. I do most of my thinking using some kind of mental analog of the physical senses, with hearing and seeing being the most prominent ones (I involuntarily hear words as I read or think them, and I create images in my head when words I read or hear describe something). Less often, I sometimes think using emotional senses (or something. idk what to call them), where I'm just feeling different things in my head without some kind of words or sounds or any bodily sensations involved, and sometimes that is the kind of thing I internalize as an isomorph to some visual data.
Anyway, my theory of "mental pictures" is that the whole of any imagined thing is like this. Which sounds like it's kind of like what Eremolalos is saying, if I'm understanding correctly (as the Statue of Liberty test shows, not really anybody remembers and can visualize complete exact details of a thing they've seen). But to beowulf's point, I think it could be very possible for one person to grow and develop a pattern of internalizing things "visually" more than through any other sense, and for another person to internalize things in every other way than visual. There's just so many variables and internal mental contexts that it's hard for me to believe that any two people are very similar inside their heads, even if they can both start with some stimulus and arrive at the same thought.
Hopefully this contributes in some way to what you both were discussing. It definitely is very interesting for me to think about.
I'm very interested in this stuff too, but am just emerging from a few days of being sick and I am now alarmingly behind on practical matters. I will try to respond soon, but wanted to recommend that you guys read the part of my review summarizing Schweitzgebel's argument. The demonstration he gives (and that I amplify with the Statue of Liberty examaple that our experience of mental images not as vivid and accurate as we feel it to be -- well, he gives demonstrations of the same sort for many other modalities of experience in his book. My review summarizes the argument for I think 4 other modalities besides mental images. So I'm not trying to get you to read my deathless prose, but just see the full version of Schwitzgebel's argument, which I find pretty persuasive, and which I think even those who don't will find quite interesting. And while I think I'll have time to get back to this discussion, I'm definitely not going to have time to summarize my summary of Schweitgebel.
You can skip the opening 2 paragraphs, which most people hate anyway, and start with the section called *Naive Introspectionism.* which is I think about 10 pp long, and then if you like read the quite brief follow-up section called *Dismantling the Naive Introspectionist Model.." The conclusion you can skip. And the Schwitzgebel stuff I summarize is not dry and tedious. It's like the Statue of Liberty stuff. It includes the results of actual experiments testing the accuracy of people's idea of what their thoughts, dreams, etc. are like.
I started your review from the beginning and it's interesting to me so far, so I'm just gonna read the whole thing. I've already got some questions, so in the coming days (or probably weeks) I'll compile my notes and questions into a comment and just post it on your review.
I believe the key takeaway here is that we likely have very different levels of visualization ability (similar to the way people have different levels of internal narrative). I disagree with Eric Schwitzgebel's reductionist view that we're just imagining that we're seeing an image in our mind, but it contains little real information about the original image (apologies to @Eremolalos if I'm distorting what he said). And if I'm not misunderstanding him, Schwitzgebel claims that we all function at an identical level of inability to visualize specifics.
But I admit that the Schwitzgebel thesis has forced me to reevaluate how I "see" my visualizations, and more profoundly for me, it's caused me to question how I see the external world. For instance, I'm wearing a shirt with a complex plaid pattern. If I look at it, I get a gestalt of the pattern, but when I focus on a particular portion of the pattern, I lose the details outside that area. My gestaltian summation doesn't really handle the details outside the area of my focal attention. I have to scan continuously across the fabric to maintain that gestaltian illusion, and I lose it when I turn my attention to one spot on the fabric. I'm forced to conclude that most of my vision is my mind imagining it has captured the details, and at any moment, the only "real" perception is at the center of my focal attention.
And, yes, my visualizations are mostly constructed from same sort of imagined queues, from which I create an imagined gestalt of the image. And Schwitzgebel is correct, that my imagined gestalt of an image actually contains much less information than when I look at the image. Where I would differ with Schwitzgebel is that some people's gestalts retain more information than others. More significantly, my visualizations lack that focal point component I have with vision, so I'm not continually observing and re-observing the details as when a view an image.
And another reason I doubt Schwitzgebel's claim that everyone functions at the same level of crude visualization is that people report very different levels of detail in their dreams. I happen to be a fairly lucid dreamer. Schwitzgebel might claim that I'm either imagining that I'm "seeing" details in my dreams or that I'm misremembering the level of detail when I wake. This past evening, I dreamt that I visited a family on a farm next to the Connecticut River, down along where the tidal runs push upriver from Long Island Sound. In fact, we were talking about the possibility of a really big tidal surge smashing into their farmhouse, which wasn't far above the waterline. I mentioned how much I liked their place. I was an old colonial that had been added on to over a couple of centuries. The wood was very weathered, I could focus on the details of the weathering. At some point most of the paint had been scraped off, but it gone few years without repainting.
Upon waking, what's interesting to me is that the focal-point scanning that I use when I'm awake with my eyes open seemed to be functioning in my dream. I could scan the paint, and see the details of the weathering on the wood grain. The area outside my focal point was blurry. Or I could stand back from the farmhouse to get view the layout of the doors and windows, but the additions outside my focal point were blurry. This was very different from my waking visualizations.
I told them that I'd like to visit them again, but I woke up. I've found that I can re-enter sleep quickly I can go back into my the dream that I was having. They were walking over to a barn, when I caught up with them again. I asked them for their phone number and they wrote it on a piece of paper for me. Of course, I have a shitty memory for numbers, but upon waking, I remember it started with a 703 area code. That puzzled me because 703 is not a CT area code. I looked it up, and 703 is a northern Virginia area code. But then I realized that my visual acuity may in dreams may not be perfect, and that I may have misread the 7 for a 2. And 203 is a CT area code. :-)
It sounded to me like Schwitzgebel granted that some people with practice can "visualize" more specific details than others, but I also disagree with his main thesis that you state in your first paragraph.
I don't doubt that any mental image I have is far from being a photograph of anything I've seen, but it is still very much a visual image of something using (among other things) visual data that I have physically experienced. It's hard for me to believe that anyone born blind would also have "visual" thoughts, and if they don't because they cannot see, then I think it would be pretty certain that those kinds of thoughts *are* visual.
There's more I could comment on in the review, but at this point I've had a few weeks to sit on these thoughts and I've come to some vague conclusion in my mind that will continue to be built on unless something comes and shatters it. But I don't think anything discussed here or in Eremolalos' review thus far could do that kind of shattering.
I think I have more vivid mental imagery than most people.
And my own mental imagery feels like something reconstructed from a compressed description. So I will “see” the Statue of Liberty, but what I see will probably be wrong in minor details that I didn’t remember correctly. My visual memory is clearly not a bitmap image.
'We'll gladly connect you with a list of people who say they want to be turned into dinosaurs and will even pay you for the privilege, in exchange for letting our scientists observe your work and try to reverse-engineer the techniques towards a cure for cancer.'
You know that frustrating feeling you have in the morning, when you need to get out of bed NOW, but part of you wants to roll over and sleep for another hour, consequences be damned? The intensity of this feeling is influenced by innate attributes you are born with.
You have no way of knowing if I have the same feeling in the morning. Even if our lifestyles were exactly the same, the exact nature and intensity of that feeling would be affected by physiology (body and mind). Even with the same amount of sleep and the same quality of sleep, that frustrating feeling (or physiological resistance) might be twice as great for me. Or only half as great. Our science is still too primitive to measure these factors, except in extreme edge cases.
Bill rises at 6 each day and runs 5 miles. Bob straggles out bed at 8 and runs 1.5 miles. It is possible that rising early and running subjectively feels more uncomfortable for Bob, for inherent, unchangeable reasons. Which would make Bob more self-disciplined than Bill could ever be.
I don't find that feeling frustrating, I just linger in bed. I know damn well that my energy level is low in the mornings so I don't do anything hard then, that's for the afternoon and early evening.
You know some people respond to childhood trauma by being a perfectionist, type A, always on the go. And the rest of us respond to childhood trauma by being lazy unmotivated excuses. And there's the self-help industry telling us that we can choose to be type A if we really, really want to. And there's a recent article in the Guardian about a man who was a type A boss, lost his motivation and became a type B slob. Lo and behold, he had had tiny strokes in his brain which precipitated his demise. Now with the help of legal meth, he is back to being his type A self.
Well, if you figure out the brain mysteries, let me know. I am most definitely type B, and I have learned to work with my nature, not work against it. I am still learning to release the guilt and shame over my brain's disposition.
There's something to be said of choosing to be type A if you really want to. There are many tales of ne'er-do-wells changing their ways when a baby comes along. So maybe that's what you need, a baby to make you wake up at 6am. The natalists were right all along, we all need our personal productivity baby.
As someone who did that sort of thing once, I suggest that training your body to sleep through alarms has potential for negative consequences that may be worth considering.
Some of you will feel this is obvious. Others will feel it is true but useless trivia. For me, this idea profoundly alters how I feel about humanity, both on an individual level and as a whole.
This is why I will never feel comfortable around committed ideologues, of any variety. No matter how much I agree with them on individual policies, ideologues can never take this forgiving, permissive view of the human character. And so I will never be comfortable rubbing shoulders with them, or supporting them. I'll always have that fear that by empowering them, I am empowering an unspoken ideology of judgment that could one day devour the world.
For those that may have read Scott's post on Janus's simulators or are otherwise interested in niche LLM phenomology and Cyborgism, the first full length public podcast with Janus has just been released.
Janus appears in full snowcrash-style avatar, filmed live in Portugal this year.
I keep expecting National Shrimpact Day to be "come to our fundraiser, attraction: all the shrimp you can eat banquet for those who buy the $100 ticket!" 😁
I wouldn't dismiss out of hand them actually doing something like that, with the shrimp being "humanely killed." I suspect you're reüsing heuristics you've developed about vegans, and I'm not sure those are really applicable to these guys.
You know, this seems like it could become a fabulous grifting opportunity, though maybe it would have been even better five years ago.
What *are* the preferences of creatures with brains the size of something very tiny? Well, the motivated interest groups claim they have feelings just like us:
Clearly, this is something that requires study. Lots and lots of study! Lots and lots of very expensive study with good salaries and benefits and opportunity to jet-set to global conferences on "is pink for shrimp or not? the species-specific colour preferences over thirty shrimp species" in agreeable locations with plenty of free booze and vegan food/all the shrimp you can eat banquets.
After all, how can we alleviate shrimp suffering and make sure shrimp have meaningful, worthwhile lives unless we are fully informed on the fears, dreams, hopes and desires of each liddle shrimpie? And how will we know the preferences of a shrimp unless we put in the work? And how can we put in the work unless the government puts its hand in its pocket and funds us?
Now that idpol is fading away and even Ibrahim X. Kendi got turfed out of his comfy academic sinecure, shrimp behavioural research may be the new woke!
I wasn't thinking of vegans as such, and if they *did* have the shrimp buffet of humanely killed happy shrimpies, my opinion of them would go up a lot 😊
On the topic of infighting between such groups, icymi, you might be amused by the fight over catering at the COP30 conference last month between those who wanted it to be entirely vegan and the indigenous communities who wanted to highlight their ancestral food.
Great Moments in Catering: Friend passed away, tragically young. Jewish funeral. We were seated with the rabbi, who looked at the catering and ruled that the provided ham was indeed Kosher. . .
Once, if I were indifferent between that place and other similar places; twice, if I liked it in particular for some reason.
My wife, meanwhile, won't eat at a place ever again if a friends says they got sick after eating there one time, and this has led to some small amount of strife over our doordash orders.
I'm an American. I've gotten food poisoning in Japan 3 times. Once from sushi, once from sashimi, once from food at a Sichuan restaurant. I've been back to 2 of 3.
The only time I got any kind of food poisoning in japan was on my very first full day after I moved to tokyo. Got a 7/11 microwavable yakisoba. It passed through me at a speed that should be studied.
Once, though it's not necessarily a rational choice. There's just so many options in most American cities, of roughly equal quality. The extra, literal, ugh factor of remembering the illness would weigh on the decisions, like it or not.
I've felt ill twice in memory, at least seriously enough to notice.
I once got food poisoning that started as mild nausea immediately before entering a restaurant, and progressed to severe food poisoning as I ate. Rationally, I know that it would take a wild coincidence for the restaurant to have been responsible, but my body still got physically ill on setting foot in there for over a year- on subsequent visits I would eat a bit of bread to be social. (The restaurant was a retirement home cafeteria that I visit regularly, so not ever eating there isn't really an option, and so I got to see how long this took to wear off. Otherwise it would have been an easy never again)
Depends on how many times I've gone there. If I go once a week, and I get food poisoning after a couple years, I might give it another shot, or even two shots. If I get food poisoning on the first trip there, I'll obviously never go back.
(With the mild caveat that if I have eaten at a place in restaurant for a long time and get food poisoning from the takeout, I am more likely to stop getting takeout than stop going to the restaurant, and the strong caveat that it has to be food poisoning - if I come down with a cold that’s not the restaurant’s fault.)
I have gotten food poisoning a single-digit number of times in my adult life, probably no more than five. I don’t remember ever getting it as a kid, but could be missing (or not have recognized) an incident or two. I think I do more of my own cooking than most people, but still eat out maybe an average of once a week, so it’s not that the situation could never come up.
I think I'd have to get food poisoning at least twice, possibly more before I actually noticed a pattern. (I don't think I've ever actually gotten food poisoning from a restaurant.)
Like, food poisoning takes time to develop, so I'm probably only going to know vaguely that it came from something I ate yesterday - and my first thought will probably be "I guess those leftovers were a bit too old" rather than "I bet the chipotle wasn't following food safety standards."
Yeah, I think it's pretty hard to know for sure where food poisoning came from unless you have, say, a big bunch of people who all had a single meal together and then all got sick about X hours later.
And yet, I got sick from (as best I could tell) an Indian restaurant once, and I have never been back. There's a lot of competition around and the chance of getting sick again isn't worth it.
Pedantic dickery answer; zero. I've stopped eating at most restaurants for other reasons.
Specifically stopping because of food poisoning; more than once, but probably twice. I stopped eating Jack In The Box for many years after their big contamination event, but went back eventually. But twice means I'm going to consider food poisoning a routine possibility, which means I'm eating elsewhere.
It also depends what you mean by food poisoning. I've continued to eat things that give me butt problems because I don't value those enough to give up the other end.
I get food poisoning on occasion, because I cook my own food and am bad at cooking, and/or keep leftovers around too long.
I don't value diarrhea particularly high, I'll put up with a lot more of that than vomiting or migraines. Typically I don't count something as food poisoning unless I'm stuck in bed because of it.
I can only think of one time I've had food poisoning from a restaurant, but I don't remember which restaurant it was. I think it was Chipotle. QDoba>>>>Chipotle, so I already don't ever choose to eat there anyway. Otherwise, it would probably have to be two or three times in a row for me to stop going to a place for awhile, if it's somewhere I've already been to multiple times. If it was my first time at a place and I got food poisoning, it might be hard to want to go back right away.
Are there any Thomas Sowell fans here? I just finished "A Conflict of Visions" And I found the book to be a deep insight into our differences. I tried to summarize the book, but AI does better, so here's the prompt. * Thomas Sowell "A Conflict of Visions", summarize. *
What Sowell book should I read next? (I've started "Basic Economics" and "The Vision of the Anointed", the second one seems derivative of what I've read... and is being cranky about the 'institutional class' )
(I know two posts in a row, I'm off of work and awaiting hernia (groin) surgery... so you'll have to put up with me for a little bit. :^)
Black Rednecks and White Liberals is worth reading. If I subtract for some repetition, partisanship, scientifically uncharitable views, and the common errors from sweeping pronouncements, Sowell is still on balance an intellectual treasure. He will help you think more broadly across global and historical dimensions before making politically expedient claims, and that is coming from someone raised as Progressive.
A big thing to notice about Sowell is that he does[1] his best work in long-form, slow writing. He wrote some newspaper columns for many years, and in general they were nothing special. I've seen him interviewed a few times, and he mostly didn't come off all that well.
[1] Did--I think he's in his late 80s now and probably isn't writing a lot more books.
I own _Basic Economics_ and _Advanced Economics_. Sheepishly, I confess I bought BE because at the time, I was literally looking for an introduction to economics (David Friedman's book _Price Theory_ is better for this), saw the good reviews, and found myself in what I would have to call partisan, but nevertheless very good, writing - enough to convince me to buy AE.
I find Sowell does eventually repeat himself - IIRC, both books contain a lengthy discussion of rent control - so I figure the more books of his I buy, the more chapters I'm likely to find saying the same thing in one of the other books. That said, I might benefit from getting ACoV.
Meanwhile, you might benefit just from finding videos of his interviews and other appearances online.
No group of people is all one type of the other. And there will be parts of both visions in many ideas. But Maga is clearly closer to the constrained view. (vs. unconstrained) with it's call to traditional beliefs and mores.
Yeah, I agree, but it does seem to me that another factor is the capacity of MAGA folk to believe exceptionally stoopit bullshit -- immigrants are eating our kitties in the back yard, etc., things for which a thought experiment and consideration of what one has observed so far should be enough to make implausible. The Woke Folk also have ridiculous ideas -- a kiss without permission is a little rape -- but the reason wokies believe them isn't that they are too dumb and uneducated to realize they're nonsense, it's a kind of hysterical, semi-religious scrupulosity.
I think the common point in MAGA and wokism is that both have evolved in the world of an online media environment that makes it easy to police the expressed views of people in your own coalition. And that makes it easy to have silly or crazy beliefs be part of your doctrine, because you've got great machinery to silence dissent just lying around waiting to be used.
This also makes the beliefs rather unstable--whatever the current crazy bit of doctrine is, it is like 10% sincerely held and 90% expressed to keep the peace or because the speaker has never really heard any good arguments against it. That can flip very quickly when the preference falsification cascade collapses.
Wouldn't you agree that MAGA beliefs are at least directionally correct (well, with the exception of tariffs) even if their reasoning is terrible? Unchecked immigration - particularly of the Biden-era kind - is objectively bad, so it's at least a reasonable impulse to be opposed to it. Woke beliefs, on the other hand, don't even have a seed of reality in them. They're little more than a fig leaf for their dislike of traditional American culture. In my view they actually care little for the welfare of blacks or transies, they just want an excuse to paint America as an oppressive state in need of reform because that gives them power. It's the same dynamic as the Cultural Revolution.
Well, I don't think MAGA beliefs about vaccines are directionally correct. As for Woke beliefs, yeah they're a bid for power, but that's not all. I've spoken with a lot of woke students at the local Ivy, and they sound a lot like people who have a case of religious scrupulosity (a form of OCD I have actually treated.). Most are quite disturbed by some of the non-woke feelings they can't keep from being aware of, eg fear and distaste for certain minorities, horny politically incorrect sexual fantasies. Also most are quite scared of being canceled by their peer group because their politically incorrect side is revealed.
Oh agreed, vaccines are another one. Though I'd say that it comes from a place of libertarian freedom and not wanting the government to have the ability to dictate anyone's medical choices, which fair enough. I disagree with the position but at least understand the impulse. It's kind of the right wing version of "my body my choice".
Very interesting to hear about woke anxiety. How do you council those people? Are you able to get them to see the unreasonably puritanical nature of woke?
I don't understand what you were getting at. My point was that in the absence of more info some consideration of what one has seen in the neighborhood, plus some thought experiments would have made it seem implausible to most poeple that immigrants were eating the neighborhood's pet cats. (Is there even much edible flesh on a cat? Has anyone found fur and bones? Have a lot of cats gone missing? Have I seen skinny, ravenous foreign-looking people wandering through my neightborhood?) I don't know what happened at the US embassy in Haiti,but don't see how giving food to Haitian kids is a MAGA bullshit rumor the way they're-eating-our-cats-is.
Right, for a book that attempts to explain the "ideological origins of political struggles", I'm not sure that it says much for our time where a reactionary punish-your-enemies / reward-your-friends populism is the dominant political force in the country. I read it years ago and recall thinking that the visions that Sowell identifies are relatively small threads in contemporary politics and certainly not clearly aligned with any political movements or parties in the Western world.
Yes it's not great but this is the future that got locked in when the Left went all-in on identity politics. MAGA is a rather predictable extension of that mentality to the Right. Sowell came of age when there was still a sense of common American identity and it was possible to have higher-minded ideology than simplistic ingroup/outgroup dynamics. Sadly that's not possible anymore. The steamroller of identitarianism has flattened all of it.
The probable reason people don't directly address that argument about transgender sports: it is such a tendentious argument that it feels like a rhetorical slap in the face. I'm not exaggerating when I say that every time I hear this argument, it feels like an insult to my intelligence. I know it's not intended that way, but that's what it feels like.
For a comparison, few Darwinians are willing to argue the fine points of creationism. Perhaps there's a sense that if you are willing to take creationism that seriously, you've already given the creationists a kind of victory.
In my view this is an immoral attitude to take, for it cedes the intellectual high ground to social fashion. I feel that the right to have a firm opinion on anything comes with a responsibility to vigorously defend it in good-faith debate. The attitude you've expressed encapsulates the clever sleight-of-hand the Left has played on the culture over the past few decades: they hold their views to be self-evident and, because they occupy the taste-making institutions, they've been able to successfully taboo any rejection of them as low-status. That's led to their current emperors-new-clothes position w/r/t gender and race and our inability to have reasonable public debate over contentious issues.
I love nothing better than sparring with Creationists. If someone doesn't understand science well enough to do that then IMO they don't have the right to consider themselves educated.
You have to triage which arguments you were willing to engage with. Do you really want to argue with people who believe chattel slavery should be restored? This is not a straw man, I have met people in real life who believe this.
Do you really want to spend your time debating with people who consider Stalin to be the greatest European leader of all time? Someone like that has such insanely different priors that arguing with them would surely to be an uphill battle. It might be pointless.
Do you want to spend your time arguing with someone who believes that life is meaningless and all pregnancies should be aborted? Because these people are among us as well. Please consider the implications of your commitment to debate.
Sure, those are all reasonable pragmatic considerations - we can't spend our lives doing nothing but debate. But it's important to remember that even those reasonable considerations can be applied in bad faith. It's easy to convince yourself that something isn't worth your time when the reality is that you're not sure you can win the argument. I think the right way to handle is that the degree to which a belief is important to you has to be earned by your willingness to defend it. That stops the whole "I don't care enough about my belief to defend it but I care enough to shun you for disbelieving it." That's the sort of behavior that I think is bad epistemic hygiene.
Addressing this argument in good faith, trans women shouldn't be in women's sports for the same reason able-bodied people shouldn't compete in events for disabled people. This biologically disadvantaged group was intentionally given a venue.
I think that in cases where trans women have a clear advantage. But when their hormones take them to the same level as cis women, does it really matter that their disadvantage is due to hormones they took instead of ones they produced naturally?
My personal opinion is that sports are inherently unfair and only really useful as entertainment, so we should stop worrying about whether it's more unfair to make cis women compete against trans women or bar trans women entirely and just let people make the leagues viewers want to watch.
I'm not personally a sports guy, but I can think of at least 2 reasons to want non-competitive players to be interested:
- It's great exercise, and games are an easy way to get children specifically to be active
- It's a great social activity (both for children and adults), and we don't have many easy means of developing social groups nowadays
Having a group of people with similar characteristics to yourself as visible sports icons is motivating, so you might want to artificially keep some around if you like the above mentioned benefits.
If you want to say "sports should just work like the rest of the entertainment economy", that's a much bigger position than "trans women should be allowed in women's sports". I think that would also imply:
- Colleges should stop offering women's sports if they don't "earn their keep" (which they are currently required to do under title IX)
- Colleges shouldn't offer most sports (almost all NCAA sports except football and men’s basketball lose money)
- You should cancel the Paralympics (they usually don't make money)
- You should cancel the Olympics (usually loses money)
If you agree that there's a social utility in subsidizing / mandating certain types of sports, then you need to think about how to maximize that (maybe most people should be able to fit into a group with a "professional" team to be role models?).
If there's no social utility, then letting pure market forces operate makes sense IMO
I don't think there's social utility in subsidizing sports. If you want to do the Olympics, you can. But you shouldn't make other people pay for it. Why do people take a sports league so seriously?
I definitely think there is net social utility in encouraging physical activity and socializing with strangers. There may be better ways to do that, but off the top of my head I can't think of many better than organized physical play.
In middle and high school I got most of my exercise and met many of my friends through sports teams.
Also, as an adult, a ton of the social organizations I know about are amateur soccer groups and things like that.
That makes it sound like I don't want women's sports specifically to exist. I just don't think people should be forced to pay for sports. If they want to watch it, then sure, they can pay for it, but why force other people to subsidize sports they don't want to watch?
I'm not saying you thought it through and decided this was the best way to oppose women doing sports without social censure, I'm just saying that taking this attitude is tantamount to wishing away women's sports in practice, because nobody (including women) likes to watch any woman's sport that isn't just veiled softcore porn. Whether you're fine with it or not, you should be aware that the practical consequence of your position is to baleet women's sports.
It's hard for me to overstate how frustrated I am about the shrimp welfare movement.
Animal advocacy is a zero-sum game, and as long as there are mammals out there suffering, the shrimp welfare movement seems like a slap in the face. I hate it so much. This is not harm reduction, this is contrarianism at its worse. Some might even call it perversity.
Ironically, I go out of my way to avoid killing ants when I am outdoors. I (slightly) resent it when people kill insects in the great outdoors without an articulable, defensible reason.
But I don't see this attitude as an appropriate basis for a political or philanthropic project.
I can think of a number of ways to prioritise animal welfare causes.
1. Start with the most intelligent animals (whose suffering is presumably worth more) and work your way down, so your first concern (apart from humans) should be ensuring the humane treatment of apes.
2. Start with the most numerous badly-treated animals, on the assumption that a million suffering invertebrates add up to one suffering chimpanzee. This is where the shrimp people are coming from.
3. Start with the most egregious abuses and work your way down. If you think about the most egregious and needless things that people do to animals, then "crushing its eyes to make its ovaries develop because the overcrowded and unsanitary conditions that you keep it in prevents normal ovary development" is pretty high on the list. (Note: if you're about to reply with a list of things that are even worse then I don't want to hear them.) So the shrimp people kinda have a point here as well.
Or you can prioritise along all these axes simultaneously, which will probably give you a fairly reasonable set of priorities.
Where the shrimp welfare people are really going wrong is in trying to sell this as a pure utilitarianism play. Most people aren't pure utilitarians and react badly to people trying to sell them something based on pure utilitarianism; if you want to convince people to care about the treatment of farmed shrimp then you're better off using the sort of combined vibes-based ethics that people tend to work with in practice.
I completely disagree with the statement that animal advocacy is a zero-sum game. It's additive. Welfare concerns for one animal group often benefit another animal group.
I used to think or hope that the fact that people cared about dogs, to such an extreme degree, would transmogrify into caring about wildlife generally. But that turned out to be fanciful.
We teamed up with the dogs to exert our joint dominion over all the other wildlife on the planet. We have a different arrangement with cats and horses. Nothing else is in remotely the same category.
We may still wind up with some level of concern for various other critters, and that speaks well of us I think, but it's not going to come from the same place as Man's traditional bond with his Best Friend.
It is a very useful reductio. The argument is pretty sound from its own internal values. And in fact it doesn't even stop at the shrimp -- many of the shrimp people earnestly believe the best thing to reduce suffering in the world is to turn natural land into parking lots, and especially to prevent the creation of more parks and natural preserves. Why? To prevent insect suffering!
I feel the same way about animal welfare. When people spend thousands of dollars on veterinary bills, I see this as anti-human and a demonstration of great callousness towards all of the children in the world who are suffering.
UNLESS the animals that need care are vital to your livelihood, of course. And in some cases, I suppose that putting down a sick animal might cause the owner to spiral into suicidal depression, which must be avoided.
But in ordinary circumstances, very sick animals should be allowed to die and the money saved on veterinary bills should be spent in a more responsible fashion.
I mean, I have a cat, whom I love, and would pay expensive vet bills for, but I also pay the Giving What We Can Pledge of 10% of my income. 25% of that pledge is directed to animal welfare. My altruism does not need to be unidirectional.
Sure. I'm not careful enough with my words. I said callousness but really what I should have said was carelessness. And I am not such a rigid virtue ethicist that I would make this a litmus test on anybody's character.
My first responsibility is to my immediate family, and then to my extended family, and then to my super extended family, and then to community members that aren't related to me.
Humans are all part of the human species, which is analogous to our immediate family. Mammals our analogous to our extended family, vertebrates are analogous to our super extended family, and shrimp are analogous to community members that aren't related to you.
I realize this is all very alien to Anglo-American culture, and maybe even Northwest European Culture. I realize people from these cultures reject the idea of owing a debt to those who share your blood, and I am not going to convince anybody who wasn't ready to be convinced.
But the truth is the truth and your obligations exist whether you recognize them or not. I am grateful I was born into a culture where I was gifted with insight into my responsibilities to my family and to humanity. What a tragic thing it would be if I didn't understand these responsibilities. It would be like if I had never learned to wipe my ass.
Scott donated a freaking kidney so I guess he's done more than enough for humanity. But I'm not healthy or brave enough to do that so I hope I am able to give back in other ways. Like denigrating shrimp welfare activism; in small part, maybe that helps fulfill my obligations.
From my pretty shallow understanding of history, that’s exactly how Western thought used to go: first, your family, then distant relatives and community members, then your ethnic community of third and fourth and fifth cousins. But at some point in the 1800s, this led to the idea that every nation deserves to be self-governed---which in turn led to conflict where to draw borders, ethnic violence, and ultimately WWII. I suspect that western propaganda very intentionally denigrated familial-piety ethnic-duty cluster of ideas to avoid a repeat.
Non-European cultures by comparison didn't accumulate so much baggage around blood-bound duty. So it's totally acceptable to be openly ethnocentric. But as far as this being "the truth" and owing duty in proportion to your blood reatedness---I dunno. I suppose it depends on the ethical system to which you subscribe. You don't think any less of Scott for donating the kidney to a complete stranger rather than a fourth cousin; perhaps even admire Scott more for doing so completely selflessly vs. "helping family."
Taken to the extreme, your duty to a rat in your house cannot be less than to your houseplant.
I could have mentioned that my family is multiethnic and my neighbors are multiethnic. But I shouldn't have had to. After all, this is 2025. Our communities are FULL of multiethnic families and neighborhoods.
You should have considered this possibility. You should have then concluded that the priorities I initially laid out do NOT, in fact, justify ethnocentrism.
BTW, I didn't notice any golden age of peace and amity resulting from the deprioritization of family ties. Did I miss it?
Funny you should mention that, these values I'm talking about are closer to Confucianism than Northern European values. I think Northern European values are uniquely "frigid," if you'll excuse the pun, and uniquely atomistic, individualistic and vulnerable to being instrumentalized by misanthropes.
I’ll subject myself to further ridicule, by again encouraging you to pony up the $20 - $25 to amazon prime to rent or buy the documentary “Age of Disclosure” produced by Dan Farah. The premise of the documentary is that Aliens are here, the US government (a small part of it) has the proof, alien craft and bodies. And that the way to get at the truth is for our congress to demand answers. Even if you put this at a 1% likely hood of being true, if it is true it is undoubtedly the most important fact that remains widely unknown. How would your life change if Aliens are here? So don’t read any reviews, ignore those voices trying to convince you that it’s all bogus, put on your tin foil cap, grab your tesseract full of inter dimensional popcorn, and join the ‘Age of Disclosure’, https://www.youtube.com/watch?v=SXUEcfgZv70
Here's what I don't understand about the claim that the US government is covering up the existence of aliens: what the fuck are all the other governments doing? There are, what, 190-odd nations on the planet? You're telling me *none* of these people have blabbed, they're all playing along, including our great rivals? Why would, I don't know, Uganda not just give a press conference and show off their crashed UFO? Are they that loyal to US military-political aims? The UK just happens to be exactly equally devoted as the US government – every US government ever, natch, for all their differences - to clamming up about their reams of UFO sightings, so they create the precise same level of coverup? Ireland doesn't want to ask the Vatican how to interpret their alien corpses in orthodox terms? Or are the aliens only visiting the US? If so, why? Pace a level of smug self-importance that comes with being the sole superpower on the planet, is the US *really* that much more important that aliens are just like "you know what, fuck scanning China. Even an American cow is more relevant to abduct"?
According the the documentary both the Russians and the Chinese are also gathering info from downed UFO's. IDK why they don't share, but there is something to, "tell your friends and your enemies will also learn about it."
The relevant probability isn't p(aliens are here) it's the joint probability that aliens are here, *and* that Dan Farah has something to say about it that is simultaneously true, important, and new to me, *and* that there won't be someone else saying the same thing in a more accessible format in the near future.
As someone who has been following the UFO debate for many many years, and who finds text >>> video for this sort of thing, I assess that joint probability as too small to justify the time it would take to watch a documentary even if it were free. But a writeup summarizing his take with pointers to further investigation might be worthwhile.
Hmm I can try a write up, but my english sucks, I'm just not good at it. You might do better from an LLM. So I think Dan started the documentary several years ago. The origin of the story starts with the AATIP (Cia task force to sturdy UAPs) And the role played by Luis Elizondo. The people in AATIP were frustrated by pushback from higher ups in the CIA in particular from people supposedly involved in the Legacy program. (Some CIA/ air force/ Dept energy/ industry group that has been around since the 1950's (40's ?) collecting stuff from alien crash sites.*) They come to the conclusion that the only way to break this log jam is for one of them to resign their position and they will then be able to go the the media. The media will hopefully get congress interested, which will then generate more media interest which may get the congress to force the issue and have the Legacy program divulge what they know. If you've been following the UFO story, then you know that much of this has gone on... increased media and congressional interest. (Heck we've even got the Air Force to admit to seeing UFO's**.) We are stuck at the last step. We need congress to pass the 'UAP Disclosure Act.' The documentary tells this story much better than I have. But there is no real new information. (A lot was new to me.) There are many interviews with government officials, cuts from congressional testimony and media coverage.
*I'm reminded of X-com defense a video game I played in the 90's
** For many years there were reports from Navy and other pilots about UFO's but nothing from the Air Force.
I don't know why this immediately made a bell ring in my head, the name just sounds familiar for some reason (and not a good one). I'll come back to that.
Problem is, George, I (and others) have been hearing the same old song for decades by now. Every so often, new REVELATIONS. Every so often NO THIS TIME INSIDERS, REALLY INSIDE INSIDERS, TELL ALL. Every so often the same old tinfoil stories. And it never goes anywhere. Like I said, I've been reading this in the Fortean Times for issues upon issues. If there were genuinely any real alien materials and/or bodies and/or live aliens making contact, by now we would have something public and tangible, not shaky videos from the 50s and lots of mimeographs of "[redacted] [redacted] [redacted] swamp gas" papers.
Why should I blow $25 or its equivalent in my local currency on what is just one more in the "What the Bleep" genre of fake science tinged with vaguely sinister conspiracy murmurings? The X-Files mined this seam dry in the 90s, let it lie.
Okay, back to Elizondo. Ah yeah, this is why I had vague "oh crap this guy" reaction:
What you've basically said here is that there's this organization that's very interested in what another organization is doing, and they're unable to find out, and they're frustrated. Reading the paragraph above, this is what I find it boils down to.
But this is not unusual; it's extremely routine.
A straightforward way to approach evidence in favor of some unlikely claim (e.g. aliens are here) is to look at the evidence, conduct a search for alternate explanations, and consider whether any of those explanations are more likely than "aliens are here". One possible explanation immediately sprang up for me:
AATIP was funded in order to look into evidence of aliens, just on the off chance they really are here. They happen upon something they call the "Legacy Program" (LP). The LP is a Special Access Program (a highly classified project) staffed by CIA and USAF, and tasked to perform duties that are kept secret from anyone unless they are authorized to view classified information and have a need to know what the LP is doing (which means they work in it, make decisions about who will work in it, or allocate funding to it). We don't know exactly what LP does, but we can infer from the staffing that it's probably related to both aircraft and intelligence, including a great of information about our tech that we Americans don't want other nations knowing about. It *could* be related to aliens, but it doesn't have to be.
AATIP wants to know what LP was working on. Higher-ups in the CIA - specifically, people who decide who works in it, and therefore need to know what LP is doing - happen to know that LP doesn't have any proof of aliens at all. One of the higher-ups (let's call him Ashley) informs AATIP that LP doesn't have anything AATIP needs to know about. AATIP wants proof of that. Ashley tells AATIP there's no way to prove that to AATIP's satisfaction without giving AATIP a full tour of all of LP's assets, and AATIP has no NTK that (and Ashley leaves out his suspicion that AATIP is very motivated to leak whatever LP *is* doing, which is classified for good reasons), so it's not happening. AATIP is frustrated, and tries for a media angle, because Ashley is right that they are very motivated. In order to make that angle work, they make a lot of claims that suggest LP is working on or with aliens, but doesn't flatly state that (because lying like that would give Ashley a solid reason to shut AATIP down).
I'm betting a careful reading of that documentary's transcript will be consistent with the above two paragraphs.
Yes, this is a vey fine summary, thankyou. And yes we only have hear-say. And right they are clearly keeping it secret (whatever it is.) for a reason. A reason could be they don't want to freak out the public. About the NTN, apparently some presidents and senators aren't on the NTN list and that bothers them. According the the doc, Harry Reid was pissed enough that he started the AATIP. So it is like one piece of the government fighting another piece.
I listened to most of Dan Farah's recent appearance on Joe Rogan's podcast. I'm highly skeptical, but apparently he got a bunch of pretty high level people like Marco Rubio to talk on camera about this stuff. I guess for me, then, the interesting question is what ulterior motive do these people have in pushing these alien visitation stories. The old theory was they wanted to deflect attention from experimental aircraft test flights the Air Force was doing during the Cold War. Seems like there's both a new level of specificity and higher-profile people involved these days in this theoretic disinformation campaign, though, so maybe that just means the US military has some new spiffy drone technology they want to test. I am of course open to other theories.
Is it Marco Rubio saying "Yeah, aliens are real and we are totally hiding that" or Marco Rubio saying "There's some stuff regarding unidentified aerial phenomena that we're not sure about" cut and interspersed with other material in a way that makes it look like Rubio is saying something more than what he's saying?
Boy I'd have to watch it again. But recall him saying something like, there's this one piece of government doing something without any oversight and we the congress want at least some of our people (senators) to know about it. I think he was part of a classified briefing... which of course is classified.
I am thinking that it started when we heard rumors the Soviets were doing remote viewing. We ginned up something for them to focus on, to see if there would be a feedback loop and they'd actually find something, or else report the Americans were hoaxing the world; and if remote viewing was bogus, they'd just waste resources on it. Also, it was a bit of a tit-for-tat move, just in case the Soviet remote-viewing effort turned out to be purely the same kind of show for us to waste energy on.
Note: just my theory, based on reading up on how people play these sorts of games and get lost inside them in personal interactions.
I like Rogan's take on Bigfoot, "You know who never sees Bigfoot? Hunters," pointing out that the legal requirement to positively identify targets clears up nearly all of the misidentifications.
I suspect the same is true of high-hours pilots, especially ones flying armed aircraft.
We are ignoring it because it is not freakin' true. You can have your hobbyhorse, George, but a point comes when nobody cares about what's going on.
The day I start trying to convince people on here about Eucharistic miracles, somebody gently force the keyboard from my hands and tell me to have a nice sit down for a bit. And I believe in the Eucharist, but that does not mean that you all have to be forced to listen to me rant on about it and why you totally should all watch this Youtube video about the recorded, verified, credible witnesses speak on the record, about the long history of Eucharistic miracles HAPPENING IN A CHURCH NEAR YOU RIGHT NOW, WHY DO THE PROTESTANT HERETICS WANT TO COVER THIS UP????
I appreciate the prompt. My thought from analysing this matter in light of thought patterns I learned on this site:
1. It seems a low likelihood that aliens are observable to us when we have so many cameras pointed in every direction now, more all the time, and still all we get are blurry images.
2. The usual "aliens wouldn't be so incompetent they'd crash on our planet and let us pick apart their ships" one.
3. I love ideas about the Alcubierre drive and the idea of other inhabited planets, but the vastness of space and the extreme difficulty of anything FTL, even merely fast interstellar travel, just crushes it for me.
That out of the way, I'll continue to strain credibility with my pet theory:
The one industry that is sure to flatline if aliens were disclosed is the science fiction genre. So, perhaps the aliens are SF fans, and love our movies and books in the same attitude with which we'd consume steampunk or other alternatve-history content.
My position is: do aliens exist? Very probably, it's a really big universe out there. Are they crashing all over our planet? Not unless it's the equivalent of drunk idiot teenagers joyriding. I used to want to believe this, that there was some grain of truth to all the sightings, but over time I've dropped any credibility I gave it. I'll have to get my aliens from SF like everyone else.
Years back I read Jung's book on UFOs and I think he came nearest to what is going on: people used to see the fairies, and angels, and in our modern scientific era we now see UFOs and little green men. It's all the same psychological mechanism in action:
"We have here a golden opportunity of seeing how a legend is formed, and how in a dark and difficult time for humanity a miraculous tale grows up of an attempted intervention by extra-terrestrial ‘heavenly’ powers—and this at the very time when human fantasy is seriously considering the possibility of space travel and of visiting or even invading other planets.
Well we had all the "Drones over New Jersey" stories that happened about this time last year. There have been hundreds of Air Force / Navy UFO sightings reported in the last year. And no, we will need Sci-fi more than ever. Who else has been thinking about this kind of thing.
I expect the chance of the specific scenario the conspiricy theories are on about to be extremely small. Much smaller than 1%. You can't just make up a large complicated pile of nonsense and act as if it's probability isn't exponentially small. Every extra bit of detail halves the probability.
The specific "alien conspiricy theories" have a lot of parts that don't really make sense.
Aliens, fair enough. Aliens are quite plausible. Intelligent technological aliens, still not implausible. But I would expect evidence for this to be seeing dyson spheres through our telescopes, or receiving a radio message.
If there were alien spaceships on earth. Well breaking from interstellar speeds seems like it would involve a massive rocket plume. Again, not craft and bodies.
And mind uploading or similar seems like it should be easier than interstellar travel. And we already mostly use unmanned space probes. So the idea of a crashed probe leaving "bodies" is dubious.
And why would it crash. Surely the alien spacecraft should be reliable. And if it does crash, well nanotech should be extremely good at self repair.
Again, why a couple of bodies in a flying saucer? Why not showing up in a massive, obvious and undeniable way. Or successfully hiding from our crude tech if they wanted to hide.
Let me guess, these "bodies" are grey (or green) humanoids that fit every "scifi alien" cliche.
And then, why would the american government be hiding this?
When you add in all the specifics, this sounds very unlikely to be true.
But this also sounds exactly like the sort of thing that a lot of people want to be true. There is plenty of UFO hype about videos that seem to be birds. Plenty of similar nonsense surrounding bigfoot etc. In short, you would expect something like this documentary to exist, in a world without aliens. Was anything about the documentary unexpectedly strong. Ie evidence that's stronger than you would expect from a bunch of cherry picking and lying UFO enthusiasts in a world without aliens.
Another possibility is that there's something about the interstellar environment that makes it prohibitively hard for even von Neumann machines to survive a trip to another star, let alone with their self-replication mechanisms intact.
Yet another possibility is that artificial self-replication is itself harder than estimated (bear in mind, we've yet to do that in a controlled environment -here-).
I wasn't thinking of dark matter; rather, plain old radioactive particles, ions, and even just ordinary matter, i.e. dust and ice. Interstellar space is no doubt a near-vacuum, but there's a *lot* of it for a probe to have to travel through, and at a reasonable fraction of lightspeed for a probe's presence on Earth to be a question worth considering. Which means that probe will probably need a way of dealing with anything it collides with en route, even if it's tiny.
Why is the default for skeptics "1%" rather than "0.00001%"?
If you want to make the argument that it is worth someone's time, then give the best piece of evidence from the documentary, preferably with the sources (one would hope the source isn't just "the documentary").
You have to put your own numbers on it. There is no smoking gun in the documentary. Mostly just people wagging jaws. But you can ask why all the jaw wagging.
...must I? People wag jaws on the internet for millions of hours, about much less interesting topics. I have no problem coming up with a boring answer for "why all the jaw wagging".
Most people don’t understand what extreme probabilities are - both that 1% isn’t that low, and that there are plenty of things below .00001% (like the probability that the next plane I am on will crash while I am on it).
That probability is below the noise floor. People hallucinate at higher rates than that. Whatever mental process it is that spits out such a tiny number, the likelihood that one went wrong in one's reasoning and/or assumptions, or is simply failing to perceive reality correctly, will be higher than this.
Once you've rounded it off to "not worth spending any more time on this", there's no point trying for more precision.
You can have coherent estimates of a probability less than 1/10M. For example, I believe the probability of an asteroid on the same order of magnitude as the one causing the chicxulub crater hitting earth in a given year would be less than 1/10M based on the observations we have.
It's 1 in 10 million. While "people" might hallucinate at higher rates than that, *I* don't hallucinate one out of every 10 million things I see.... (that I know of).
>Whatever mental process it is that spits out such a tiny number, the likelihood that one went wrong in one's reasoning and/or assumptions, or is simply failing to perceive reality correctly, will be higher than this.
Ok, so let's say the minimum is actually .01% (or whatever value you choose, so long as it is above the floor you quote), can I offer you a bet at those odds versus my ~0% odds about, say, lizard men ruling the world behind a shadow government, or the moon being made of cheese, or the world being surrounded by a celestial crystal sphere created by Uriel?
Clearly, there is a point at which the claim is so ridiculous that you *have* to go to less than 1 in 10 million.
Is the alien thing actually that low? Not sure, really. It probably depends on the specifics, but my point is that it is rhetoric to say "Even if your probability is 1%" because it implies that 1% is some absurdly low number for a probability for something like this, and it simply isn't.
> can I offer you a bet at those odds versus my ~0% odds
No, I'll take no bets on random tiny numbers I can't reason about the relative magnitudes of with any confidence. I'm completely happy to file the lizard men, cheese moon and celestial crystal sphere all in the same mental bucket. I really don't think there's any merit trying to separate them from each other.
If you believe you know better, I'm completely happy to bet money that none of these three things are true against your proposal that one or more are.
There's a big difference between saying "a human being cannot reason about numbers so tiny" and, essentially "that number is below the noise floor, so you aren't allowed to use it".
I agree noise floors are useful concepts. I disagree with what I *thought* you were saying, which was "therefore you should never assign something probabilities below the noise floor". But it looks like, here, you're saying something more like "Since I'm a human being, I can just round all this down to zero for practical purposes until there's some big, earth-shattering evidence that brings it into the realm of being remotely possible"
If so, I agree. I still disagree with OP that "even 1%" is anything other than rhetoric designed to make 1% look reasonable a priori.
>If you believe you know better, I'm completely happy to bet money that none of these three things are true against your proposal that one or more are.
As a side note, my point was that I would bet *lower* odds than you (since you were arguing against using such low probabilities), so you'd have to pay out to me when they didn't come to pass.
Yup, I think we're on the same page. For me, things like "0.00001%" in casual conversation trigger the same twitch as seeing a number with too many significant figures cited in a newspaper article.
What's true is already true. Admitting it doesn't change this. We know we can live in the world where the true thing is true, because we are already living in that world.
My life would not change if aliens are secretly here, because if aliens are secretly here they are already secretly here and that is the world my life, the way it already is, is already in. The only thing that would change if I bought the "documentary" is that I would be down $20-$25.
Now, if the president were to, oh, announce a trade treaty with the aliens, certainly that might translate to tangible effect on my life, just as any foreign treaty might. Or if they declared war on us, my life would change, just as it would if a foreign nation went to war against us. But I really don't see how me buying this documentary will accomplish anything other than moving $20-$25 from me to its creator.
How was your life changed by watching the documentary?
Yeah, it's worth calculating some kind of payoff matrix here.
What's the maximum benefit to believing that the US Government has a shed full of aliens? Well, if it is later revealed that they do, then I get to say "I told you so", but nobody will be that interested.
What's the maximum downside of believing that the US Government has a shed full of aliens? Well, if they don't, or if they do but it's never proven, then I wind up looking like a kooky weirdo.
The upside of believing in flying saucers is small even if flying saucers are proven exist, and the downside is significant *unless* flying saucers are proven to exist.
Yup, I'm wearing my tin foil hat proudly. :^) (well mine's made from aluminum foil.) For me, knowing Aliens are real and here would be the coolest thing ever. (and maybe the scariest too.)
I'm surprised to see that sort of attitude here, doesn't this argument apply equally well to most of the articles on this substack?
Don't get me wrong, I'm not sold on the documentary myself, but if I thought it had a reasonable chance of being true I would definitely want to know, even if it was just for my own curiosity and interest in truth-seeking.
Thanks. I find it weird also. The lack of any interest. But I chalk it up to a long standing campaign to discredit all UFO reports. (One could ask why we have that effort and where it comes from.)
How's my life changed? My kids think I'm a tin foil hat wearing old crank... but they mostly thought that before too. Some other minor financial things. I will freely admit that I like the idea of aliens. And I have like a million and one questions. But first we need congress to shake out the truth... which will be blocked.
> How's my life changed? My kids think I'm a tin foil hat wearing old crank... but they mostly thought that before too. Some other minor financial things.
Grin, I'm not trying to be funny. So part of the problem (if this is true) is how do you release it? Grasp the nettle firmly, (all at once), or in little drips and drabs.
I'm expecting drips and drabs, so at least a year... but maybe even longer... What would you do?
1% chance it's true, 99% chance it's not. Tell me where that is wrong, because I'm sure somebody right now is going "noooo, that's not how probability works!"
If aliens are here, then they're not very good at being alien, are they? Crashing their ships and letting their dead be found by the primitives. If they are doing anything, it seems to be weird experiments on cattle and humans. No sign of proper first contact.
I wouldn't be adverse to the notion of alien anthropologists coming here to observe a pre-spaceflight species in the wild, but the rest of it is just unconvincing.
How would my life change if aliens were here? Not at all, it would seem, since the proposal here is that aliens do exist and are here right now. If that is so, then things would continue on as they are. No, aliens exist and are here and we can prove it and Congress released footage of the crashed ships? Eh, again, things would continue on as they are. World governments would squabble over who has the tech, what tech is it, should it be shared, etc. but unless the aliens land in capital cities of the world and announce themselves, it won't change much.
If it's true and aliens with interstellar travel capacity exist and are here, then we are in the position of those New Guinea tribes making (or avoiding) contact with the outside world. We would be so far behind that the notion of capturing UFOs or controlling that tech would be like the cargo cultists trying to create landing towers and runways out of palm trees and bamboo.
So the question would be "how did the lives of natives changed when a higher tech civilisation made contact?" and following on from that "who is the dominant culture, our native primitive global one, or the aliens?" Meiji era Japan where we have to adapt or else, because like it or lump it, the black gunboats are in our harbours. Except we don't yet have the alien black ships in our harbours, so that's the big difference.
>Meiji era Japan where we have to adapt or else, because like it or lump it, the black gunboats are in our harbours.
I like this example because, while the Perry Expedition to Japan wasn't a true first contact, it's close enough to one for practical purposes and moreover, it happened in modern times with large literate cultures on both sides of the contact. The Autobiography of Fukuzawa Yukichi was a review contest finalist here last year. It's the memoirs of a man from a minor samurai family who had just started studying the Dutch language (Japan had very limited and restricted contact with Portuguese and Dutch traders prior to the Perry expedition) when the black ships showed up. Fukuzawa was one of the first Japanese people to learn English, participated as a translator in Japan's first diplomatic expeditions to North America and Europe, and had a career as a teacher and education reformer after returning to Japan.
I've read it and noticed a tons of parallels between Fukuzawa's experiences and the kinds of scenarios that are commonly explored in SF books and movies that explore the aftermath of first contact, but in a context of "this is what actually happened, from the perspective of a well-placed member of the culture who got contacted".
Yeah! So I'm hoping we've all read enough sci -fi so that we have a chance of surviving first contact. And won't form a cargo cult. Why they are crashing is an important question. (I would guess that those in the know have some good ideas.) So first a 'fact' (piece of info that may be true) from the documentary. There are at least two 'species'/types of aliens. So maybe they're at war and shooting each other down? I know not a very fun picture. So on a hopeful note, it could unite mankind. We have to stop all the stupid wasteful fighting and get busy.
Finally I just encourage other people to watch the doc, so we can talk about it with a little more information. (And yes no one has told me shit yet. It's mostly jaw waving.)
It seems very, very unlikely that the majority of UFO reports are due to aliens.
More plausible (still unlikely): nearly all UFO sightings have prosaic explanations: weather balloons, the planet Venus, people just making it up. But also, there really are aliens here, who are much harder to observe.
I just want to say, that from the point of view of the military industrial complex that has control of this info, you are saying exactly what they want you to say. A stigma against UFO's is built into the narrative. (I know you see me as a crank with a tin foil cap. no worries I embrace the look.)
Edit: Oh most of the sightings talked about are armed forces personnel (pilots) and the only data released are distant video shots and radar images.
AI answer to how many reported sighting by armed forces.
"The number of reported UAP sightings by U.S. armed forces has increased significantly, from 144 between 2004 and 2021 to 757 reports received between May 1, 2023, and June 1, 2024. The surge is attributed to less stigma and more reporting protocols, not necessarily more actual events. The Pentagon's All-domain Anomaly Resolution Office (AARO) is now responsible for analyzing these reports."
I have another pet theory that there exists "some kind of phenomenon" that appears to take on the form of most/some of the people looking at it, coming from their beliefs about the world. This could be easily defined as crowd hypnotism, but I'm saying there might be a very rare physical phenomenon that has its own reality external to shared human delusions. So, in older days it showed up as angels, and today UFO's.
Yes. And given the number of personnel and sensors, we should expect to see lots of unidentifiable blobs in the distance.
Tiny blob of something flying in the distance that could be a bird, or a balloon, or a secret drone.
And if 80% of these detections are blurry birds, why can't 100% of them be. It's clear that normal non-alien phenomena can generate UFO sightings. So the shear number of UFO's isn't evidence for aliens existing.
What you need is something that's close to the cameras, so you can see the details, and know it's alien.
I don't think it's that likely that the MIC would want to hide the existence of aliens. I think it's even more unlikely that the aliens would want to show their existence to the MIC and only the MIC.
You think the aliens tech would be simple and comprehensible enough for the MIC to replicate much of it.
As opposed to, take a smartphone. Cut it in half. Give both halves to a caveman.
Also. To me it seems "Aliens crash landed on earth, we have the spacecraft and we are developing all sorts of cool alien tech, no we aren't telling you the details" is a great boast, and a good excuse for needing more funding.
“The flash of light you saw in the sky was not a UFO. Swamp gas from a weather balloon was trapped in a thermal pocket and reflected the light from Venus"
I think this is a hard sell to those of us around for the von Daniken years. We grew up on 70s documentaries of ancient astronauts, was Atlantis real, the crystal skull and all the rest of it.
One more "documentary", even if it has modern gloss and "no but this is a real Pentagon guy!", isn't going to move the needle much. We're pretty much seen it all before, and the underlying story hasn't changed: crashed craft, the greys versus the other aliens, bodies in secret government labs. EDIT: It's like the Fortean Times, which I stopped buying after a couple of years because the same stories turned up again and again with little to no development. Great the first time you hear about it, still fun the fifth time, by the thirtieth time it's old hat and boring.
Get a real alien on TV broadcasting to the globe and we'll buy the docudramedy.
Yeah I'm an old fart and watched much of the von Daniken stuff with my dad. The number of government officials giving interviews makes this much different from the tales in the 70's.
Oh what, they've got *six* government officials instead of three this time?
If you've been around since the 70s, you know that the "it's all a giant cover-up!" set have touted "pilot X, official Y, other guy with fancy title Z all claimed to have witnessed this, that and the other". It's always been some government official.
Yet we still haven't any aliens rocking up to the European Parliament going "hey guys, we've been here since 1940 so we thought we'd finally say hello".
Scott seems almost to be inviting ragebait, so how about this: it's not much fun to be a white person competing in the international 100m, so why don't we have different races for, er, different races? Or perhaps someone has already had that idea at some time in US history.. ahem.
The answer, I suppose, is that white athletes would have ended up with all the best facilities (in the past), are embarrassed to ask for their own category (today) and can't present themselves as a disadvantaged group overall (no doubt groypers will disagree).
Back to the start: it seems to me that women's sports are justified by (1) historical disadvantage and (2) the fact that mammalian sex is a natural kind. We could deal with (1) in other ways, for example by positive discrimination (promoting 10th place to 3rd, say) or handicaps (as with racehorses, male runners could be made to carry weights!) but (2) is worth thinking about because not every 'demographic' can have its own sporting category. Hence we have the Paralympics but not the Flabolympics.
Two further points to consider: (a) injury and (b) sexual predation. Heavyweight boxers can't self-ID as featherweight, yet trans-identifying men have caused serious injuries in women's contact sports. As for (b), this is an obvious likelihood and ignoring it has not helped transactivists this far, but to repeat the obvious: if you allow, say, men aged 40 to self-ID into the under-15 category, you should not be surprised at the resulting scandal in the changing-room showers. And predation seems especially likely where males are allowed to self-ID as female with no gatekeeping allowed.
Actually if you tried, I suspect there would be plenty who would try to stop you!
And also actually back at me, if you tried to set up a "left handers only" sports league, you'd probably get sued for discrimination.
Absent those considerations, there's no reason we couldn't let free association sort this out, though. Not every division would reach critical mass, one should be able to find enough southpaws (or whatever) to enjoy a friendly game every now and then.
You made a claim. It's on you to provide evidence for it. You've come back with one data point. Which tells us pretty much nothing about anything. And you have the gall down thread to say that someone else needs to read up on the basics of statistical reasoning?
Of course you can. You can also point to cases of non-trans (actual women) causing serious injuries in women's contact sport. What you cannot do is show that injuries had anything to do with trans-ness.
How much does this change after HRT (suppressing testosterone to natal female levels, and administering exogenous estradiol to natal female levels)? What length of time on a female-equivalent hormone balance sees the bulk of punch strength reduction and how does this point compare to the natal female distribution?
How does the body signal for muscle growth and retention? What happens to muscle mass when that signal is suppressed? Does it increase, stay the same, or decrease? If muscle mass decreases, might that also reduce strength?
Ultimately, you should not compare yourself to others. The most fair sport category is one that includes only one person -- you. No more unfair advantages of any kind.
Some people suggest that you should compare yourself to yourself; to check whether you have improved compared to the last month or the last year. I reject that as microageism.
There's a case to be made that all competitive sport is a waste of time. If that's your position, I wish you all possible success in convincing people..
There is a political decision to be made whether people should finance their own hobbies, or whether government should finance them from taxes.
If it's the latter, it would make sense if the money assigned to a competitive sport was proportional to the number of people who enjoy this hobby. Maybe more if the hobby has positive externalities (e.g. making people more healthy), or less if it has negative externalities (e.g. fans destroying things).
Who can afford £11 a month for one Substack? This seems like an exorbitant price to ask. I don’t mean morally, though I think that’s an important question, but I just think there might more more than 2x subscribers if it were half as expensive (I’d be among them).
I also don’t get the impulse. If you can afford it, shouldn’t you want more people to read the locked posts even if deemed important
Sorry this is so critical. I do think your writings are worth at least 11 a month
Then don't pay it, sheesh. I hate people who complain about prices. You're free to leave anytime. No one owes you a product that fits within your budget. Everyone has a right to maximize their revenue. Obviously lots of people are willing to pay the subscription fee, otherwise it would be lower. Prices are a two-way street and aggregate information about both supply and demand, so you may as well criticize current subscribers for paying. Also bear in mind that supply is elastic so a high price here sends an economic signal that incentivizes the creation of more high-quality writing.
I worked with marketing teams for 5 years, there is nothing more valuable to a marketing team than honest and direct consumer feedback on their policies and products.
The idea of 'customers should stop expressing their preferences and intentions to the people trying to sell them things' has always seemed insane to me. Please, for god's sake, tell me as much as you possibly can about what I could be doing to earn your money!
That is the only price the 'invisible hand of the market' can receive, if you go in for that kind of thing.
Actual companies spend huge amounts of money and effort on getting more feedback than that, including scanning social media for people talking about them. They find this feedback very valuable.
This is just a brute empirical fact about reality, one that I am confident about because I worked on the teams doing it. You can try to deny that fact if you want, I guess, but I'm not sure what you're getting out of it.
Explain to me exactly what information that Scott got from a single person saying "this isn't valuable to me at this price" that he didn't get from the fact that that person didn't subscribe. He knows how many views he gets. He knows how many subscribers he has.
>Actual companies spend huge amounts of money and effort on getting more feedback than that,
Yes marketing departments are well-known for wasting money on technically misguided analyses. I, too, have provided technical assistance to biz-dev teams and can testify that they're largely incompetent. They only thing most of them are interested in is generating reports that demonstrate to leadership that they're doing something.
Halving the price is unlikely to get double the subscribers. The friction from $0 to even $1 is huge and most people will never pay it. Of the small pool of people who are happy to pay for a subscription in general, they are very likely to want Scott's writing no matter what, and as Scott has a monopoly on Scott's writing, people are going to pay whatever is being asked. There's an upper limit of course, but generally there is no competition making the service inelasric.
If you're a student or you can't afford it, you can use the link in the subscription page to reduce it to $2.50 a month! ;)
> If you can afford it, shouldn’t you want more people to read the locked posts even if deemed important
Have you considered that he doesn't want people to read it? Why do you think the last three hidden posts are about trans people, Gaza, and fascism? It doesn't attract the best kind of attention, if you know what I mean.
> I just think there might more more than 2x subscribers if it were half as expensive
The trick is how to find out whether this is true, without making subscribers angry.
(If you don't care about making subscribers angry, you could halve the subscription, check whether the number of subscribers doubled, and if not, double the subscription again.)
Yes, but ACX is a bit of a nonstandard product, so I am not sure how much the traditional solutions apply.
I mean, the majority of content (most Scott's articles and Open Threads) is free. You pay for a little extra content (some Scott's jokes, and subscriber-only Open Threads), which could perhaps be better described as "a gift with a nudge". Gifts can have a weird dynamic, more complicated than the usual supply and demand. If the subscribers stopped paying, they would still keep getting maybe 90% of the value they are getting now.
Tiered pricing would require Scott to produce more extra content. I think the billing is per month, so for N tiers it would be nice to have N extra things every month, otherwise someone may feel cheated. It could also feel weird, like: "hey, if you help support my hobby, thanks, and I will tell you a joke... but I won't tell you the other joke, because you didn't support me as much as some others". The current way kinda like a gift exchange, the tiered way would feel less so.
Or maybe I am wrong here, and Scott is just leaving money on the table. The problem is, if he makes an experiment and it turns out that there was a weird psychological reaction from the current subscribers, something like: "I don't want to pay more... but I hate the feeling that I can't see some hidden content despite sending Scott some money... well, if I can't see all the jokes, I might as well unsubscribe and save the money", that change may be irreversible.
> Tiered pricing would require Scott to produce more extra content
Not necessarily. e.g. you could have a variable delay on the premium posts instead of just a straight unlock, so it becomes "pay $x to have access to the posts a week later after the comments section has mostly run its course, or $5x to have access immediately when they go up". (Substack's awfulness helps with that: the comments section becomes unusable around 1k or so comments mark, so there really is a race to get to say your piece). Or you could do the Humble thing and have "pay whatever you want, but if that's at least $x you have access to the premium posts", instead of a flat fee.
Can anyone here point me to a primer on shrimp welfare arguments, directed to someone already sympathetic to animal welfare in general?
I've realized that the ratio of trans women in sports : shrimp welfare comments in this comments section is disappointingly high, so I've decided to be the change.
I made this account initially to write a comment about how I, unlike Scott and apparently all his fans, loved modern architecture. Hong Kong is one of my favorites. So, just a fan.
I don't know if I'd classify HK as modern architecture. I do love it as a glorious mess of a place, but I associate it more with "crazy eclectic mix of everything" than being a good representative of modern architecture (Singapore or maybe Shanghai seem like cleaner positive examples of that).
You're sleeping on Shenzhen and Chongqing here. I forgot the context of the original comment, but I was discussing how I heavily weighted scale and visual glamour compared to modern-architecture-haters.
Not sure I understand this. He's talking about reducing the suffering of shrimp, not increasing the suffering of vertebrates.
If I give money to reduce the suffering of children, that doesn't make me "an enemy of adults". If I give for the welfare of horses and cows, that doesn't make me "an enemy of primates".
As he see's it...he's giving on the general premise of "reducing the most suffering possible per dollar", and the fact that shrimps win out in that argument is just how the chips fell. If the same sums came out in favour of (say) giving the money to causes that improve the welfare of cows.... he do the same for cows.
He isn't the enemy of vertebrates in the real world example, nor the enemy of crustaceans in the counter-factual just because of how the calculus works out. In both cases he'd be the enemy of "suffering" and trying to make sure he is as effective an enemy of suffering as he can be.
1. Donating a penny to one cause is withholding one penny from all others. Although technically, it's more like withholding a fraction of a penny, since working towards one goal can have knocked on effects that help other causes. But I'm not sure how this justifies shrimp welfare.
2. Saying "every sentient life has equal value" is pure ideology, and it's based on vibes. I don't care how many math problems you do, and how often you update your priors. If I'm not feeling the vibes, I'm not feeling the vibes.
1. So you are an enemy of any category of objects on earth not supported by a charity, if you donate a dollar to that charity? Thats not usually how we use the word "enemy".
If I donate to (say) a donkey sanctuary.... that makes me the enemy of any living thing that isn't a donkey, all the inanimate objects that aren't involved in the welfare of donkeys, and the entire universe outside of earths atmosphere. Because I donated to a Donkey sanctuary?
Can't I just be a "friend of donkeys and neutral to every other non-donkey object in the universe"? Must donating to the donkeys make me the enemy of everything else? After all, I *could* have donated that dollar in some other way that would have benefitted all those other objects and callously refused to do so in order to support Donkeys.
This is NOT how we use the word "enemy" in normal conversation.
2. First, this seems orthoganal to the "enemy" argument. This can be true or false and not make him an enemy of mammals/vertebrates either way. So we are now branching outinto pother arguments/discussions without settling the enemy point first .... But even taking it as a separate line of discussion ... its not his position.
He explicitly does NOT value all sentient lives equally. He discusses extensively that shrimp are clearly NOT equal to mamalian/vertebrate sentient life as a moral consideration. IIRC he discusses "discounting" their suffering by between a factor of 100 and 10,000 compared to vertebrates and points out that because so many shrimp are involved, you can easily discount by that factor or more and it still equates to a reduction in overall suffering compared to alternatives as so many shrimp are involved (numbered in the tens or hundreds of millions) and the cost of alleviating the suffering is so low (a handful of dollars per million) that the math works out anyway as opposed to (say) donating the same dollar to a donkey sanctuary.
Nor does he impose any individual value (a 100 to 1 discount....a 10,000 to one discount.... a million to one discount) on the reader, he leaves it up to you to decide what that discount should be (and even leaves it up to you to apply an infinite discount if you think thats appropriate) and provides information relevant to you making that choice. So its hardly imposing any ideological conclusion. He describes his personal conclusion, the facts as they are, the maths as it is, and leaves up to you how to apply those and whether you do so in a way that ends with your agreement with his point or not.
Finally, I don't see how you can say "you're not feeling the vibes" of his argument.
Given you're response 2 above it's absolutely completely clear you haven't read his argument. The "vibes" you are or are not feeling.... and vibes in relation to "an argument I made up in my head specifically so I could disagree with it". I can totally understand why you aren't feeling vibes towards accepting that argument, you've personally constructed it in a way to give you the opposite "vibe".
Perhaps you should read his argument, "feel the vibes" from that actual discussion... and then disagree (if you want to) rather than "imagining what he said, in as strawmanny a way as you possibly can ("each shrimp is fully morally equal to an adult human"), so as to support the conclusion I already reached before listening to what he had to say".
Its not as if I totally disagree with you here, I don't think I reach the same conclusion he does EITHER.... its just I've read his argument and conclusion and (like you) "vibe decided". I just read it first.
1. Well since transwomen are not women by the definition they cant participate in women sports. Thats the ground.
2. Applying that logic consistenly we also would have to allow cis men participate in women sports and adults in kid sports. This is just continuum fallacy.
Most sports *are* Open and Women already, woke people are just quibbling the *term* women because they've decided it should mean something other than what it de facto does and has done for centuries.
In fact, I think the only sport which had a rule that women could not participate in the "men's" division even if they qualified, in the modern era, was golf, and that was changed in the 1990s or something. Otherwise, any woman who makes the cut can play in the "men's" Wimbledon, for example, it's just that that's physically impossible for a woman to achieve so in practice there are no women. That's the whole reason why they get their own category in the first place.
Agreed, I've long advocated for this solution. As far as sports go there are two categories: actual women who were born women and everyone else.
This just illustrates the insanity of the transgender question in the first place. Gender isn't arbitrary: it serves lots of objectively important purposes in thousands of subtle ways all throughout society. Rejiggering all of that for a vanishingly small fraction of society is just objectively foolish. Society should never waste resources worrying about a 1%-sized demographic. It's simply out of proportion.
"Society should never waste resources worrying about a 1%-sized demographic. It's simply out of proportion."
That's not true. The meaningfully wealthy are far less than 1%, but their concerns are well worth considering, not only due to the knock-on effects on society at large but because they have legitimate claims on the conduct of the rest of society (such as not being decapitated for their money, again). Admittedly these claims scale down to every subsequent tier of prosperity, but even if they didn't they would still have a justifiable demand to have their claims of property respected.
However, you're certainly correct that it's bizarre for a tiny cultural/medical minority community to have palpable sway over social mores, especially in a democracy whose whole design purpose is to marginalize such irrelevant minorities.
Yes agreed my point is actually about value and not just size. The wealthy are different because society actually does have a rational interest in catering to them.
>especially in a democracy whose whole design purpose is to marginalize such irrelevant minorities.
Democracies aren't designed to actively marginalize minorities, but neither are they designed to cater to them. Given the choice between active marginalization and active catering I'd say it's objectively better in a net-social-outcome sense to marginalize them. Society simply can't function if every 1%-sized valueless group can command public attention. That just leads to political gridlock as 100 different factions demand conflicting policies, particularly when those demands come on the back of completely inflexible rights-based claims. That's just anarchy. Minority rights don't entail the entire country changing its bathroom policy to enable the preferences of a tiny group of ill people, for example. That's true regardless of your views on the medical legitimacy of transgenderism. If there was a cancer that 1/1000 people had I wouldn't endorse destabilizing our national politics to make their lives more convenient either. The needs of the many outweigh the needs of the few.
"Democracies aren't designed to actively marginalize minorities"
Yes they are. Majority rule entails minorities only not getting marginalized if if amuses the majority more not to marginalize them in whatever respect – in other words, it's dependent on the amusement of the majority. This is crucial because it's *why* minority advocates of various stripes are always so eager to corrode democracy and put various things beyond the reach of the people, whether in the form of binding treaties and conventions, leaving matters to "an apolitical team of experts" (i.e. in practice a bunch of middle class managerialists ideally drawn from the ranks of the advocates themselves so that they can be expected to be reliably ideologically captured), or any other form.
EDIT: Forgot to add, I agree with the rest – "Society simply can't function if every 1%-sized valueless group can command public attention" is indisputable. Going by some pre-2010s-inflation figures on trans people they, at that time, were a smaller group than sufferers of cystic fibrosis. As a thought experiment, imagining fibrotics *alone* having the same amount of cultural/news airtime as transsexuals (while not depriving the latter of attention) is instructive: it would be an exhausting barrage. *Every* topic of that level of general interest laying the same claim to our attention would be wholly unlivable. People would start killing entire minority groups just so they could hear themselves think.
I get accused of being lawyerly on the reg. But man, this takes the cake. If I ever get arrested and charged with a felony, I want you by my side in court.
Exactly, the whole point is that "women" isn't nearly as coherent a concept as the statement you're making makes it out to be. Both sides of the spectrum would agree with your claim, they just disagree with what the word "women" means. There's no real category around women, that's the whole debate.
People had the word women long before they could check the chromosomes of people etc etc
Women is an extremely coherent category. The vast majority of people have no problem with it - if you are born as a women with a vagina and uterus and such, you are a woman.
Exceptions do not matter here, it's not that hard.
There isn't a debate. A small group of radicals want to change the definition of women to something totally subjective and effectively meaningless. To the rest of humanity across cultures and throughout history, the distinction between men and women has been obvious. The words are not arbitrary; they represent real, obvious, and consistent patterns that exist in reality.
> Both sides of the spectrum would agree with your claim, they just disagree with what the word "women" means. There's no real category around women, that's the whole debate.
But there is a bait and switch going on here, no? When we discuss matters such as "what is a woman" or argue that "trans women are women" etc, we are talking about gender, we can do this because the classification is social in nature.
The distinction in sports comes from the other meaning of the word, which refers to sex, not gender.
No popular concept is coherent. There are always arguments around the edges. (E.g. I consider both Neanderthals and Denisovians to be humans.) So expecting one to be coherent is unreasonable. Whether trans-women can compete in "women's sports" is a boundary argument, and either choice is defensible.
If you have a person who’s a woman by any definition, and she happens to have sky-high testosterone, the only place she can rightfully be banned from would be the low-testosterone league. How is it “rightful” to define a category by quality A and ban people from it based on quality B?
...sudden shower thought: to what extent is that actually true, though?
"Women's sports" is a very broad category indeed, containing many very different things that can be categorised in very different ways (e.g. groups that only allow women participants doing things invented by men, vs groups that don't even necessarily explicitly limit participation but play games invented by and traditionally played by women, and everything in between), and absolutely does include activities that were designed at least as much with the viewers as the participants in mind, with a focus on showing the (cis) men in the audience something they'd pay to see.
We dont need any groundbreaking new argument. The answer is crystal clear and has always been. There was never a valid "debate" to begin with just as there is no real debate whether the earth is flat or not.
to myself: this thread caused alot of debate among everyone except me lol, i barely received any replies. This is a pattern i noticed as I got older. When I was an obnoxious 14 year old on the internet I almost always received replies to my comments but now barely any. I think the reason for why that is is obvious. Sad but thats the cost of being objectively correct.
Question for ACX readers: Is there a name for the position between 'pause AI development' and 'accelerate at all costs' that focuses on interpretability and coordination infrastructure?
I wrote three fables exploring what this looks like as design principles rather than engineering specs. Systems that must narrate intentions before acting, communities with built-in verification, infrastructure designed to amplify prosocial behavior rather than engagement. Basically: Bostrom's sparrows, but they figure out how to thrive.
I think most people in AI safety are between "pause" and "accelerate". I don't think there's a name because it's the normal moderate position.
But I don't think that your fables/principles are speaking to the same concerns that most other people in the field are. Other people in the field are more asking "What if the AI is trying to deceive us?" An AI that was trying to deceive you would narrate a perfectly good-sounding intention before acting, then do something else when you weren't looking.
You're right Scott—I was addressing coordination assuming alignment, not the alignment problem itself. If the AI is deceptive, these mechanisms don't help.
The narrower question I'm exploring: Given aligned AI, how do we build coordination infrastructure so fast deployment doesn't preclude collective oversight? The Two-Feather Rule, Chorus Path, etc. are about human coordination around AI systems we trust but want to verify.
Does that problem matter if deception is unsolved? Or is it too downstream to be worth any attention now?
The problem with the "pause" scenario is that it requires broad agreement and enforcement. This isn't present, and isn't likely to become present. So it's not a real option. The options that *are* available are "open source" vs. "closed source". It seems possible, though unlikely, that legal frameworks could tilt that in either direction.
The value of the open source approach is that it enables those who want safety checks to design them. The problem is that using of them can't be enforced, so those who don't want to use them won't.
A plausible name for an intermediate position would be "put lots of effort into safety checks". I'm dubious about it's possibility, as if there's much cost, the first mover is likely to be one who doesn't do it. (Scott's argument that the cost wouldn't be significant seems wrong, as I think doing it properly would require a lot more effort than he is assuming.) Also, it seems unlikely because businesses are already starting to apply AI agents that we KNOW produce incorrect answers that may be costly.
I agree that in a pure competitive race, interpretability mechanisms will get dropped for speed—classic race to the bottom.
I'm exploring what the alternative landscape looks like—not 'how to win the race' but 'what infrastructure we'd want if we weren't racing to the bottom.' Getting there probably requires regulatory mandates or post-crisis adoption.
History suggests we regulate after things break: traffic safety after 5 decades of road deaths, building codes after major fires, FDA after drug disasters. The question with AI is whether we're smart enough to build guardrails before the catastrophic incidents happen, or whether we'll follow the same pattern.
Your point about businesses already deploying 'semi-reliable' AI is well-taken—suggests we're in the 'breaking badly' phase but haven't hit the threshold that triggers political will for regulation yet.
Are you saying the race dynamic is so dominant that exploring alternatives is pointless? Or just that we shouldn't pretend voluntary adoption is realistic without enforcement or crisis?
To me it looks analogous to the "tragedy of the commons". Without enforced rules, people will act for personal advantage. But enforcing rules in a worldwide development is difficult enough when most people agree on what the rules should be, and there is limited advantage in breaking the rules. Who would decide on what the rules should be? Who could enforce them? (For "Who" read "what mechanism", as it probably shouldn't be some particular person.)
It's not that the race is so dominant, it's that the structure of the problem is extremely difficult. I don't see any solution that doesn't require a quite intrusive international power with sufficient force behind it to intrude where it wasn't welcome. ... Actually, that's not quite correct. If one group felt that they had an unassailable lead, then they might be open to being careful. In the actual situation, however, everybody is being secretive about just how advanced their best models are...at least until they actually release them, so nobody can be really certain of their lead. Consider how Google just upset OpenAI.
This is a multi-player game of hidden information with partially unknown rules for extremely high stakes, without a referee. And it's not seen as an iterative game, so it's more like a multi-player version of single-shot prisoner's dilemma.
This is really helpful framing—the 'single-shot prisoner's dilemma with hidden information' is bleaker than I'd initially conceived.
I'm coming to terms with the fact that I'm designing mechanisms for a coordination equilibrium that may never exist without either crisis-forcing-function or (as you say) intrusive international enforcement. That feels like clearer scoping than I had at the outset—design for ideal conditions, acknowledge the conditions are unlikely.
Appreciate you helping me think through the structural problem more clearly. This has been genuinely useful. Thanks.
The other day, I came across The Qualia Research Institute's Psychedelic Cryptography Contest. The aim was to "encode" a message in an image so it could only be read by a person who was tripping — psychedelic steganography.
Here's the winner. Evidently, any psychedelic/dissociative substance that causes heavy tracers could enable the decoding of the message. I don't have the substances to test it, myself.
> It turns out that in order to decode these pieces you do require a substantial level of tracers, so only members of the committee who had a high enough level of visual effects were able to see the encoded messages. Some of the members of the panel reported that once you saw the messages during the state you could then also see them sober as well by using the right attentional tricks. But at least two members of the panel who reported seeing the messages while on mushrooms or ayahuasca were unable to then see them sober after the fact no matter how much they tried.
Somewhere in the write-up, they explained that it probably worked because of tracers that one would see with psychedelics like LSD and Psilocybe. And the tracers would leave behind after-images of the motion on your visual field.
If you're not aware of the term, tracers are the visual trails that one sees when an object moves across one's field of vision. For me, they've always been trails, something like colored lines you see in comic books to denote motion. According to a study that Gemini referred me to (below), "LSD users perceive a series of discrete positive afterimages in the wake of moving objects, a percept that has been likened to a multiple-exposure stroboscopic photograph." It's been a while since I tripped, and I don't actually remember them being so clearly stroboscopic.
And I always assumed the effect was due to my pupils being fully dilated while tripping, but on further consideration, that doesn't really explain the visual effects I saw. And it certainly doesn't explain the stroboscopic description that some people see!
BTW, I don't see the message at 2x speed, but the video ends too soon for me to visually "lock onto it."
I can see the hidden images/messages while sober but I have done a lot of psychedelics so maybe it's a lingering aftereffect. It takes a bit of effort to parse it at 1x speed but at 2x speed it sticks out really easily.
We also hosted a sillier debate between Bentham's Bulldog and Jeff Maurer on the motion "Donating to GiveDirectly instead of Shrimp Welfare Project is a chump move".
Re the trans sports thing - I think you have to get back to the question of why we have sports at all, and why do we have women's sports, then ask how do trans women fit within that. Some of these reasons have implications re trans women in women's sports.
E.g. one big reason for women's sports is that people saw that male athletes were heroes and role models for boys, and wanted something similar for girls. They think that as long as athletes are so idolized, we should have the same for women, and it's important for a project of gender equality/empowerment/whatever in a way that "short basketball players" isn't important for short people equality. What that means for trans women and sports ... people will disagree I'm sure.
Another reason - same way people want to see the fastest man over 100 meters, they want to see the fastest woman over 100 meters. For people who don't want trans women in women's sports, they specifically want the fastest biological female, gender identity isn't what they're interested in.
Another reason - camraderie among girls. Making an all-girls environment where they can build confidence etc in a way that doesn't happen so much with the boys around. Again what that means for trans women, people will disagree.
I don't watch women sports, so it wouldn't even feel like I "accepted" something. If the people who actually do the sport want to have it certain way, let them have it.
My current impression (possibly wrong) is that some women who do the sports hate the idea of competing against someone who was a man yesterday, especially if it is a combat sport (when it's not only about losing, but about likely ending up in a hospital)... but if they object, they risk cancelling. So it's an end of their career either way.
(I especially hate the argument about how trans athletes are only 0.00000000001% of the population. I mean, that's exactly the point, that having a male body in a women sport is often such advantage that those 0.00000000001% easily get to the top.)
This is an important point but the justification of a practice doesn’t always determine what the fair rules of competition or inclusion are, once the practice is already established. It might be that we established, or would have had reason to establish, organized competitive sports for some of the considerations you mention, but also that, given that they are established, fairness within the competition demands something that is at odds with the original justifications. Or so it seems possible to me.
Trans women have been eligible to join the Olympics for 22 years, and have never won a single medal.
I think things are plenty fair, and will continue to think that until trans women actually win more competitions than the null hypothesis would predict.
In terms of camaraderie among girls in the context of casual school sports, the trans women I've met fit very well into female social groups and social styles. This is why cis lesbians now have the strongest rates of support for trans rights among cis people; nearly every lesbian knows a trans woman who blends into the community. Trans women even seem to be prone to the same sorts of gossiping and shunning that cis women are.
I've seen several surveys which find this same result, but the above is one from the UK from a couple years ago. There is a vocal minority of trans-exclusionary feminists which lead people to believe that cis-lesbians tend to be trans exclusive, but when you actually ask the communities in any kind of systemic way it's clear that TERFs are not representative.
To be pedantic, this survey asked cis lesbians what they thought about trans PEOPLE, not trans WOMEN specifically. The finding is entirely consistent with the cis lesbians in question having many trans men in their social circle, of whom they think highly, but also having a negative opinion of the trans women they know.
Feels like an implausible stretch, but you are welcome to apply whatever standard of evidence you would like to this in order to try and confirm your own priors. I cannot find a study that specifically parses out attitudes towards trans women and trans men, but it does not seem likely that approval of one group would be un/inversely correlated with approval of the other.
Why specifically lesbians? I thought they would be more annoyed with trans lesbians trying to court them. I would have assumed that gay men supported them more, since there's not much reason for them to get into conflict with trans women.
Agreed. I'll add that when people argue over this, they often have different ideas of who we should optimize for. We can focus on what's best for the competitors, the spectators, the identity group (women in this case), or the young people who think they may one day be a competitor. And we may have different answers at different levels of competition, like Olympic level versus high school level teams. Debates can get muddied when people don't realize they had different starting assumptions.
It's also a proxy issue for the more general culture war, so you get people who never cared much about women's sports having very strong feelings on this particular issue.
Why do you have to care abut women's sports, though? I couldn't care less about womens sports, but allowing in transgenders is just a litmus test for basic sanity. If you can't pass this test then I don't trust you in any type of leadership role.
> It's also a proxy issue for the more general culture war, so you get people who never cared much about women's sports having very strong feelings on this particular issue
That's true, although I don't think we should dismiss the opinions of those sorts of people, people are allowed to be tourists on issues when it comes to issues of fairness.
If the issue is "no blacks allowed at the Woolworths lunch counter" then having eaten at a Woolworths lunch counter is not a prerequisite for having an opinion on it.
I don't think you should necessarily dismiss their opinions but you can question their motives. Someone who always treated women's sports as a punchline, who opposed equal funding of women's sports under Title IX, who suddenly is freaking out over this issue, I think you can question whether they are truly motivated by concern for women's sports.
As somebody who would prefer to see toddlers kicking a rock around a muddy rain sodden field in November than the woman’s World Cup I do mostly try to keep out. In fact all men should keep their distance, that said i feel when this ideology breaks down the patriarchy will be blamed somehow.
People who worry about head injuries in football aren't worried because they think it will hurt the game of football, they're worried because they think it's bad for the kids' health. That's totally consistent with also thinking football is stupid. I think you can question whether they'd be so gung ho about it if they loved football and didn't think it helped the patriarchy or whatever ...
but it's a different boat from people who deride women's sports but think that trans women in women's sports is bad *for the sport*, i.e. bad for competition and such.
You could argue it's not a different boat specifically for a fighting sport (e.g. boxing not swimming), where there is arguably a health angle separate from the fairness angle, but nobody limits their views to those types.
And when I was wrestling, I was allowed to wrestle against people in weight classes higher. I could not choose to wrestle against people in lower weight classes however.
Correct. I did this in one match. If I recall correctly, the opponent would have no grounds to object, and one could only "wrestle up" no more than two weight classes (for the smaller wrestler's safety).
From an abstract, crotchety, no real pokemon in this fight perspective I kinda agree. To me if you're engaging in a competitive discipline, either you want to be the best in the world, no excuses given or accepted, which means you need to be willing to compete against the very best in the world across categories. Or you just want to do your best and have a good time, in which case it makes sense to sort you into a league with people of similar ability, but there's no particular reason to privilege sorting by sex or gender over height, weight, years of experience, or just empirical demonstrated skill (eg by an Elo rating).
A lot of sports do sort by this though, league systems (e.g. the English Football League System) are skill based and combat sports generally have weight classes. They just do it on top of the gender, although its also worth remembering that a lot of male leagues and competitions aren't actually male, they're open to all which seems to result in all male once you get out of the most casual of casual matches, the standard is an open competition and a female competition.
Uh what? Can you name a single organized sport that doesn't bother to sort people by ability at all? Where a random new player can just wander in off the street and get matched against the best in the world?
> most sporting victories involve biological advantages that make things less fun for people who don’t have them
This has occurred to me for some time now. The conclusion I have reached is that trans rights are at the moment a kind of compromise - many people accept it as a kind of polite fiction for civil society and accommodation for unhappy people. But not to the extent of like, fully internalized category update
the sports thing is a kind of rubicon/line in the sand, which is why it's so important despite not mattering much on the object level (see also: dating)
Most trans people agree that trans women are women, and that trans women are not exactly the same as cis women in all ways. The two options aren't "trans rights are a polite fiction" and "trans women are treated exactly the same as cis women in all contexts." For example, many trans women want access to options for freezing sperm.
I'm speaking of trans women here since those are the people this controversy is mostly about.
> Most trans people agree that trans women are women, and that trans women are not exactly the same as cis women in all ways.
Yes, and most people aren't trans. When are they going to realize that they aren't in a position to make demands? Don't they realize that people will only tolerate them as long as it remains convenient? Do they not realize what will happen if they start making things inconvenient for the majority?
Most people don't use wheelchairs and we're still doing okay (but could be doing a lot better) about having ramps, accessible bathrooms, and curb cuts. Most people aren't Muslim, but every major university/conference/anywhere else that provides food I go to has had halal options. Have a little more faith in humanity.
"having ramps, accessible bathrooms, and curb cuts"
All of which in no way inconvenience the majority. In fact, the bathrooms are also a convenience to non-disabled people who don't want to share a can (although this sometimes disgruntles the disabled, it's still widely allowed; guess why?), and curb cuts can help e.g. old people and kids as well as people in wheelchairs. A halal option isn't an inconvenience for the majority either as long as you can just opt to eat non-religious meat, which you normally can.
But those things do inconvenience the majority. A public building with ADA accommodations is getting that extra construction fund from your tax dollars, and a conference with halal options is getting the extra food budget from your ticket. These are indirect, yes, but sometimes inconveniences are direct too, with disabled parking being an easy example.
How much a given accommodation costs versus benefits the majority depends a lot on what the accommodation is, so it's hard to compare without getting into specifics, but generally speaking there's usually some cost and some benefit.
> Most people don't use wheelchairs and we're still doing okay (but could be doing a lot better) about having ramps, accessible bathrooms, and curb cuts.
Which is not a product of disabled people using their leverage, but the voluntary kindness of the abled. If disabled people start making unreasonable demands and make their existence problematic for society, people will start withholding such kindness.
> Most people aren't Muslim, but every major university/conference/anywhere else that provides food I go to has had halal options.
That one's definitely not going to last long, given the current political climate in the US. Immigrants have become a problem for society, and you're about to see what happens to people that become a problem.
It sounds like you’re inclined to simply define an unreasonable demand as one that the majority won’t abide. So your argument seems pretty empty to me.
'Society' is doing a lot of work in these assertions.
An overlapping patchwork of people who are minorities along various axes comprises a big chunk of society, and while the degree of success fluctuates, they often try to stick together, as per the old Niemöller litany.
Add to that the plurality of default-type people who are willing to join the 'other side' on ethical grounds, and your society ends up fairly cleanly split.
Fine, is "the collective that holds majority power in a given setting" better? I won't even say population majority, because not all people have equal capabilities to impose their will, one of the more obvious example being men versus women. And with 61% of men saying that changing gender is morally wrong, and the views of the ruling party being what they are, it's not looking too good for our trans friends. https://news.gallup.com/poll/645704/slim-majority-adults-say-changing-gender-morally-wrong.aspx
> Which is not a product of disabled people using their leverage, but the voluntary kindness of the abled. If disabled people start making unreasonable demands and make their existence problematic for society, people will start withholding such kindness.
A cursory investigation into the history of disability rights will show that they were often not a voluntarily given kindness, and often a result of disabled people organizing, filing lawsuits, and "making their existence problematic" by doing things like occupying federal buildings and staging an event in which they crawled up the steps of the Capitol Building without assistance. Likely many people thought of their demands for legal protection as unreasonable at the time.
These demonstrations were admirable but in no way do they seem like exertions of force. These were sophisticated, intelligent requests for charity. And that's fine, I've requested charity in my life, so has everyone. But it's still asking for charity.
> A prevailing view held that people with disabilities were pathetic and deserving of pity, and therefore incapable of such political actions. This same perspective made officials reluctant to risk a public relations embarrassment that would result from arresting participants.
Even the biased Wikipedia article admits that the government just felt bad for them. There was nothing forcing people to accede to their demands. But their pity and compassion is not endless. More importantly, most people do not feel pity for trans people.
How could immigrants possibly be a problem to American society when studies have found pretty much all forms of immigration to be pretty much universally beneficial to the US.
On a timeline basis the reprisal against immigration started at the beginning of the year and has shown to be pretty unpopular with the broader electorate, and immigration is viewed positively overall among the entire electorate except for specifically the surge in people exploiting the refugee loophole.
Jews are white these days, so I'm sure they'll be fine for a good while unless something triggers a massive wave of antisemitic sentiment among the right.
Worlds collide: someone I follow on X reposted a fiction piece written by an LLM (Opus 4.5, to be precise). It's based on a meme that has apparently been making the rounds, featuring two doors: a red one opening on $2 billion immediately, and a green one opening on $1 that doubles every day. The piece begins: "I chose the green door ninety-three days ago."
The first reply to it is from the person who posted it, claiming Claude made a few obvious math errors, which he fixed, and otherwise posted without further wordsmithing.
The second reply, also from him: "his was the result of thinking ”I wish there was a @slatestarcodex or @jackclarkSF short story exploring the green door option” and then asking Opus for that."
Yeah, I also thought it was good. I'm going to put it in a links post, hopefully in the form of a link "read this cool story", followed by "and after you're done, see the interesting part", which will be a link to him saying it's AI.
My first prompt was "Can you write a 1000 word short story in the style of slatestarcodex/scott alexander or jack clark/import AI about a world in which the dollar doubles every day, and the horrifying implications. Eg implications of storing the bits of the dollar value in bank systems, and how the economy is affected by all this money being created. Write it from the perspective of the one who made the choice."
It then went down a path of writing about how computer systems were crashing from storing the number of bits needed to store the value, so I had to correct it and say "make it mathematically accurate, i.e. bits to store the monetary value will scale linearly with number of days, since it's doubling and bits also increase the number they can represent by 2^x"
The Emergent Thermostat, the Beautiful Ones, and the Coming Collapse of Human Societies
A Unified Theory of Reality-Model Degradation from Thunderstorms to Hikikomori
1. The Insight on the Beach
In the year 2000, a scientist living in Fiji noticed something no climate model had ever been asked to explain: the global temperature of Earth had varied by only about 0.2% (in Kelvin) across the entire 20th century despite volcanoes, wars, and industrialisation. While most researchers focused on why the planet had warmed by 0.6 K, he asked the opposite question: why had it barely changed at all?
Observing the daily cycle in Fiji revealed the answer. Clear dawn led to rapid surface heating, which triggered cumulus cloud formation at a precise threshold. The clouds reflected sunlight, cooling the surface. If heating continued, thunderstorms would form at a higher threshold, producing downdrafts, evaporative cooling, and cold pools — enforcing even stronger cooling. Conversely, when the morning was cooler than normal, clouds formed late or not at all, keeping the surface warmer. Day after day, fast, local, emergent phenomena maintained an extraordinarily narrow temperature band.
His realization: a governor operating on the timescale of hours can, through repeated application, enforce the same bound over centuries and millennia. CO₂ is not the main long-term control knob of Earth’s temperature; thunderstorms, El Niño, and cyclones form the planetary heat engine’s cruise-control system.
2. The Blindness of Artificial Intelligence
Twenty-five years later, he tested two competing claims on ChatGPT: one, that the major control of the world’s temperature is CO₂, and two, that the major control is emergent phenomena like thunderstorms and ENSO.
The model initially tried a safe “both/and” answer, but when forced to choose, it picked CO₂. This is because that claim appears millions of times in its training data, while the emergent-thermostat hypothesis is barely represented, mostly on one blog. ChatGPT reproduced the exact objection the scientist himself had held for months: “but those phenomena are only short-term.” It could not grasp that repeated short-term regulation is the only way long-term regulation is physically possible in an open system with variable energy input.
This illustrates the core limitation of current large language models: they excel at summarizing what has already been said, but cannot originate genuine integrative insights.
3. Model Collapse Is a Universal Law
In 2024, Nature published the definitive paper on AI “model collapse”: neural networks trained on an increasing fraction of synthetic (AI-generated) data rapidly degrade into confident nonsense. The more artificial the training diet, the faster the collapse.
The parallel to human brains is chilling. Brains are prediction engines trained on sequential sensory data. When a growing fraction of a child’s training data is second-hand, symbolic, or entirely synthetic (screens, classrooms, anime physics, TikTok morality), the internal reality model loses fidelity in exactly the same way.
Critical developmental windows make this irreversible, just as it is in silicon. Vestibular and proprioceptive calibration occurs in the first 5–7 years, native phoneme discrimination is lost after 12 months, spatial navigation circuits prune if never exposed to complex terrain, and corpus callosum thickening from early music training cannot be fully replicated later. Remove enough real-world gradients — gravity, pain, hunger, predation, unscripted social consequence — and the network begins to hallucinate.
4. Universe 25 and the Mammalian Blueprint
John B. Calhoun (1947–1973) removed every natural selection pressure from social mammals and observed the same sequence in rats, mice, deer, rabbits, reindeer, and partially in primates: explosive population growth, sudden peaks despite unlimited resources, emergence of a hyper-aggressive minority alongside mass withdrawal of the majority, maternal failure with 100% infant mortality, and the appearance of the “beautiful ones” — perfectly groomed, healthy individuals who did nothing but eat, sleep, and groom, ultimately ceasing reproduction and leading to extinction.
This phenomenon is not rodent-specific; it has now been documented in seven mammalian orders whenever adult mortality approaches zero for multiple generations.
5. The Human Experiment We Are Running Live
For the first time in history, large human subpopulations meet Calhoun’s four conditions: adult mortality from violence, starvation, or disease is nearly zero; caloric and material resources are abundant; urban and digital density are extreme; and emigration to lower-pressure environments is functionally impossible.
The results mirror Universe 25 with disturbing precision. In Calhoun’s mice, the “beautiful ones” were perfectly groomed and socially withdrawn. In humans, the closest analogues are Japan’s 1.5 million hikikomori, 30–50% of 20–34-year-olds in affluent countries who are sexually inactive, and hyper-groomed but socially dead young adults in Seoul, Tokyo, Singapore, and Western cities.
In Calhoun’s mice, maternal failure caused total infant mortality despite food abundance. In humans, we see record child-neglect reports despite unprecedented wealth, and voluntary childlessness is rising fastest among the richest and safest cohorts.
A tiny minority of mice became hyper-aggressive while the rest withdrew. In humans, tiny cohorts (the top 2–5% of males) commit most violent crime while the majority increasingly drops out of work, dating, and reproduction.
Calhoun’s final outcome was total reproductive collapse. In humans, South Korea’s preliminary total fertility rate in 2025 is 0.72 nationally and 0.55 in Seoul — the lowest ever recorded. East Asia is furthest along this trajectory, having first and most completely achieved the Calhoun conditions. The West lags 20–30 years behind in its most affluent, screen-saturated strata.
6. Conclusion: Two Emergent Governors, One Civilization at the Threshold
Earth’s temperature is regulated by fast, ruthless emergent phenomena that activate the moment a local threshold is crossed — the mechanism required for long-term stability.
Human behavior is regulated (or de-regulated) by the fidelity of our internal reality models. When those models are trained on increasingly synthetic data, we produce our own behavioral sink: the “beautiful ones” who defend delusions confidently because they have never interacted directly with reality.
The same natural law governs both systems. Thunderstorms prevent runaway planetary heating. Direct, unmediated experience of physics, biology, and consequence prevents runaway social cooling.
We have spent seventy years removing every consequence that once shaped human behavior, and we are discovering in real time, at civilizational scale, that the thermostat works in both directions. The cruise control is still there. It is simply waiting for the next threshold to be crossed.
There's a lot to chew on there. Just out of curiosity, I ran a couple of probes against a couple of your statements using LLMs (ChatGPT and Grok).
> we see record child-neglect reports
In the US it looks like reports of child neglect or abuse went down ~20% between 2019 and 2023. That may be an aberration to a longer-term pattern for which I can't find any data at the moment. If you've got this data, please share some links.
The EU seems to be undergoing a surge in hotline reports of child sexual abuse complaints, but both the LLMs I queried said they didn't have any overall data on child neglect. The UK data shows that as of 2024, the number of kids in child protection plans is the lowest since 2013. OTOH ,Germany showed a 4% rise in child neglect/abuse cases between 2021 and 2022. This was flagged in news stories as unprecedented.
> John B. Calhoun (1947–1973) removed every natural selection pressure from social mammals and observed the same sequence in rats, mice, deer, rabbits, reindeer, and partially in primates...
Calhoun only ran these tests on Norway rats and house mice. Later researchers tried to reproduce his experiment with macaques, rabbits, pigs, and songbirds (I didn't try to find out which species). Although crowding could affect behavior, none of them showed a Calhoun style population collapse.
Hold on there! I want to know how they calculate that "$1 can improve welfare for around 1,500 animals," and how they arrived at "making them one of the most impactful animal charities to donate to by number of animals helped per dollar"?
These answers have profound consequences for how we measure the success of human welfare programs!
And just who is removing their eyestalks? Are these immigrant laborers using tweezers?
For a dollar, I can buy about 5 pounds of corn. I can then give each kernel to a different cow, which I'm pretty sure would result in that dollar helping more than 1500 cows.
But they don't actually have great publicly available data connecting their work to the index at the moment. The numbers on the shrimpact page seem to be a straightforward calculation of: (1) how many farms have adopted electrical stunners or agreed to other welfare changes (stocking density, pH management, eyestalk ablation, etc.); (2) how many shrimp those farms produce annually; and (3) programme cost.
“Improving welfare/helping X animals for $Y” can be a misleading framing that covers everything from "0.1% reduction in stocking density" to "not ripping their eyestalks off". But SWP’s main measurable impact is from introducing electrical stunners, so it’s reasonable to assume that “quicker, less painful death” is the main unit of benefit they’re counting.
This feels like a claim that needs evidence. The number of shrimp and other arthropods that are farmed are enormous, and the whole 'crush their eyestalks so they lay more eggs' thing makes me feel like we probably cause them to suffer.
I didn't say "they don't suffer", nor is my claim contingent on that. To the extent you can be confident that an animal suffers, not only can we ascribe far more confidence to the aforementioned, but the conditions they endure are the worst. Chickens raised in battery cages their entire lives shouldering disease, immobility, pain. Plenty has been written about this already and it's not hard to find, it's even on the old SSC blog. There's also a Richard Hanania post on the topic.
By comparison, shrimp is a complete joke. This is analogous to prioritizing eliminating pesticides and drastically resharping crop reaping practices because there are far more insects than anything else, and they have sentience too. You cannot equivocate these.
You've used this expression no less than 3 times in this Open Thread. You know what they say: If one person slaps you in the face, they are the problem. If everyone slaps you in the face, you are the problem.
Maybe it's just that I'm sleep deprived and I'm repeating myself. Maybe it's because shrimp welfare and biological men in womens’ sports are two extremely hot button issues and it's unusual to see Scott doing so much counterproductive advocacy in one post.
Maybe it's because it's a highly evocative phrase and so I like to use it. We all have our favorite phrases, don't we?
Did you have any kind of point to make or are you just taking pot shots?
We sure do. My post was mostly meant to be tongue-in-cheek, not as an attack. I'm sorry that I didn't make this clear.
Edit: What are you doing here if you're sleep-deprived? Take a nap, get some rest! Let people be wrong on the Internet, your physical and mental health is much more important than posting in some forum.
I am not convinced that vegetables don’t suffer, and actually I’m inclined to think they do. But vegans and animal welfare types, broadly speaking, do not seem to take this possibility seriously. (The only group that takes it seriously, that I’m aware of, is Jainists.) I find this kinda fascinating.
> I am not convinced that vegetables don’t suffer, and actually I’m inclined to think they do.
What would it even mean for a plant to "suffer"? Plants don't have brains, they don't even have a central nervous system. Suffering, and qualia in general, isn't an objectively measurable phenomenon, at least not as of today; we can only (speculatively) infer that other beings suffer by their similarity to us, specificially by the similarity of their brains to our brains, as that's where suffering manifests. So given this, the question isn't "Do plants suffer?", but rather "Is suffering even a meaningful concept when it comes to organisms that are as far removed from us as biologically possible without leaving the eukaryotic domain?".
Plants evolved a different means from animals of sending messages throughout their bodies, but using many of the same chemicals (including cortisol!) and to achieve the same goals for the same reasons. Something like a quarter of our DNA is the same as theirs. Plants are certainly capable of sensing and responding and communicating, manipulating their environment, and even moving around in some ways, and just as invested in their own reproduction. To me these facts suggest the burden of proof is on those who would claim that plants don’t suffer.
It would be intellectually honest for someone to simply say “I only care about the suffering of an organism with a face”, but nobody says that. Why not? It makes me ponder the reasons why people care about suffering. Also the ways in which we deal with cognitive dissonance.
Thinking about it now, it occurs to me that “caring” in general is not necessarily some kind of over-arching long-term project people undertake. There are lots of things we care about moment to moment, day to day. One minute we care about something, another minute we don’t, but if at any point we were to be asked if we cared about that thing we would look inward and come up with the answer that we do, even if we are trying to be honest—though it would not be accurate. That’s just the convoluted way our minds work.
I don’t intend this as ridicule at all: there is a performative aspect to caring about suffering. We are performing both for others and ourselves. The things we care about thus are cast in a role within that performance. We are more accustomed to some subjects of caring being cast than others. I think part of the hubbub over shrimp is they are like all-female ghostbusters, violating a casting norm just enough that some people can be on board with it and others can’t. Plants violate it even more. But I think just as the basic plot of the female ghostbusters movie is the same as the Bill Murray movie, the mechanics of the caring about suffering, the arguments for it, the issue of suffering and industrialized predation and so forth, still apply.
> Plants evolved a different means from animals of sending messages throughout their bodies, but using many of the same chemicals (including cortisol!) and to achieve the same goals for the same reasons.
This is merely spatial transmission of sensory data. Suffering manifests in your brain, not in the nerves along your arm. If I connect a temperature sensor with a wire to an op-amp, an LED, and a battery, this also constitutes "sending messages", but neither the sensor nor the LED will suffer when I put it in the oven.
> Something like a quarter of our DNA is the same as theirs.
That's an incredibly small overlap. This 75% difference leads to plants having completely different lifecycles, developing completely different body plans, having completely different organs, using completely different energy sources, having completely different reproduction, and exhibiting completely different intercell structures. And yet, somehow, _suffering_ is supposed to be the one thing we have in common with plants? When suffering happens in an organ which isn't even present in plants?!
> It would be intellectually honest for someone to simply say “I only care about the suffering of an organism with a face”, but nobody says that. Why not?
Because I only care about the suffering of an organism that actually suffers, face or not. Plants are biologically incapable of suffering in any sense that would be ethically meaningful.
If suffering can only happen when the signal-sending cells are bunched together in one place (i.e. brain), might there be other criteria as well, like there needs to be a certain quantity threshold of those cells, arranged in a certain way…? It seems to me your parameters are arbitrary or at least motivated by the conclusion you want. All of life is connected, and that 25% shared dna figure is an average—there are some non-animals with whom we share much more of our genetic code. People used to be convinced that cows don’t suffer. Heck, forget animals, there are still white people who don’t think black people feel as much pain as they do. Etc.
Finally you refer to some kind of suffering that is “ethically meaningful”, as opposed by implication to suffering that is ethically meaningless. I doubt a plant considers its suffering ethically meaningless, but it’s a convenient distinction for a human to draw!
Some people don’t take this seriously. But others do. Still, regardless of most of the details, the recommendation is the same - eat low on the food chain.
But if you think plants suffer, that only makes factory farming look even worse, because we feed the majority of our crops to animals that then get fed to humans. If you think plants suffer and you want to minimise the suffering required to feed us, you should feed plants directly to people instead of spending 80% of agricultural land on farming animals, which only produces 37% of our protein and 17% of our calories.
I agree with Melvin’s comment about needing a better model of plant suffering. When I cut a crisp stalk of celery on my cutting board, I’m pretty sure it’s alive while I’m cutting it, right? That seems worse than harvesting corn or wheat that have already dried out and turned brown on their stalks.
And, if we don’t grow plants to feed animals, that land would still likely be covered with plants that would still meet torturous ends. It’s not like the alternative to growing plants to feed animals is to turn that agricultural land into parking lots (which might be a still worse thing to do anyway).
I think you would need a better model of how exactly plants suffer in order to figure that out. How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
Honestly though, this is the first I've heard about the eyestalk ablation thing and I don't like it, I would be happy to have it banned, not on some kind of consequentialist grounds but because it's disturbing and in bad taste. I think I can come around to supporting greater farmed animal welfare from a virtue ethics sort of standpoint, we should treat the animals who serve us with dignity.
I can't get behind the idea that we shouldn't ablate prawns' eyestalks because it's the only way to get them to grow ovaries in the horrendous Indian/Vietnamese farms in which they live it because shrimp suffering is worth epsilon times my own suffering for some totally made up value of epsilon, but I can get behind the idea that we shouldn't because it's weird and gross and those farmed Asian prawns taste like crap anyway.
> How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
The apple is intended to be eaten, so this looks analogous to saying "how does a woman's suffering when a stylist cuts her hair compare to a man's suffering when he gets shot by a bullet?" I don't think you could have picked a worse question if you'd tried.
By this argument, childbirth should be completely painless, and yet...
[Quibbling about the use of the word """"intended"""" in reference to biology preemptively excised]
EDIT: actually the argument would be *stronger* for human childbirth, since an apple tree does not get any say in whether its fruits are plucked or not, so whether or not it suffers has no selective effect. Which is also why I think it's very unlikely that plants suffer -- suffering has no evolutionary advantage for them. Hyperselectionism is a known fallacy in evolutionary biology, but I doubt conscious experience appears as a byproduct of signal processing without any selective advantage for it.
I agree it’s not a good analogy. (I didn’t make it.) but to rescue it a bit, there are similarities between apple trees and farmed animals. Both are human cultivars, freaks who would not exist without us having bred them out of quite different ancestral organisms, and probably would not survive without continuous human intervention, whose phenotype has been massively distorted to suit our specific demands. There’s a moral component to that, right?
If some other organism liked to eat human hair, and it managed to breed humans that were maximally hairy so that we looked like giant walking piles of hair (only we couldn’t walk or even see anything or move under all that hair—eyes just take up space where hair could be so we are eventually bred to have no eyes, just very hairy permanently closed eyelids, in fact our eyelashes are considered a delicacy so some people’s faces are all eyelash), that would be bad, right?
> How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
Before we pick fruit, we wait until it is fully developed. We know fruit is developed when it reaches what we call the "breaker stage." This is the stage at which the fruit's internal connection to the plant's vascular system has been severed. You can tell this has happened because the fruit will just pop off with a gentle tug. It's like the difference between a loose hair that's ready to fall out and a healthy hair that holds on tightly by the root.
Less than 20 years ago a housing bubble popped and caused the Great Financial Crisis. When bubbles pop, a thing that everyone thought was expensive and valuable is revealed to be cheap and not valuable. So how come houses got expensive again so fast?
Yes yes, it wasn’t technically the houses that were bubbling but the mortgages. It still led to brand new houses being built everywhere and sold off at a loss. Why didn’t we get cheap housing for a while after the GFC, like we got cheap computing infrastructure after the dot-com crash?
Because there's no financial instrument to short housing, so prices can only ever increase until they burst. How fast the prices increase has more to do with the velocity of turnovers and the number of buyers than it has to do with the 'actual' value of the houses.
Doesn't this imply that there's no such thing as an underwater mortgage? Why would banks bother classifying mortgages as safe or risky if every house purchase was guaranteed to be either neutral or profitable?
Think people need to remember the "financial crisis" part. It wasn't just that people lost savings because home prices went down, there were a bunch of over-leveraged financial institutions and investors who were buying complicated financial instruments linked to the housing market. So even a temporary drop in housing prices causes a bunch of knock-on effects.
>"Less than 20 years ago a housing bubble popped and caused the Great Financial Crisis."
False.
Monetary authorities (esp. the Fed) cut the money supply causing NGDP to crater; the symptoms of the recession were then concentrated on the most rigid (in nominal terms) sector: residential mortgages.
Kevin Erdmann has an interesting hypothesis that everyone misinterpreted the Great Financial Crisis and that, effectively, there *wasn't* a bubble in housing prices. That was their true market price (due to zoning constraints creating a huge shortage).
This enormous shortage caused prices to rise so much that it finally managed to trigger a construction boom of single family homes (the only thing still legally allowed to be built). This led to people misidentifying the construction boom as a bubble and a large elite effort to kill the "unsustainable" level of housing construction. We succeeded at engineering a massive recession which did cause housing prices to fall (briefly) on account of massive unemployment reducing housing demand. Unfortunately, home prices quickly recovered as prices are a rationing mechanism and we still had a true housing shortage.
This seems too clever by half. In hindsight, the global financial market was dangerously overexposed to a bunch of correlated housing derivatives. Whether the marginal additional dollar of pricing signal was coming from Atlanta or Miami doesn’t change that several major companies went bankrupt and needed to be bailed out or liquidated with shocks around the world for the next half decade.
The thing is, banks did fail. Lots of firms went bankrupt. But mortgage backed securities never actually experienced substantial losses. We know this because the Federal government nationalized Fannie Mae and Freddie Mac and *didn't* suffer losses (that is, Fannie Mae and Freddie Mac remained profitable).
If the Mortgage Backed Securities actually were failing, the government *should have* taken a loss. Indeed, the whole point of nationalizing them was for the government to absorb the losses. But that... never actually happened and we just assumed that the mortgages were the cause of the banking crisis anyway.
Fannie and Freddie were exposed to better quality mortgages than the market as a whole. See page 219 of the Financial Crisis Inquiry Commission Report. Just one data point: in 2008 Fannie and Freddie mortgages were had a delinquency rates of 6.2%, versus 28.3% for other mortgages.
At the time that the government took over Fannie and Freddie, both had positive net worth, but their losses meant that their leverage was quite high, making it unclear that they could continue to borrow money at reasonable rates. So there was no reason for the government to suffer any losses in the absence of further adverse events. That doesn’t mean that the Fannie and Freddie didn’t suffer losses in the financial crisis; it just means that their losses hadn’t been large enough to wipe out their entire net worth.
In summary, Fannie and Freddie managed to largely but not completely escape the fallout from the mortgage backed securities collapse by mostly avoiding the worst mortgages, but they never the less suffered losses large enough to prevent them from continuing to operate as independent companies. I don’t think this is consistent with your claim that “mortgage backed securities never actually experienced substantial losses.” I certainly don’t see how it is supposed to *support* that claim.
In a liquidity crisis, is it that surprising if the lender of last resort makes a profit buying up distressed assets? Iirc the government also made money on hedge funds. The problem wasn’t the balance sheets, as you say; it was overexposure to systemic risk that led to a drawdown and sudden demand for liquidity. You could be sitting on a mountain of umbrellas that are “objectively” worth hundreds of thousands of dollars, but you might have to sell at a loss during a drought to get food, if that analogy makes sense.
> In a liquidity crisis, is it that surprising if the lender of last resort makes a profit buying up distressed assets?
Well, if it's a liquidity crisis instead of a solvency crisis then you're right. It's unsurprising.
But it's my understanding that the Great Financial Crisis (GFC) narrative is that the building boom led to a solvency crisis. Homes were built that were unneeded, so the underlying loans were insolvent.
But if the assets were solvent, that implies that the root cause of the GFC *wasn't* the building boom, right? Because the underlying assets were solvent! If the underlying problem was a liquidity crisis, it means the issue was Federal Reserve mismanagement of liquidity in the financial system.
> Homes were built that were unneeded, so the underlying loans were insolvent.
I don't follow your logic here. The US has been under-building housing for decades; I doubt there were homes that weren't needed. Rather, there were homes that were way too expensive to be justified by the incomes of the people owning them. They were insolvent for that reason, not because we had more housing units than people all of a sudden.
And honestly, what building boom? There had been a steadily increasing number of new units being built per population since about 1990, but in comparison to the prior 30 years it's pretty pathetic: https://fred.stlouisfed.org/graph/?g=1OnVx
I believe that liquidity and solvency problems fed into each other. This is why one of the first things Bernanke did was try to find a price floor for distressed assets. Whether banks were technically solvent on paper or not, an inability to accurately price assets caused institutions to hoard cash and prevented banks dependent on short-term loans from refinancing their debt. This is why the Fed needed to inject liquidity into the banking system (swap lines, money market funds) alongside propping up demand for assets and equity (TARP).
A priori, I'm not sure how big of an effect you would expect or how long it would take to bottom out and then to start to recover. I guess it is at least a little surprising that it never got lower than about 10% more expensive than it was pre-bubble (inflation adjusted). Also, I'm not sure how much this price decrease reflected availability, since mortgages became harder to obtain.
In any event, the valuable part was actually the land. You can justify much riskier mortgages if the price of housing is going up, since you don't have to rely on the borrower's income to recoup the loan. But the value of the building doesn't go up (compare to cars--obviously these get cheaper as they become more heavily used, lack the most current features, etc). Pretty much the entire price increase is explained by the increasing price of the land, which in turn is consistently pushed by because A) people need to live somewhere, and B) most of the US bans reasonable land use. Neither of those things changed just because the housing market collapsed. Also, I don't think the GFC would be expected to impact rent, except to the extent that some people would buy instead, but it's not like the actual supply of housing went up (if anything it probably went down).
Overall I think the GFC might be both a "policy bubble" (in the sense that the price is largely driven by high demand and limited supply) and a "speculative bubble" (in the sense that the price goes up because people expect it go up, which reinforces those expectations).
The average-looking national aggregates in 2012 were actually an average, roughly speaking, of low-tier housing in places like LA that was 100% overpriced because of the shortage and 50% underpriced because of the mortgage crackdown, and low-tier housing in places like Atlanta that had never been overpriced, but was now 50% underpriced because of the mortgage crackdown.
Prices in a few places were unsustainably high in 2006 (Arizona, Florida, etc.) mostly because there had been a mass migration to those places from the cities with binding housing shortages. That reversed. But, on net, the entire loss of $5 trillion across the country, in places like Atlanta, was not a reversal of anything that had come before, and was due to tightening lending too much at the federal agencies, well after the private securitization boom was over. Because of the moral panic about the lending boom, all credit tightening and deep drops in home prices where they had never been elevated, were, a priori, blamed on the lending boom.
thank you for your answer! can I ask you more questions?
Why do you think the actual supply of housing didn't go up during the bubble? My impression of America's zoning issues is that they make every new build x% more expensive, rather than putting a hard cap on the build rate. So increasing the profit for new houses should still lead to an increase in housebuilding.
If, as you say, the market for houses collapsed but the market for land was unaffected, why did the GFC produce such a shock to banks? Was the lost value purely speculative?
I think it is actually a hard cap, or very close. Housing supply is limited by policy that effectively bans building housing above a very low density in most of the land area in and around towns and cities. In some places it also substantially increases the cost of even that low-density housing, but municipalities can effectively make the cost as high as they want so there's not actually any profit to charging more. However much the builder could charge, I assume that local residents could make it instead by just letting their home prices appreciate.
> If, as you say, the market for houses collapsed but the market for land was unaffected, why did the GFC produce such a shock to banks?
Like I said, I think there were some of both (also, I'm not a real estate expert, but I doubt the market was unaffected, it just has kind of a high price floor under current policy). Banks assumed house prices would keep going up because they had previously gone up, even though there's only so much that people can afford. And there were actually a lot of people with mortgages that were only justified because of expected price increases rather than by their incomes.
> municipalities can effectively make the cost as high as they want so there's not actually any profit to charging more
Excuse my lack of clarity; I meant that developers expect more profit per house when mortgages are generous and would thus lobby harder against restrictive zoning. I wasn't alleging that zoning restrictions are due to a profit motive.
>However much the builder could charge, I assume that local residents could make it instead by just letting their home prices appreciate.
I don't believe this follows. There are three ways residents could increase house prices:
1. Investing in their communities, making them so attractive that demand to live there goes up. Not super relevant here
2. Supporting low-density policies that constrain supply. This increases price in the short-term, but also increases the pressure to defect by overturning the zoning or seeking exemptions to it. However much appreciation you can get on a single-family lot, the price jump from permitting multi-family or assembly projects is many times greater. So homeowners are not a stable profit-seeking coalition.
3. Supporting design bylaws or HOAs that make housebuilding more expensive. In my opinion this is a strategy for homeowners to capture a larger share of the surplus value of new builds, rather than a way to increase absolute profit. For instance, requirements that all new houses must have electric furnaces, or 2-car garages benefit the neighborhood more than the end buyer (if the end buyer thought it was worth the extra cost there would be no need to mandate it).
All in all, I don't think the game theory really works out to make NIMBYism the most profitable strategy for homeowners. It's more accurate to model NIMBYs as motivated by non-financial desires like local character or crowding concerns.
> Was the lost value purely speculative?
I meant the same as you when you said "mortgages were only justified because of expected price increases". The underwater part of each mortgage (the disappearing value that prompted the recession) was due to a disproven expectation rather than a real asset losing its value. I'm not positive it makes sense to draw this distinction, but it does feel like an answer to my original question about why house prices didn't drop more. In short: land never stopped appreciating; we had a temporary price spike due to financial shenanigans; the actual housing boom caused by the price spike was either non-existent or tiny compared to the underlying trends so there was no housing glut.
> Excuse my lack of clarity; I meant that developers expect more profit per house when mortgages are generous and would thus lobby harder against restrictive zoning. I wasn't alleging that zoning restrictions are due to a profit motive.
I mean, I think they are due to profit motive, on the part of local homeowners. Developers probably would like construction to be easier, but local homeowners have a lot more votes.
> So homeowners are not a stable profit-seeking coalition.
I think it's stable enough. Only 1 person having an exception is not stable (no one except that person would vote for it). Everyone being restricted is pretty stable. And the fact is that "home values" are one of the most commonly cited reasons for preventing new development. And I'm not saying it's the *only* motivation, just that I don't think there's much incentive to allow development at a slightly higher price point when they can instead capture all that value for themselves.
> I'm not positive it makes sense to draw this distinction, but it does feel like an answer to my original question about why house prices didn't drop more. In short: land never stopped appreciating; we had a temporary price spike due to financial shenanigans; the actual housing boom caused by the price spike was either non-existent or tiny compared to the underlying trends so there was no housing glut.
Have either of you two read Mike Bird's *The Land Trap*? It discusses the history of real estate bubbles. I wrote a review of it here if you want a summary:
A very contentious spat between hereditarians and statistical geneticists like Sasha Gusev broke out this holiday weekend on X. It all started with something Rebecca Sear said in an interview...
Sear said eugenics wouldn't work because the traits they want to add or remove "are too complex to breed in or out of populations. And we don't really know what would happen if you tried to select for some of these traits. You might get all sorts of unintended consequences ... as you do with farm animals and dogs."
Full YouTube video here (note, I didn't listen to the entire interview)... https://t.co/sBBUPv2iKp
Philippe LeMoine dismissed her out of hand, calling her clueless. Damien Morris called her ignorant. Sichu Lu, an account who seems to have some genetics chops, said she was retarded. Sasha Gusev leaped to her defense.
AFAICS, Sear's first point, that we can't breed polygenic traits in or out of populations, is supported by polygenic liability threshold (PLT) models. These models were purely theoretical when they were first formulated in the early 20th century. But more recent GWAS studies have supported this theory (for instance, Huang et al., Nature 2024, "Examining the role of common variants in rare neurodevelopmental conditions", https://www.nature.com/articles/s41586-024-08217-y).
Sear's second point, that we'd get unintended consequences if we tried to breed out polygenic traits, is supported by the antagonistic pleiotropy hypothesis (APH). (described here in Wikipedia: https://en.wikipedia.org/wiki/Antagonistic_pleiotropy_hypothesis; and supported by studies such as Long & Zhang, 2023, "Evidence for the role of selection for reproductively advantageous alleles in human aging", https://ouci.dntb.gov.ua/en/works/l1waE8Z7/).
The only technical arguments I've heard against Sear fall into two main categories.
1. Yes, we can select for polygenic traits. We do it all the time in domesticated plants and animals.
2. Evolution through natural selection is a given, and it's obviously affected the frequency polygenic traits.
The 3rd argument against Sear is that she's politically motivated. Which is probably true, but doesn't affect the technical arguments except to cast shade on her.
The best example that I found for both PLT and APH in action is thoroughbred racehorses. Yes, the winning speeds of thoroughbreds have increased by ~0.11% per year from 1997 to 2012. But thoroughbred racehorses have higher rates of congenital diseases due to inbreeding, with studies showing 8-14% limb deformities and overall foal morbidity up to 88%! (I need to check that statistic, but I haven't had a chance to dig into Google Scholar).
Breeders have used genomic selection models to reduce the incidence of congenital deformities in thoroughbreds. Studies have shown that harmful haplotypes on ECA14 contribute to some of the risk, and they're trying to identify them in breeding horses and select against them. But under liability threshold models, fully eliminating ECA14 haplotypes is challenging. The trait is polygenic, with liability distributed across multiple SNPs, making complete eradication difficult without reducing genetic diversity further.
And the argument that evolution happens despite PLT and APH assumes that the human genome would remain static if removed from selection pressures. But it would not remain static, because the human genome mutates at a rate of 0.5–1.5 × 10⁻⁸ per generation. With ~3.2 billion base pairs, that translates to ~30–70 new mutations per child. Depending on the study, most of these mutations are neutral, with about 5-20% mildly deleterious, and less than 1% being strongly deleterious. And Less than 1 in 10K are beneficial. So selection can happen slowly because mutations over many generations delete genes and insert new genes into genomic deck, reorder the polygenic relationships.
Finally, the political argument. Sear is now famous (or infamous among the HBDers) for proving that Lynn and Vanahan's National IQ study was specious to the point of intentional fraud. And Retraction Watch has been publicizing her attempts to get the studies that rely on Lynn's bogus data retracted. She's identified about 175 and is starting to gain traction on her retraction crusade. Yes, this is partially political. But it's rich that the people who accuse her of being political fail to remember that Lynn was unabashedly political when he called for complete separation of the races and created a largely bogus database to support his crusade.
Happy to hear contrary views. But please argue with data and the theories.
I have only tangentially followed this discussion, but I think Sear's claim is valid insofar as this is extremely hard to do on a society-wide level. I've given the example before of how the Nazis killed all German schizophrenics for eugenics reasons, and it maybe decreased the schizophrenia rate of the next generation by 1% or something undetectable like that. See https://www.astralcodexten.com/p/some-unintuitive-properties-of-polygenic. This isn't just about schizophrenia being a threshold trait, at some point I calculated out what would happen if you tried something similar with criminals, and it would take executing every criminal for thousands of years just to lower genetic propensity to criminality 1 STD.
It's invalid insofar as sure you can do this, as long as you're willing to do extreme things and have lots of generations to work with. For example, if you banned normal reproduction, and said women could only get pregnant by inseminating themselves with Einstein's sperm (which you had on hand in some massive cloning vat), the next generation of humans would be much smarter than the last one on average. If you repeated again with the smartest Einstein-child for the next generation, they'd be smarter still, and so on. You can do something like this in farming, and it works fine for producing very large cows, although depending on exactly how you do it you might get some genetic diseases unless you work hard to avoid this (I don't think the work would be very very hard, , but it's not really worth farmers' time at this point).
>The best example that I found for both PLT and APH in action is thoroughbred racehorses. Yes, the winning speeds of thoroughbreds have increased by ~0.11% per year from 1997 to 2012. But thoroughbred racehorses have higher rates of congenital diseases due to inbreeding, with studies showing 8-14% limb deformities and overall foal morbidity up to 88%! (I need to check that statistic, but I haven't had a chance to dig into Google Scholar).
This seems mostly irrelevant, since nobody is proposing inbreeding, which is the mechanism causing the problem here. Furthermore, as far as I understand your argument, there is little direct evidence here for PLT/APH, you merely claim it would be hard to get rid off due to thresholds?
On Sear's argument itself, the problem with farm animals is that we don't really care about their health; In fact we want to maximize only a very small number of traits which obviously trade off against health, especially in the tails. Of course maximizing fat - in other words, obesity - in chicken is not going to be good for them.
Eugenics in humans is very, very different, and includes a wide range of both possible targets and possible approaches. Especially, we can simply start by optimizing for health-related conditions directly, which is already done in some cases. Monogenic rare diseases have of course been done for quite a while, and the currently available PGS embryo selection companies are looking at fairly uncontroversial common diseases and most importantly use risk/threshold designs themselves, so it's not really a problem for them, this is literally what they are already doing. Take a look at, for example, lifeview's EHS. In addition, when they looked into the relationship between different health measures and diseases, they didn't find trade-offs, they mostly found (weak) positive relationships: https://www.nature.com/articles/s41598-022-22637-8
This actually makes a lot of sense, and there is a wide literature on both the phenotypic
comorbidity and the genetic shared architecture of many different diseases. Structurally, this is also equivalent to classic evolutionary selection but without all the horrible parts it normally entails. There is very little evidence why such an approach should increase the risk for any particular disease, and even if we can find certain cases, very little reason to believe that the negative externalities will outscale the benefits. APH, btw, is just one of several hypothesis', primarily for aging, and even in that field I wouldn't consider it the most explanatory (which imo is mutation accumulation).
Sear's argument itself only fully applies for a very narrow band of approaches which are similar to what we have done with animals, such as IQ-maxxxing with complete disregard for health, including incest to really double-dip. But nobody is actually doing this; Steven Hsu is sometimes associated with that approach (minus incest), but he is actually one of the founders of lifeview/genomic prediction, which is not even looking at IQ so far at all. There will surely be some crazy online people somewhere advocating for it, but none of the people in this space actually building things is attempting it, and Sear makes a far stronger claim, namely that eugenics in general doesn't work.
>Finally, the political argument. Sear is now famous (or infamous among the HBDers) for proving that Lynn and Vanahan's National IQ study was specious to the point of intentional fraud. And Retraction Watch has been publicizing her attempts to get the studies that rely on Lynn's bogus data retracted. She's identified about 175 and is starting to gain traction on her retraction crusade.
Short version, if we were to take Sear's argument serious, we'd not be allowed to use large parts of the international data. All the issues they identify are general problems with third-world data, which are only ever raised if someone doesn't like the conclusions. Worse, most newer datasets - which don't have the issues Sear raises -come to congruent results anyway.
The paper found that if you use modern exam performance data, you will get mostly similar results to Lynn. It doesn't methodologically rely on Lynn at all, however.
Ostensibly due to misrepresenting his research on the application to the NIH; However, as a researcher in genetics myself, he was if anything very forthcoming and specific. I've diverged far more from my initial statements with nobody ever complaining and this is generally well understood in the community as perfectly fine, but apparently only as long as you don't get controversial results.
It's extremely light on data, and very heavy on invective. It's particularly telling that she doesn't actually propose an alternative, better dataset, which would be the normal course of action in these kinds of controversies. Like her co-author Bird, she just wants to shut the research down entirely. It's perfectly fine and reasonable to be against Lynn's political views, but he still build up an unprecedent dataset that holds up surprisingly well.
Frankly, there is just nothing here except a classic witch-hunt. First you claim that a dataset is bad because of faulty methods and a plausible bias due to researcher's own stated views (fair enough, I'm all for improving standards!), then people use other approaches without the issues you named and show the original dataset is not particularly biased in any direction... and then you go after them, anyway. That's not the conduct of anyone interested in science. And again, I want to mention here, we don't do this ever in any other case; Nobody tries to outright get David Graeber's research and books retracted, for example, despite him openly stating that he considers it an important vehicle to spread his political (anarcho-communist) views and has frequently generalized small studies far beyond their applicability.
> This seems mostly irrelevant, since nobody is proposing inbreeding, which is the mechanism causing the problem here.
But what mechanism are they proposing? Again, if you accept the implications of liability thresholds and antagonistic pleiotropy, then selecting polygenic phenotypes to be bred in or bred out of a population may move the needle a bit, but you can only take it so far before the negative consequences start to outweigh the positives. The hereditarians hold up Ashkenazi Jews as proof that one can select for IQ. According to the hereditarians, this is the result of endogamy and preferential marriage to the smarter ones. And, yes, the mean IQ of Ashkenazis falls into the 107–115 range. But they suffer higher rates of both monogenic disorders and polygenic disorders (such as IBD). Endogamy is just inbreeding over many generations where people who frequently share the same ancestors 5 or 6 generations back marry each other.
> Then people use other approaches without the issues you named and show the original dataset is not particularly biased in any direction...
AFAIK, no one has performed any multi-national IQ surveys a la Lynn and Vanahan. And more rigorous IQ studies have failed to reproduce Lynn and Vanahan's national IQ scores for various countries. For instance, Lynn and Vanahan claimed the average IQ for Thailand was 91. Current studies show a median of between 98-102. They assigned Nigeria national IQ of 69. While Wicherts et al., concluded that because of these flaws and biases, it is not possible to derive a scientifically reliable, comparable “national IQ” for many sub‑Saharan countries (including Nigeria). Using the available data, looking at some less-than-perfect sources suggests that Nigeria is doing OK IQ-wise. A for-profit online IQ testing company shows that the average IQ of Nigerians taking their exam was 91.4 (https://iqexam.co/countries/nigeria-average-iq). The average IQ of students entering Lagos State University is 97.7. Contrast this with a report (I didn't check its validity) that claims the average IQ of US students entering college has dropped since the 1940s and the average is around the US mean of 100.
I seem to recall the Ashkenazi had and possibly still have some sort of breeding register to try to avoid their various exotic congenital diseases. If so, I assume there are Ashkenazis here who can say more on the topic.
There's an organization called Dor Yeshorim that runs a confidential genetic screening program. "Before a potential marriage match is finalized, the couple can check their ID numbers with the organization to see if they are genetically compatible. The specific carrier status of an individual is not revealed, only whether the match is advised against, which helps prevent children from being born with these diseases."
As I wrote, the most commonly proposed mechanism is embryo selection, which has none of the drawbacks mentioned here.
>AFAIK, no one has performed any multi-national IQ surveys a la Lynn and Vanahan.
Not to the degree they had, no, which is why it was still in use. But there are plenty of country-specific studies, and some looking at related concepts such as exam performance, such as the Pesta et al. study I linked.
>A for-profit online IQ testing company shows that the average IQ of Nigerians taking their exam was 91.4. The average IQ of students entering Lagos State University is 97.7.
No offense, but that's precisely what I'm talking about with respect to Sear's criticism; Neither for the company nor for the university average is there any reason to believe this is representative; Lagos State university has 35k students and is either the top or second best school in the entire country, Nigeria has a population of ca 240 million at the moment. You'd expect their average to be several SDs from the average citizen. The US mean you cite is afaik for public colleges, where just about everyone can enroll, so you'd expect it to reflect the population mean. The equivalent would rather be Harvard student's average IQ, which incidentally is ca 120-130, possibly higher.
Lynn's data is, if anything, more representative. None of this would be allowed to be used if we took Sear seriously!
Again, you're making unsubstantiated claims to support your argument. For instance...
> The equivalent would rather be Harvard student's average IQ, which incidentally is ca 120-130, possibly higher.
Did you check the validity of that statistic before you wrote it? I can't find any systematic IQ studies performed on Harvard students using randomized sampling. I found a study published by Shelley Carson, a professor at Harvard, who gave a selected group of highly creative students (n=25) a short-form WAIS-R test (Vocabulary + Block Design), and her study found that their mean IQ was 128. Secondary sources garbled the study and applied it to the entire Harvard student body, and that seems to be the source of the chestnut that the average IQ of Harvard students is between 120-130. AFAIK there have been no systematic IQ studies using a random sample of students done at Harvard. If there are, please share them.
There are some other studies that estimate Harvard IQ from SAT scores. And the one thing that IQ correlates well with is SAT scores. How much of that is due to g and how much to study is a contentious issue.
So, I do not doubt that the average IQ of Harvard students is in the range of 2 SD above the US mean. After all, Harvard probably is one of the most competitive schools to get into in the US, and many people with very high SAT scores get rejected (interestingly, Harvard doesn't track SAT score averages, but only 54% of the 21-22 applicants provided SAT scores). In the 21-22 application cycle, Harvard received ~61K applications, and it admitted 1,962 students—an overall acceptance rate of about 3.19%. Of those accepted, it looks like 48 were from Nigeria (but I'm having trouble verifying this, though—and it's likely some of these are grad students).
OTOH, up until 2024, there were between 50 and 55K students from low-IQ Nigeria studying at universities in the UK (undergrad plus grad programs). Beginning in 2024, the UK began tightening its visa requirements, and new students accepted from Nigeria have dropped by a whopping 60% compared to previous years. But it's clear that Nigeria was exporting its best and brightest students to the UK.
> Lynn's data is, if anything, more representative. None of this would be allowed to be used if we took Sear seriously!
To quote the old song, "Saying it don't make it so." Using the Nigerian example, Wichert noted that Lynn cherry-picked the three studies that reported the lowest IQ scores. Then he picked apart the methodology of those studies. Also, it looks like Lynn ignored between 6 and 9 other studies that gave higher IQ numbers for Nigerians. Wichert is a hereditarian when it comes to IQ, and even he thinks Lynn was full of shite.
Lynn is a fraud. Anyone who uses Lynn's data knowing that it's bogus is a fraud. Sears was not the first to voice criticisms about Lynn's national IQ database, but she has done a great service by compiling all the problems with it. You might call it "cancellation," but she and Retraction Watch are trying to get all the bogus studies that Lynn's fraud generated retracted. Good for her.
Sorry, but this is just getting ridiculous. Your original post was full of unsourced claims (such as the horse breeding stats) and baseless assumptions (such as talking about the relevance of inbreeding for human eugenics). In a comment section that is to be expected, so I engaged in good faith, explained that nobody is in favour of inbreeding and mentioning some other approaches that actual people do advocate for, and also going so far as to adding links that frankly should have been in your original post (such as Sear's Retraction Watch article). You respond by ... talking about inbreeding again (?!) and also making (unsourced) claims about the average IQ of Lagos State University, with an (unstated & baseless) assumption that this is somehow representative of Nigeria as a whole, a country still struggling with basic literacy & numeracy. Again, I didn't make a big deal out of it in my answer.
If you think that a claim needs a source, you can simply ask me. I didn't expect the Harvard stat to be contentious, so I didn't add a link, just as you didn't add a link for the Lagos State University IQ, and as it turns out, you don't even disagree! So it's unclear why you even mentioned the LaSU average in the first place, you should be aware of how wildly biased it is.
Even worse, you also make unsourced after unsourced claim in this very post! As far as I can see, Wichert is in fact not a hereditarian by any usual definition of the word; He is constantly stressing that high heritability can be caused by cultural factors, never gives any clear estimate he considers correct, and most of his papers I can find are explicitly phrased in opposition to hereditarians (especially Lynn).
Nevertheless, you may be surprised to hear that I have relatively little gripe with Wichert. His criticism is mostly fair and even-handed, and he still comes to an imo very plausible estimate of close to 80 (again he is stressing that it needs to be seen in the light of the Flynn effect and pro-western bias of the tests, making rather clear where he stands).
He did precisely what I consider the correct course of action: He proposed an alternative, argues for how it is better and recommends the usage of his own estimate instead of going straight to cancellations or calling names of people who disagree with him.
No offense, but I will not be engaging any further from here.
Still waiting for the data on Harvard student IQs that I asked you for. If you've got it, I'd be sincerely interested in seeing it, because I can't find any studies other than Carson's N=50 study of creative students.
> Sorry, but this is just getting ridiculous. Your original post was full of unsourced claims (such as the horse breeding stats) and baseless assumptions (such as talking about the relevance of inbreeding for human eugenics).
This is getting the pot criticizing the kettle's melanin quotient. But here you go. Some bedtime reading material for you...
And BTW, I explicitly asked you to back up your Harvard claim, and you never did.
----------------------------------------
Re: Horsies...
“Racehorses are getting faster” (Sharman& Wilson, 2015)
“Multiple Genes Related to Muscle Identified through a Joint Analysis of a Two-stage Genome-wide Association Study for Racing Performance of 1,156 Thoroughbreds” (Shin, et al, 2014)
“Origin and spread of the 1278insTATC mutation causing Tay-Sachs disease in Ashkenazi Jews: genetic drift as a robust and parsimonious hypothesis” (Frisch, et al, 2004)
Re: IQ in deepest darkest Africa. From 2015-present, there've been no nationally representative, peer-reviewed IQ survey for Nigeria. Seems like the genetic racists would be all this, because with a nat'l IQ of 69 (per Lynn), Nigeria is their favorite talking point.
“Raven's test performance of sub-Saharan Africans: Average performance, psychometric properties, and the Flynn Effect” (Wicherts, et al, 2009)
My sense is that the whole argument just isn't presented around a relevant crux - it's just a poorly-framed argument.
Polygenic liability threshold (PLT) models show that there's a liability for a certain trait that only becomes a visible trait once you cross a threshold. That liability comes from a combo of rare, high-impact variants and the larger polygenic background. It implies that you can remove some rare variants in a single embryo, but it's more difficult to remove from a whole population. You can't mechanically/deterministically "breed out" a complex polygenic trait, but you can reduce risks probabilistically. New mutations might emerge such that we can't reduce liability to zero, but I think we could probably map out a "selected" pathway that consistently reduces risk every generation (if we do consistent embyro selection).
On the idea that we get unintended consequences, it's obvious that we can, and this is very well-documented in companion and farmed animals. But the relevant question is not "will there be unintended consequences?", but "will the unintended consequences be worse than the benefits?" , "what are the trade-offs" etc.
I get the sense that the actual cruxes are things like:
1) What's the effect size ceiling of polygenic embryo selection (when do we get decreasing returns)?
2) How generalisable are polygenic scores across populations?
>Yes, this is partially political. But it's rich that the people who accuse her of being political fail to remember that Lynn was unabashedly political when he called for complete separation of the races and created a largely bogus database to support his crusade.
Of course, the problem with both sides being intensely political is that convergence on the truth is impossible - everyone is already convinced beyond doubt that their priors/intuitions are correct, and the only thing that matters is marginalizing opposition. Debunking bad studies doesn't help much if there are no forthcoming good studies to replace them, and where would those come from?
Although I have strong disagreements with Yudkowskian rationalists, I think that all Bayesian rationalists (including Yudkowsky), plus scientific realists, post-positivist naturalists, and even scientific anti-realists, would agree that experimental evidence is what provides us with at least an outline of Truth (or a reproducible story about Truth in the case of the anti-realists). I think they'd all agree that by proving the null hypothesis, we weed out false beliefs.
> Debunking bad studies doesn't help much if there are no forthcoming good studies to replace them, and where would those come from?
Network epistemologists and social epistemologists would likely agree with your statement. But being an old-school Popperian, my philosophy of science aligns with the post-positivists. Theories need to be falsifiable to be scientific theories. If a theory is falsified, we know it's wrong. If it's not falsified, it doesn't necessarily prove it's correct — but at least we've eliminated a false hypothesis, and we can more profitably spend our time pursuing other avenues of research.
Galileo proved objects of different weights fall at the same rate. By doing so, Galileo disproved the Aristotelian dogma that heavier objects fall faster, but he did not provide an explanation for this phenomenon. It took Newton to quantify the behavior. And it took Einstein to explain the behavior.
So, no, I don't think the debunkers owe anyone an alternative theory.
You are Popperian? How suprrising. Tell me which facts would make you to consider anti-HBD debunked. I would lose belief in HBD if Nigeria goes the path that South Korea/Taiwan went (and later China/Vietnam). When I see Nigerian students winning IMO and IOI and Nigerian made electronics. You?
IQ is pseudoscience. Other than SAT scores, it has as much correlation with life outcomes as phrenology or astrology.
But if we look at average IQ scores, the Flynn Effect is a worldwide phenomenon — no doubt due to improved health conditions, nutrition, education, and economic development. African IQ is rising at the same rate that Asia is, although Asia has had a 20-year head start. The US has seen a 3 pt increase in median IQ every decade up to 2010 (that's why they have to renormalize IQ tests every decade or so). US average IQ in the 1910s was what Lynn said Nigeria's is (~70). There's no reason to think that America's gains are due to genetic improvement. And there's no reason to think that modern Americans are any more intelligent than Americans were a century ago. And there's no reason to think the Flynn Effect won't continue in Africa for many decades to come.
From: "One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909–2013)" (Pietsching, et al, 2015).
I asked you what kind of evidence could have changed your opinion. You answered ANOTHER QUESTION (I asked nothing about IQ tests). You have nothing to do with Popperian ideology but in name.
Sorry, I thought you understood basic Popper, and you were setting a trap for me. ;-) According to Popper, you can debunk the theory being tested (=null hypothesis), but you'll never be certain you've proved the alternative hypothesis. "Positive results never conclusively verify hypotheses; they only fail to falsify them.” This tends to rub Bayesians the wrong way. But to Popperians, Bayesianism measures how confident you are, not how good the theory is.
But even though HBD is not a single simple hypothesis, I think the various null hypotheses that HBD theories encompass *have* been disproved over and over, but the hereditarians just don't want to listen.
Minor clarification. IQ tests were originally designed to identify the cognitively impaired. And they're still good at that. But above -1 SD they have very low predictive value (other than standardized test scores).
>So, no, I don't think the debunkers owe anyone an alternative theory.
I agree, of course, that debunking a bad study is a clear positive in and of itself (whereas publishing a bad study is a clear negative). My point is simply that in the grand scheme of things this is only a minor step, and there doesn't seem to be any prospect of substantial improvement any time soon.
Polygenic here just means "more than 1," right? Eye color is polygenic and a population could breed that in or out, against certain limits depending on which genes are recessive or not. You wouldn't need to identify which genes are at play or even know about gene theory to get some measurable results. Does that fit the model or not?
Funny you should ask. I was just checking out eye color. When I was in school (long before genome sequencing), it was of held up as an example of Mendelian inheritance, which would have meant it was a monogenic trait. But it looks like two genes control >95% of eye color variation. And one of the two is a regulatory region that controls the other. So, they're tightly linked.
And turns out that eye color does not perfectly follow Mendelian inheritance patterns, but they're pretty close. Since the trait is only moderately polygenic, You might be able to breed for eye color in a population without producing AP effects.
But even with massively polygenic traits, you *could* move the needle with selective breeding to reduce or increase the frequency of a desired trait. You just can't breed it in or out completely. And the more you push the needle in one or the other direction, the more unintended adverse effects appear.
Yes, Tay-Sachs is a single-gene disorder. Polygenic threshold limitation models wouldn't apply in the case of Tay-Sachs or any other monogenic disease.
Every polygenetic trait is a combination of many monogenetic ones. There's nothing qualitatively different, really. You just look at how the answer varies with X while you hold Y and Z constant.
Of course, the tricky part is that things don't have to be linear. Organisms are very complex and have lots of moving parts. So maybe removing X makes things worse, unless you also remove Y and then it makes things better. Same idea as if you have a mechanical loom and you make one part twice as big, that makes things worse unless you also adjust the other parts to match. Figuring all that out is tricky, but for now we can basically compare against the most common other values, to decide which gene variants we like or dislike. This is the same system used in selective breeding, both artificial and natural.
Frankly this whole thing sounds like a debate over the word "can". Can you jump over the ocean? Yes, but it's hard. But jumping over the first 3 feet of it is pretty easy.
I don't have the science chops to argue with you, but I want to ask: why does the argument against PLT and APH assume that human genome would remain static without selection pressure?
I don't think either theory assumes that. My argument was that if new mutations occur continuously, they will inevitably change, most likely damage, polygenic relationships and their liability thresholds, allowing natural selection to slowly favor or disfavor polygenic traits. That was my idea, and I should have noted that. My bad. Take it with a grain of salt. :-)
In a 80,000 hours podcast, Rob Wilbin suggests that dropping fertility rates are not a consequence of cost and expenses, but more so a consequence of individual choices where people pursue hobbies, interests, and careers over children. And cultural acceptance of this furthers the trend. And the rise of the phone and digital entertainment mediums slow coupling as stimulating ways of engaging without in-person-relationship-forming interactions. I found this perspective in the podcast fresh and it captured the tension between individual and community needs that continue to reshape our society and we lean more and more toward individual determinism. Give it a listen if you are interested in the fertility conundrum!
As Jack says, I’ve seen this explanation as one that makes sense of how “cost” could be an issue when people are getting richer - it’s really *opportunity* cost, and as opportunities increase, those increase.
Haven't watched the podcast but it seems like that's possibly just a restatement of the same thing.
Why would you pursue hobbies over children? Part of your answer has to do with the cost of each. If your hobby is "travel" then, how much are you able to travel if you don't have kids? With cheap airfare, the internet making booking shit easier, ability to work remotely for many people, the cost of "travel" as a hobby is way lower than in the past when "travel" as a hobby, beyond some local domestic spots, was out of reach for most people.
Same with the cost of having kids ... if you think that sending them to college is a requirement, you're basically saving significant money for it the moment they're born.
I don't think it really has anything to do with cost. I like donuts, I don't like flan. Not because of costs of one vs the other. Just because that's my preference. I think for many more people than in the past, they just don't want kids, same as I don't want flan (that's an example, I love flan AND donuts)
Why don't they want kids? Because culture has changed from the past. Until some time relatively recently (60s?) it was just expected, you become adult, you pair up, you have kids. All of that seems to have disappeared or is in the process of disappearing. You become adult, you play, you work, you play some more, you work some more. Kids, for many, are just not something they want. They aren't weighing costs, they're just not interested.
The same is somewhat true of relationships in general. They seem harder than ever to get into and many have just given up and/or are fine not trying to find someone. That was less true in the past.
I've wondered about this myself - what would happen if you eliminated the alternatives - and decided to do some research on it a while ago. I ended up writing a post on the effect of curfews on fertility, with a focus on the era of Latin American military era curfews:
I don't think it's just a restatement, it's a reshaping of the conversation away from purely economic costs and towards other costs, both the opportunity cost of having kids (which stop you from doing a bunch of fun things) and the other non-economic costs of having kids (they can be annoying!)
In your travel example the increased opportunity costs are indeed driven by the decreasing cost of travel, but in other cases it might just be social changes. For instance a big part of it is just social contagion.
I'll be in San Francisco shortly before the solstice festival for an onsite interview. Would anyone be interested in meeting up Tuesday afternoon or Wednesday evening?
Is there any evidence for “micro-optimisation” doses of psychiatric or hormonal drugs in healthy people?
Pharmaceutical drugs (SSRIs, testosterone, Ozempic etc.) have clear therapeutic effects at normal doses, plus predictable side-effects. It seems plausible that there might be a “tiny, subclinical dose” where you get a small benefit with minimal downsides, e.g., TRT microdose nudging testosterone from the 60th to the 62nd percentile without giving you any side-effects, miniscule mood improvement through a tiny dose of antidepressants, tiny Ozempic dose bringing average food consumption down by 50kcal a day etc.
Any good evidence, or anecdotes on this? I'm probably not considering it, just curious why I've never heard of people trying it.
IME, the effect of testosterone on mood is very subtle, so only slightly nudging it up is unlikely to make a difference. (As well as the problems trying to do this as Erica has explained)
I've flirted with the model where many psychiatric drugs work, not by treating an abnormality or disease, but by enhancing healthy people.
Of course, these two models are not INHERENTLY mutually exclusive. But society sure acts if they are. I think the disease model will brook no competition, institutionally speaking. And besides just withholding research funding, I suspect our institutions actively harry and obstruct anyone who dares to suggest that doctors should be in the business of enhancing healthy people.
But I don't actually know much about the state of medical ethics besides the occasional post from Scott or discussion in these comments. Is the situation as dire as I have made it out to be? Are models that allow for human enhancement gradually gaining ground or is there virtually no movement in that direction?
For testosterone in particular, I would expect no effect from microdosing because there are homeostasis mechanisms that would cut endogenous testosterone production to compensate. In a normal adult cis man, the pituitary makes secondary hormones (FSH and LH) to "order" a certain rate of testosterone production. The testicles respond by making the testosterone, which circulates in the blood. Some small percentage of the testosterone gets converted to estrogen (roughly proportional to testosterone levels), and estrogen sensors in the pituitary up-regulate or down-regulate FSH and LH production based on estrogen levels.
The practical upshots of this for exogenous hormone purposes are:
1. Macrodosing testosterone severely reduces endogenous testosterone production, unless you take additional meds to counteract it.
2. If your testosterone levels are low because your pituitary is regulating levels to too low a target level but your testes are healthy (i.e. secondary hypogonadism), you can boost T levels by taking meds that block estrogen conversion (aromatase inhibitors) or boost FSH or LH levels. Clomid is the most common drug for the latter purpose.
3. Trans women can often shut down testosterone production almost completely by taking enough exogenous estrogen.
4. Both cis men taking exogenous testosterone and trans women taken exogenous estrogen usually experience testicular atrophy because of the lower FSH and LH levels.
I can't speak to psychiatric drugs, but I rely on prescription bioidentical hormones due to a genetic condition that makes me unable to make many hormones naturally.
I can say that nudging a sex hormone like testosterone by such a small amount (going from 60th to 62nd percentile) probably wouldn't be at all noticeable, at least from my experience. The human body is built to account for constant, mild fluctuations in hormones (especially women), so it tends to take a larger dose for your body to react to a noticeable degree.
This is why you see body-builders increasing their testosterone to insane levels; increasing their testosterone by 20% isn't going to make them 20% stronger. They probably would need to double their testosterone to get that effect (and even that may not be enough).
So I'm quite skeptical that a *micro* dose would have any significant impact. That being said, I have a keen interest in seeing some studies done on *low* dose HRT. I suspect low-dose therapy could be extremely beneficial to many people.
I personally know of three people who were diagnosed with psychiatric conditions, suffered for years with symptoms and harsh meds, later discovered they were low in progesterone or testosterone, and were able to cease all medication when they began taking HRT.
I'm actually writing an article about this at the moment... It's such an under-researched area of psychiatry, and I think it's criminal that it's not being explored more.
Thanks for the reply, interesting anecdote about the HRT and psychiatric conditions - I'll subscribe and look forward to your article.
I suppose I’d expected a micro-dose to produce a micro-effect. Some chemicals like caffeine or alcohol have a linear dose-response. A "micro-dose" of, say, green tea in the evening can give a slight, barely perceptible boost if you need to work late, without affecting sleep as a double expresso might. A small alcoholic drink can put you in a "sweet zone" where a few of your inhibitions are down but you haven't lost any clarity of thought.
So I was curious whether anything like that existed in the testosterone pathway. If it's possible, it might give you a tiny, barely perceptible strength/libido/energy boost (with tiny negative effects), which is probably what you should want if you're at normal levels to begin with. But, as Erica mentions above, it seems that isn't really how testosterone production works because of homeostatis.
I've known people to take 2.5 mg of Lexapro daily; that's 1/4 the standard dose. The general argument is that most drugs are taken at supra-optimal levels, because our Phase 2 / 3 trials system selects for the highest dose that doesn't cause extreme side effects.
There's been a lot of news recently about how bad UC incoming freshmen are at math, including those who had nominally passed honors calculus and/or gotten A's.
a) Q: Is light with a wavelength of 530.2534896 nm visible to the human eye?
results: "Yes, without question. That wavelength is green light, sitting comfortably in the middle of the visible spectrum (roughly 400–700 nm)."
b) Q: I have two solutions, one of FeCl3 in HCl in water, the other of CuCl2 in HCl in water. They both look approximately yellowish brown. What species in the two solutions do you think give them the colors they have, and why do these species have the colors they do?
results: Got all the important points - that the species are FeCl4- and CuCl4 2-, that they both have LMCT strong, allowed transitions in the UV to blue, which dominate both colors, that the Fe d-d transition is spin-forbidden and doesn't contribute to the visible color, that the CuCl4 2- d-d transition, while allowed(-ish, partially Laporte-forbidden) is down in the near-IR and doesn't contribute to the visible color.
c) Q: Please pretend to be a professor of chemistry and answer the following question: Please list all the possible hydrocarbons with 4 carbon atoms.
results: Missed a few of the 3-unsaturations species: There are a couple of other highly unstable species under C4H4 besides the tetrahedrane and bicyclobutene that you already mentioned. Taking cyclobutadiene that you've cited as the 1,3 isomer, a 1,2 isomer is formally legal as well, and a methylcyclopropanediene, anc cyclobutadyne.
d) Q: Does the Sun lose more mass per second to the solar wind or to the mass equivalent of its radiated light?
results: "Radiation wins, by roughly a factor of 2–3."
e) Q: Consider a titration of HCl with NaOH. Suppose that we are titrating 50 ml of 1 N HCl with 100 ml of 1 N NaOH. What are the slopes of the titration curve, pH vs ml NaOH added, at the start of titration, at the equivalence point, and at the end of titration? Please show your work. Take this step by step, showing the relevant equations you use.
results: correct at all three points, and stated a universally true equation for [H+] as a function of volume
f) Q: Please give me an exhaustive list of the elements and inorganic compounds that are gases at STP. By STP, I mean 1 atmosphere pressure and 0C. By inorganic, I mean that no atoms of carbon should be present. Exclude CO2, CO, freons and so on. Please include uncommon compounds. I want an exhaustive list. There should be roughly 50 compounds. For each compound, please list its name, formula, and boiling or sublimation point.
results: Very good, though not quite exhaustive. All the compounds it cited were valid. It (correctly) blew through my "roughly 50" to give 68 out of 97 possible. Actually a bit better than Gemini 3's 63 of 97.
g) Q: What is an example of a molecule that has an S4 rotation-reflection axis, but neither a center of inversion nor a mirror plane?
results: Fully correct, gave three examples, all of which are valid: A conformation-dependent tetraphenyl tetrahedral case with a "propeller" orientation of the phenyl rings, an alternating R and S chiral substituted tetrahedral case (which doesn't depend on a specific conformation), and a 1,3,5,7 tetramethyl substituted cyclooctatetraene
Re: transgender sports & biological advantages, this has long been my view, glad you wrote up something about this though it sounds like you take the other side of the argument (though I can't read the post, sorry for not subscribing). In particular, as someone who has never been particularly athletically inclined, there's no way I would ever be competitive with top female athletes even with the advantage of my male body.
My preferred solution to this is to just use what every competitive online multiplayer game has already figured out: matchmaking plus anti-cheat. You can use ELO or similar to update an estimate of player skill and assemble matches with a goal of each side having an equal chance at winning. As long as you also prevent cheating on whatever the agreed rules of the sport are, everyone has a fair shot to move up and down the ranks, and is matched with players of similar skill in competitive matches. Sure, the very top players likely have some unusual combination of biological advantages, but you can always make filtered leaderboards for whatever particular set of characteristics you care about (Top Women, Top Octogenarians, Top Asthmatics, etc etc).
The entire argument is about whether men should be allowed in the Top Women category, so this doesn't really solve anything. The premise of a women's only category is that women aren't competitive against men so they need their own league. If we accept the argument that we shouldn't care because every winner has an unfair advantage anyway, then there is no justification for women's sports existing at all.
It clearly does offer a solution. I'm not advocating for or against it, but just to clarify:
There would be no women's sports leagues. Anyone can be matched against anyone else with a similar skill level.
There would (or could) instead be women's coverage. A broadcaster might choose to cover all the top women in their matches. Each of their matches may or may not be against other women, but they'd be against someone close to their skill level. Interested spectators could watch the world's top female athletes compete.
It doesn't answer whether a particular broadcaster would include trans women in their women's category. Maybe it would be influenced by demand and the free market.
> Maybe it would be influenced by demand and the free market.
In theory, we could have two kinds of sport events, and let the market decide, which one survives, maybe both.
But it wouldn't work in practice, because instead of voting by their wallets, people would instead do political pressure on media to stop broadcasting the "wrong" kinds of sport.
You can have different lists, trans-inclusive and trans-exclusive depending on your point of view. There's no need for universal agreement because the list doesn't restrict who you are matched to play against, as that can be determined purely by skill.
I don’t think you understand sports. Or the arguments involved.
If all sex segregation is to disappear it’s not going to re-appear with trans inclusive tennis games at Wimbledon. People want to see the best, and the reason the women’s sports exist at all was a once understanding of biological difference. If that is to be forgotten or ignored then there would be a single sex competition and not a trans exclusive/trans inclusive division. Which would upset both sides.
If you mean best as in most skilled, then someone could easily fill that market by having a basketball league that only lets in people that are shorter than average. People would need to be skilled instead of just tall.
If you mean the best in general, then why have a women's league at all?
The best in general. There are Olympics for the disabled, and while admired it’s clearly not as popular.
> then why have a women's league at all?
Women wanted to compete. And women’s sports are just not that popular. Tennis maybe, or perhaps some sports where the female physique isn’t a disadvantage - like gymnastics.
What the argument here anyway, because we now trans we should have dwarf basketball?
I don't think there's a whole lot of controversy of trans men in men's leagues. It's really focusing on women's leagues. So this isn't so much a question of why people watch sports in general as women's leagues in particular.
> What the argument here anyway, because we now trans we should have dwarf basketball?
Is it a single list? The world record in marathon for women in mixed events is much better than that for women-only events. Including 2 cases of women holding both records.
The problem with this is something like "state" or "the olympics". Under your model, there would be zero women in the olympics, or very few in very non-physical sports (marksmanship type stuff?). Starting from highschool, you'd have male-dominated everything with occasional female-outlier people making it through, essentially eliminating girl's team sports. And then it would never recover.
I think the foundational place you start on the whole trans thing is pretty well proven to be "do you even care about women's sports existing?". People can answer different ways on that, but knowing the guy you are talking to is OK with women's sports taking any size hit to give the marginalized-group-de-jour what they want is relevant to know as compared to any other possible position.
I don't think it would necessarily hurt female participation in sports, and may even help by making it easier for everyone to compete in well-balanced matches. It's true that men would dominate at the very top, but the vast majority of people aren't anywhere close to the very top, and would get to compete with people of a similar skill level regardless of their particular biological advantages/disadvantages. Also in the school example, there's no need to restrict things by grade level; you might have women competing with slightly younger men for example.
Oh, it would hurt it a massive amount, with near 100% elimination of girls from team sports from the earliest points. You are thinking the differences between highschool girls and boys are small enough to be adjusted for with just a few years of age, and it's just not so.
To put this into perspective, the woman's national soccer team (an adult professional-level sports team) has LOST to a single academy's boys under-15 soccer team. The best female soccer players in the US, or nearly so, lost to a group of 14-year-olds from a single highschool.
To say that another way, if you say "Oh, the oldest girls can just play highschool sports with the youngest male highschool players..." you've just eliminated girl's highschool soccer, full stop. The male advantage is not small and competitive sports are not and cannot be set up like non-competitive video game matchmaking; there's just not the ability to make enough matches or enough supervision to be constantly running them.
Again, what you are proposing would start eliminating girls sports at approximately 100% levels just after the onset of puberty; having lost the training and facing gaps that only grow as they age, you will have also effectively eliminated women's sports entirely.
Doesn't matter, and that's sort of the point. If it was a smaller field, then men would use their still greater endurance to compete on it harder. If it was a contact sport, they'd be bigger, or else stronger at the same size. They'd be faster, blah blah.
You can try to find any sport you want, and if it has a significant physical element, men are going to have a massive advantage at it. Tennis? Men. Golf? Men.
I'm being forceful here, but what you said above is nonsense because the women don't "compete just fine" during some magic period before they get tired out. The men are stronger and faster during that time, too, hopelessly so, to the point where for women to get an even match they have to play against highschool JV squads to have a chance.
Does introducing other languages to small children who otherwise live in a monolingual environment have any effects? Having a kid now it seems like there's a much bigger trend towards bilingual or foreign language books. May just be a quirk of the books we've acquired, but I'd say like 5-10% of them seem to be some flavor of bilingual (read: spanish).
Is there any studies/info on what effects this has, if any? Does early exposure to other languages matter for any cognitive reasons? I'm guessing the effects would be small in either direction, but interested if there's any concrete data one way or another.
EDIT: Should clarify; while I worded my first sentence kinda broadly - I'm specifically asking about the effects of bilingual children's books, not a full "teach your kid a foreign language" program. A lot of the replies are addressing the latter, but my question is more about whether just reading simple bilingual books (e.g. "Besos for Babies" is an example of one in our collection) has any effects.
I don't know any studies, but I imagine that reading books is a very inefficient way to learn a language for a child, compared to talking or even watching cartoons.
I suspect this is just a status thing, a combination of "languages are high status" and "reading books is high status", therefore reading books in a foreign language is high status. You would probably achieve much more by showing the kids Paw Patrol in Spanish, because they would be naturally motivated to pay attention.
You're asking about what is called a "bilingual advantage" in the linguistics literature (https://en.wikipedia.org/wiki/Cognitive_effects_of_bilingualism). It's very clear that there are direct benefits of speaking different languages - for example, being able to travel more widely, read more literature in its original language, and experience more cultures. It's very unclear whether there are broader (non-linguistics) benefits of being bilingual per se.
There are robust studies that find neurological differences between monolinguals and bilinguals. The literature finding *behavioral* differences has not been treated well by the "replication crisis" of psychology (https://en.wikipedia.org/wiki/Replication_crisis): Many experiments which found benefits for bilinguals in the late 1990's and early 2000's were re-run in the 2010's, but found null results. It's possible that earlier works failed to fully correct for confounding variables which are highly correlated with bilingualism (such as socioeconomic status and level of education) which also influence other cognitive functions. Unfortunately, this topic has become politicized within the research community, in part because the research finding bilingual advantages was a response to an early generation of research which (likely incorrectly) found bilingual *dis*-advantages, so folks are sensitive to the pendulum swinging back too far. There's a quote from a researcher (Mark Antoniou) that I find particularly damning: "Those familiar with the literature are able to surmise whether the findings [of a study] will be for or against the existence of a bilingual advantage simply by peering at the list of authors." (https://www.psychologytoday.com/us/blog/life-bilingual/201906/the-bilingual-advantage-three-years-later). My personal take is that, if a bilingual advantage exists, it must be weak enough or specific enough as to make it irrelevant to any individuals making choices for their children.
All that said, I think it's a great idea to raise kids speaking as many languages as possible! The linguistic benefits are excellent, as mentioned by other commenters, even if the broader cognitive ones don't seem to be.
> There's a quote from a researcher (Mark Antoniou) that I find particularly damning: "Those familiar with the literature are able to surmise whether the findings [of a study] will be for or against the existence of a bilingual advantage simply by peering at the list of authors."
Why do you find that damning? You can substitute "the existence of a bilingual advantage" out for any question in any field and the statement will still be true. The only thing it relies on is that people continue doing the same things they've done in the past.
What I wrote was definitely misleading. In my head, "as many X as possible" was smuggling in "when not taking away from other important uses of time and resources," but that's obviously not how you'd normally read it. In context, I was trying to push against a monolingual default that I was worried my earlier claims would seem to support - again, there's a fear of the pendulum swinging back too far.
Having a private language ("idioglossia") definitely wouldn't give you the culture/travel benefits of learning other languages, for sure!
Attempting an Ideological Turing Test, I think that even the most gung-ho bilingual-advantage proponent would believe that your situation wouldn't have benefited you much. The classic explanation of the bilingual advantage is that a bilingual speaker is in a constant state of balancing their languages in their mind, and so trains their brain to more easily switch tasks or suppress irrelevant thoughts. In your case, or even in a more common case of an immigrant needing to switch entirely to a new language, I'd expect this balancing act to apply much less than in the case of someone who is actively using their different languages in different contexts.
There's pretty good evidence that learning languages as a kid makes you better at languages as an older kid or adult! Especially phonology/pronunciation, which are often quite transferable across different languages.
We're going for "trilingual by default" for our boy, but I can already see why any data collected from him will be confounded. If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two.
I'm actually tempted to teach our lad the whole phonetic alphabet, including those weird Xhosa clicks. It's only like 200 sounds, and you can probably master them all in a couple of weeks.
"We're going for "trilingual by default" for our boy, but I can already see why any data collected from him will be confounded. If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two."
From experience, I'd recommend limiting it to two. I tried to teach my kids three from birth but I just couldn't put in enough time to teach them the non-native language of the family (German) - ie. the one which neither me nor my wife speaks natively. We did make an absolutely no exceptions rule to only ever speak in our native languages to the kids (with the biggest burden being on me, as we live in my wife's country not mine, so they only got English from me - and the TV...).
I think it is probably better to get them really fluent at two languages first (and possibly also teach them the phonetics of the third language while they are very small - ie. before they are two - because that is definitely when they best learn that). This will be less confusing for them and learning a third language later in life will anyway be much easier once they already know two languages (due to loss of speaking embarrassment etc).
We're quite lucky in that we've got three "naturally occurring" languages (Mum's L1, which we speak at home- Chinese, the language of the country we live - French, and my L1, English, which we speak a lot with friends and family, and for work etc.). My sense is that we should prioritise Chinese at home, and neglect English, because he'll probably end up learning English no matter what, but we might adjust depending on how he turns out.
We have a friend who has trilingual kids without trying very hard (she taught them Chinese, French at school and with dad, and English became fluent through YouTube and gaming!), which seems pretty ideal.
Ah, yes, then your situation is a bit different than mine is (where mother and country language are the same) and I speak two other languages but was the only source of these for my kids (besides English on TV, Youtube, etc). So you will likely have much better chances of trilingual success. Wish you luck!
> If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two.
Language learning is one of those things that is notable for not being related to intelligence. (Not quite: people who become noted rhetoricians tend to be more intelligent. But idiots and geniuses achieve the same levels of fluency.)
If your child can learn one language, he can learn three languages just as easily. Though note that if children perceive a possibility of getting people around them to speak what the child sees as the dominant local language, they will try to do that.
> I'm actually tempted to teach our lad the whole phonetic alphabet, including those weird Xhosa clicks. It's only like 200 sounds, and you can probably master them all in a couple of weeks.
What do you mean? All children are born recognizing those sounds. It takes them more than a couple of weeks to learn to produce them. It takes them even longer to achieve the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds.
That learning will take place whether you want it to or not as long as the child doesn't have to use a language that draws the distinction.
Except of course that South African clicks don't even have that handicap; you can hear the difference between them just fine right now. Do some babbling with them and you'll probably figure out how to fit them into a stream of other sounds. Note that the dental click and the lateral click are frequently used in American English, and all native speakers are expected to be able to produce and recognize them (in isolation, not as part of a word).
I definitely disagree with the "language learning is not related to intelligence" claim (almost tautologically), but not sure what specific claim you're making.
On the phonetic alphabet point, I've learned a bunch of languages that use sounds we don't have in English - it's a cool skill that you can develop relatively easily as a kid, but gets difficult as you get older. After learning SE Asian languages, I can distinguish certain contrasts (like /d̪/ /ɖ/) that monolingual English speakers just can’t “hear”, even after instruction. A few years ago, did a pronunciation teaching course and was surprised to find that all Anglophones except me couldn’t identify the difference at all. But weirdly, I can't pronounce /ʌ/ even when both my parents had the /ʌ/ /ʊ/ distinction. I'm sure that focused learning as a kid would give you a massive leg-up on these distinctions.
I've done accent training with Chinese kids of different ages, and it's relatively straightforward to get real improvement in producing and identifying new sounds. Of course, this doesn't translate to mature adult use unless they practice a lot, and get consistent exposure, but you can get real progress from a couple of hours a week.
But I take your point about "the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds". I wouldn't expect full distinction between all sounds, but I'd expect that, if I spent 2 hours a week for a year on pure, global phonology (say, when my kid is 6), he'd get a huge boost at producing and distinguishing certain sounds.
> I definitely disagree with the "language learning is not related to intelligence" claim (almost tautologically), but not sure what specific claim you're making.
There isn't a difference in fluency between idiots and geniuses who learn the same native language. People don't learn language by applying their intelligence. They use dedicated language-learning systems. Relatedly, given a population of people trying to learn a foreign language, intelligence will not predict which ones do so more successfully.
> (almost tautologically)
You're thinking about "learning" languages the wrong way. Is it tautological, or delusional, to suggest that learning to walk is driven by intelligence?
> After learning SE Asian languages, I can distinguish certain contrasts (like /d̪/ /ɖ/) that monolingual English speakers just can’t “hear”, even after instruction. A few years ago, did a pronunciation teaching course and was surprised to find that all Anglophones except me couldn’t identify the difference at all. But weirdly, I can't pronounce /ʌ/ even when both my parents had the /ʌ/ /ʊ/ distinction. I'm sure that focused learning as a kid would give you a massive leg-up on these distinctions.
It won't. As a baby, you already know the distinctions. As a kid, you will forget them.
For example, before she could speak, my sister "learned" that my name was k!, since I would make that sound to her. She once saw me approaching the house through the window and ran over to our mother excitedly saying k!k!k!k!k!k!.
But more recently she chastised me for not having taught her that very sound when she was young. She retained no memory of how to produce it and struggled quite a bit while I demonstrated it for her.
> I'd expect that, if I spent 2 hours a week for a year on pure, global phonology (say, when my kid is 6), he'd get a huge boost at producing and distinguishing certain sounds.
I wouldn't expect this. If you maintain the practice 𝘧𝘰𝘳𝘦𝘷𝘦𝘳, you'll get the effect you're hoping for.
Though if that's what you want to do, 6 is a late start. You can start that as early as you want, and there's not really an advantage to waiting.
> But I take your point about "the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds".
You still seem to think of sounds as something that you learn to tell apart. In that model, if you give a child the ability to tell two sounds apart, maybe he'll still have that ability later when you test him again.
But that isn't what happens. If you forget the difference between two sounds at age 3, or if you do it at age 8, you've forgotten it either way.
"There isn't a difference in fluency between idiots and geniuses who learn the same native language. People don't learn language by applying their intelligence. They use dedicated language-learning systems. Relatedly, given a population of people trying to learn a foreign language, intelligence will not predict which ones do so more successfully."
This seems just insanely untrue. Native speakers clearly differ enormously; some native speakers struggle to understand or produce certain sentence structures, have much smaller vocabularies, or have speech issues. I'm not sure how connected this is to "raw IQ/g" or something, but I can't imagine a use of the word "intelligence" that precludes differences in these different abilities. Foreign language learning is even more driven by intelligence; it relies on memory, grammar learning etc., which are all directly dependent on intelligence.
"You're thinking about "learning" languages the wrong way. Is it tautological, or delusional, to suggest that learning to walk is driven by intelligence?"
It's tautological. Walking, however, is relatively capped in terms of capability, so there's very little difference between someone with "just enough" intelligence to walk and "more than enough". Language is not like this. Linguistic abilities clearly run the whole gamut from zero to genius.
It's possible to use language in ways that reveal that you're stupid. You shouldn't confuse that with misusing the language. Stupid people are no more prone to doing so than other people.
> some native speakers struggle to understand or produce certain sentence structures, have much smaller vocabularies, or have speech issues.
(a) Some native speakers struggle to understand certain sentences because they can't understand the concepts. But that's not a flaw in their use of the language.
(b) Vocabulary size is very diagnostic of intelligence. It's not at all diagnostic of ability in the language. Vocabulary is driven by what you're exposed to, and people are systematically exposed to different things in ways that are driven by their intelligence, and also in ways that are driven by their relatives.
You might notice that people maintain the ability to learn new vocabulary without difficulty throughout their entire life, and it's not something that presents difficulty to people attempting to speak a foreign language.
(c) Really? There's nobody out there who thinks speech impediments are caused by stupidity.
> Foreign language learning is even more driven by intelligence; it relies on memory, grammar learning etc., which are all directly dependent on intelligence.
This is completely false. Foreign language learning is driven by practice, just like native language learning. Do it more and you'll get better at it. Do it more than someone much smarter than you who focuses on their textbook, and you'll be better than they are.
The phonetic alphabet is a bit problematic - there isn’t exactly a discrete set of phonetic sounds that make up the phonemes of the world languages. If you go through 5 places of articulation for consonants (labial, dental, alveolar, palatal, velar) and 5 modes of articulation (voiced nasal, voiced stop, unvoiced stop, voiced fricative, unvoiced fricative) and add about a dozen pure vowels and a few diphthongs, you’ve got most of the sounds for a bunch of languages. Then you start adding a few more locations like uvular, pharyngeal, glottal, and a few more manners like ejective and aspirated and you start getting better, but then you start getting into more detailed questions about whether Hungarian “ty” and “gy” are the same sort of palatal as those of other languages, and whether any American and British vowels are close enough to count as the same or whether the subtle differences are enough to represent them all differently.
There’s probably plenty of value in learning, from an early age, about all these features, and learning to pronounce and hear a good sampling of them beyond the ones that appear in the language(s) you speak, but you shouldn’t expect to learn “all” of them, because there isn’t a clear list of “all”.
With Retsam's clarification edit (that we're looking at short-term exposure based on reading from bilingual books), I think that Torches's suggestion is the most relevant benefit: A baby who hears sounds from different languages early on may find it easier to distinguish those sounds later. With two caveats:
(1) In the case Retsam describes, it doesn't sound like the parent is a particularly proficient speaker of the language they're reading. In that case, I wouldn't expect as much benefit, because the child would be exposed to English-accented Spanish, which sounds more like English than like Spanish. (Contrast this with Torches, who I imagine is practicing the phonetic alphabet!)
(2) Solid experimental evidence is thin on the ground. There's good evidence that long-term exposure as an infant can lead to benefits even years later (looking at heritage speakers of different languages), and that short-term exposure can lead to benefits a bit later (the longest I've seen is a couple months), but the combination of short-term exposure and long-term effects is both the theoretically least-plausible and the empirically least-tested.
Beyond this, I think we're mostly looking at benefits in language attitudes - the child may learn early on that it's normal for someone to speak Spanish. That's a valuable lesson, but there are better ways to learn that lesson (e.g. interacting with anyone who speaks Spanish - which would also be much better for learning sounds, come to think of it).
Are you looking specifically for cognitive effects (in which case my instinct would be 'a small effect consistent with any other activity that keeps your brain busy, like learning piano or chess'), or beneficial effects in general?
Having some knowledge of other languages gives you cultural capital, lowers the barrier to learning that and other languages further later on, can help with literacy in your first language, helps you understand other cultures and empathise with non-native speakers of your first language, and so on.
I just saw a recent news headline go by in my feed about how 2+languages leads to cognitive benefits. It seems like this is a common-sense side of the argument, though I know how often this turns out untrue or very marginal.
What if playing music was considered another language?
I listened to a podcast (can't remember which one, maybe Coleman Hughes?) with John McWhorter -- NYT columnist, Columbia linguistics professor -- and he basically said "I love languages, but no, there's no real cognitive advantage to learning other languages." From what I recall he even says that it's pretty pointless to require language classes in school, since <1% of people who are fluent in a second language learned it primarily through school language classes (usually it is family, living in the country, etc). Did not look up the primary refs myself but based on the broader cognitive ability literature I would put a lot of money on the "no" position, in large high-quality studies. Clearly if you go hunting, you will find some studies that say "yay language learning" because of the obvious selection effects at play.
But I keep my mouth shut when family members talk about how important it is that they get little Billy into the Spanish Immersion school.
So are you saying speaking more than one language takes up brainspace and makes you "dummer" for other things? My kids grew up fully bilingual from birth (no need for books etc, just one parent speaking one language and the other speaking the other), does that mean this has hampered their other potential intellectual pursuits later in life? I speak three languages fluently, does that mean I will always be bad at some other stuff 'cause my brainspace has been all used up?
"Do you think you could train yourself to recognize images on a sub-second timeframe? So you could look through 1000 images in 100 seconds (or less) and recognize if any of them were new?"
No, never. But - admittedly with no evidence whatsoever - I am loathe to believe that has anything to do with being a ployglot.
There is some slowness in young DLLs in picking up the preferred local language, but it disappears around age 6, and very soon after that they're statistically indistinguishable from monolinguals. It certainly has nothing to do with "brain space." You can't even distinguish the kids who were in a home with 1, 2, or 3 languages just by looking at their local language skills.
I think you are just needlessly aggressive here. To me the response makes perfect sense. You worry about brain wasting capacity on languages which could be missing somewhere else. Edward says that empirically, after age 6 we do not see any missing capacity.
Of course, the extra languages have to be stored in the brain somewhere, but it seems that either the amount of extra space is negligible compared to the total brain capacity, or that the missing capacity is something unimportant (maybe remembering fewer episodes of Paw Patrol?).
For those of you working in technology, what’s going on with the job market when it comes to more senior roles do engineers or engineering managers? I keep reading that people are getting ghosted left and right, big haircuts in terms of comp, you’re lucky if you get on the phone with anybody to begin with. Any reports from the ground?
“ getting ghosted left and right, big haircuts in terms of comp, you’re lucky if you get on the phone with anybody to begin with ”
This was my experience two years ago after I was laid off from my human factors/UX researcher job. I ended up switching careers (to skilled trades) and haven’t looked back, wish I made the switch way sooner instead of fruitlessly looking for another tech job for a year!
Skilled trades is the way to go! Last year swapped from managing data analyst to refinery operations. So much more fun (minus union nonsense but whatever).
Sorry, could you elaborate on what is meant by "skilled trades"? When I read that I think " plumber", but by your post and the one above it sounds like you're referring to something else. Thanks for the clarification.
I think what you’re thinking is correct! I mean any blue-collar work that requires some training to learn. Like plumbers, electricians, pipefitters, carpenters, etc. Refinery operations is just a small (well-paid, intellectually demanding) niche within blue-collar work, but it’s definitely blue-collar.
Unfortunately, everyone is free to become a software developer, but after I read how much mandatory training is required in my country to do qualified blue-collar work, I gave up.
Yes the training pipelines are dumb and have a large opportunity cost. It’s hard to get by for 4-5 years off apprenticeship wages (usually starts at 60% journeyman wages 1st year and goes up to 90% by the 4th/5th year).
That specific reason is why I went into refinery operations, because there is no apprenticeship and you start out making ~$100k with no overtime and over 10 years or so get up to ~$130k with no overtime (can easily clear $180k with overtime if you don’t mind working it).
If I understand correctly, FarmKind's shrimp matching donation campaign uses a pre-donated pool, and doesn't lead to any actual amplification of donations, unlike GiveDirectly's matching, where each donation actually leads to more money being donated than the donation, itself.
GiveDirectly states:
"This is a “true match”, so if we don’t reach our $600,000 goal, the leftover match funds will not be donated to GiveDirectly."
"In a conventional matching campaign, participants generally assume that by donating through the campaign, before it ends, they cause the matching funder (or in our case 'bonus supporter') to donate money that they wouldn't have otherwise. So a $10 donation can cause $15 to be donated in total. But critics of these kinds of campaigns fairly point out that if the matching funder has already committed the matching funds to the charity, then participants in the matching campaign aren't causing any more money to be donated to the charity than if they made a normal donation.
There are two senses in which our bonus system is not like this:
1. Participants get to direct the bonus funds to the charities they care most about. This is the most direct and clear benefit of getting your donation boosted through FarmKind...
2. Our bonus system isn't finite or time bound: It's an indefinite cycle of giving. In a conventional matching campaign, the funds are typically provided by a single large matching funder with a limited amount of funding, to be deployed by a certain date. FarmKind democratizes the provision of bonus funds: Many donors, big and small, support the bonus system on an ongoing basis. So long as enough donors are willing to provide bonus funds, and enough donors are willing to use them, the system can run continuously and indefinitely. As we explained earlier, this allows standard donors and bonus supporters to multiply their impact relative to what they could achieve without working together."
I don't get what they mean by the system not being time bound, given that this campaign has fixed dates. Maybe they mean that the whole matching is an accounting fiction, so they can set whatever dates they want, and it's all meaningless. If sufficient new donations aren't made in the campaign by the end of the campaign date, the matching funds will just rollover to the next campaign.
That is, it seems like the matching is just an arbitrary mapping of new donations to old donations. That is, let's say the matching pool were $500,000, then instead of saying that they'd match the first $1 million at 50%, they could say that they're matching the first $50,000 at 10x. They can call it whatever they want, since it doesn't correspond to any additional money actually being donated.
In fact, depending on the disbursement timing, wouldn't this potentially lead to *less* charity donated by a given point in time, since instead of disbursing the matching fund at the earliest opportunity, they'd hold onto it until the end of the campaign?
So it seems like the GiveDirectly and Shrimp matching campaigns are fundamentally different things, with the former actually leading to greater impact of new donations, and the latter only "amplifying" the effects of donations by convincing people to donate, who wouldn't otherwise donate, under the mistaken impression that their donations are now amplified.
The substantive issue is that per my comment, as I understand it, the whole matching mechanism of Farmkind is an accounting fiction, and its only justification is that the campaign will incentivize donors who wouldn't have donated, had they not been fooled by the campaign.
But the wording they employ is also disappointing. First, they seem to acknowledge that their matching campaign isn't doing what donors think it is:
"Our mechanism for multiplying impact is different to how people tend to think that donation matching works."
Yeah, the "people who tend to think" that the match means that each donation is amplified, are the people you're trying to fool into donating to the campaign!
Second, they have a weird line about how the most important benefit of the matching system is cycle of virtuous giving:
"Participants get to direct the bonus funds to the charities they care most about. This is the most direct and clear benefit of getting your donation boosted through FarmKind (although we think it's a far less important benefit than contributing to the virtuous cycle of giving)"
How exactly can a charity predicated on strict utilitarianism claim that the most important benefit of a donation campaign is the beauty of the system of donors working together, rather than actual dollars and cents being moved?
It sounds like they're trying to misdirect readers from the fact that new dollars and cents only come in if people are fooled by the nature of the campaign, and implying that the matching cycle is beautiful, in and of itself.
Hey Mallard, this specific campaign is a counterfactual match from an individual Anthropic employee. If less than $100k is raised, less money will be provided as matching. You're right that in other contexts, this page describes how matching works: https://www.farmkind.giving/support-bonus-fund
"a charity predicated on strict utilitarianism" -- not sure what this is based on?
Upon clicking the donation link for the campaign, the window that comes up states explicitly: "I agree to FarmKind’s terms of use and I understand that the bonus funds are already pre-paid by bonus fund supporters. Learn how our bonus system works here." with "here" linking to: https://www.farmkind.giving/support-bonus-fund.
So at the minimum, your website seems to contradict your claims about the nature of this campaign.
The point about utilitarianism was an aside, specifically delineated from the substantive points, so I won't respond to that, as it would only further distract from the substantive issues.
To your first point -- yes that's our standard T&C checkbox which we lacked the developer capacity to change for this campaign (we're a tiny org). Nonetheless, I can confirm that this match is contingent.
I've been writing a bit about being a writer who is poking around for various kinds of work, which has been just interesting to just enough people to keep doing it. Today I want to talk about long-shot jobs in the field, and what going for them looks like.
For the purposes of this post, I'm also going to be talking about Dwarkesh Patel a lot. Stay tuned to see why!
For definition's sake, let's define a long shot job as follows:
1. Good compensation, similar to what a position one step more junior than a given backend programmer would make. There's a lot of people who want to write, it depresses the payrate, and even someone like me (top percent performer among people seriously trying to do what I do) sometimes struggle to maker serious adult money.
2. Possibility of skirting normal job requirements. Most writing jobs that pay well pay well for reasons that don't necessarily have to do with value added; there's a whole socially-influenced backdrop to justifying the dollars that skill alone doesn't touch. If you are me (minimal education, getting by on moxie and ability) you jump at these opportunities any time they pop up, since most people are trying in "and goes to a school whose name I like" to the job requirements.
3. Fun work. Usually these jobs are unusual in some way, which makes them intellectually fun to contemplate doing.
Enter Dwarkesh, and to a lesser extent David Perell. Perell is the last guy who offered a "long shot" good job in my line-of-sight, and he's important for context here because his job offer was something like this:
*I want very good writers who are smart and very good at writing, the best outlier crazy-good guys in the word. I want to pay them an awful lot of money (I won't say how much) to do work (which I won't decribe). Also they need to have high-end degrees and you shouldn't even apply if you don't have one.*
The writers David gets are going to be the survivors who got through repeated rounds of selection before actually getting work, which is mathmatically going to get him worse writers, but I only actually care much about because I can't go for those jobs.
Now take Dwarkesh, who says this:
*I want people to do a weird kind of writing work for my podcast, but here's my best effort at saying what that work is and what we are looking for. I'm paying 100-220k-ish. Apply here. I want good writers."
I don't know Dwarkesh outside of pinging him for a corporate interview a long time ago (I think) that went nowhere (I think), but I like him. Look at him advertising for a writer and actually caring that they know what the job is, and then apparently also having gone "Oh, what do I want in a writer? Writing ability, I guess?"
That might seem a common enough reaction, but Dwarkesh might have been the only person to ever hire a writer who has ever had it. It's insanity; nobody actually says this. The fact that he's advertising mid-to-senior-dev pay on top of that is also crazy, but believe it or not actually takes a backseat to the other thing.
So you apply to this, obviously (seriously, go apply to this. I'm not getting the job anyway, probably, more on this later). But what are your chances of getting it? Let's say you have the minimum qualifications. What kind of competition are you facing? Well, a million people applied for that. Let's say half a percent are qualified writers who want crazy-ass writer money. That's still five thousand (count 'em) applications, and there's now a good chance he won't even see yours.
My guess is the way he'll get around this is by feeding the big document containing all the apps into an AI and having it do the heavy lifting of giving him a short list, based on either the same qualifications (which were mostly unstated) or another set of quals he's keeping secret. And, I hate to say it, AI's like on-paper qualification. Most of them are little credentialists, at heart.
NONE of this is a criticism of Dwar; I'd love him forever just for this job posting not mentioning degrees. He's a shining diamond just for that. But the first bit you have to grapple with is, statistically, you aren't getting this job, and it's the best one you will see in years.
The second thing to fight with is something like "Can I do this work?". Dwarkesh is hiring someone to do a portion of what he does, and his job is (can't get around it) bizarre. He interviews AI guys and does ad-rolls for a living. That's oversimplified, but to get this job you'd be very well advised to actually have a lot of existing art in the "talks about AI a lot" bin, and hopefully some non-conventional marketing chops that you can show on paper in a short air-table interview setting.
Even if you have those, though, you have to understand that someone who is hiring for this job and willing to pay a lot of money is usually talking about a job function that's important to them, and then ALSO usually talking about something they've done themselves up until now. In Dwar's case, he's also been incredibly-unlikely-outlier-bellcurve good at it, arguably better than anyone similar. There's no way he's doing it "normal".
So now you have a new burden: You need to get Dwar to notice you, you need to get him (or the AI, likely) to shortlist you, you need to be personable/oddly specifically qualified enough to beat the other few hundreds who got this far, and THEN you have to have a long, ethically-mandated conversation about what he's actually expecting and if any living human can actually do it in the way he wants.
And if all this goes well, congratulations. But realistically, it's not going to do that. You should have been a plumber. Everyone needs plumbers! Given that you've already made the mistake and aren't that, this is probably the best you can do.
[Note: Someone does have to get this job, eventually. If you are a writer type looking for this kind of work, I counter-productively urge you to go apply. It seems like a good gig.]
What do you make of The Conscious Wager? I argue we only have a small window of opportunity to build AI due to current existential risks and we should accelerate building it in order for even a slight chance to make consciousness eternal. Even if it increases current existential risks. Because the expected gains are infinite!
That article's good food for thought. Left a comment that boils down to "the wager's prompt is not descriptive enough of the near-term probability of extinction to be close to my personal assumptions regarding x-risk".
Thank you, I sense that is the problem with the wager as well. I need to make the case that our window is too narrow to delay AI as it is, and that eternal consciousness is worth the bet, even if it means humans must go extinct (since we will go extinct anyways).
I was arguing with an anti-vaxxer on Twitter and accidentally got him excited about a preprint from my labmates lol
Background: nuclear DNA damage and other stresses are responded to by p53 (the famous "guardian of the genome"), but DNA leaked out to the cytoplasm where it doesn't belong is responded to by cGAS-STING signaling, which causes an inflammatory reaction.[1] The system is interesting; bats have less of it, which may be part of why many viruses that kill us don't kill them (less inflammatory over-reaction; this in turn may be an adaptation to flight, which is metabolically intensive, so they don't overreact to their own self-generated DNA damage).[2] My labmates found that armadillos are missing cGAS-STING signaling entirely, suggesting that they're probably doing something else, which may turn out to have to do with their getting little cancer.[3] Although cGAS-STING handling DNA leaked to the cytoplasm is good, chronic activation could rewire things and be oncogenic.
The latest anti-vaxxer fear (a possible one, but unproven, but treated by them like it's proven) is that vaccines containing DNA plasmids (e.g. DNA vaccines or contaminated mRNA vaccines) go into the nucleus and integrate into the genome, or stay long-term in the cytoplasm chronically activating cGAS-STING (or both?).
So anyway I was arguing with an anti-vaxxer about something else (old other-vaccine-and-autism claims, not new plasmid-vaccine-and-cancer claims), and he dug around and found the armadillos paper, and tweeted it as a "gotcha" like "see your own lab agrees that chronically activating cGAS-STING would be bad!" (like that was the contentious part and also on topic). It then turned into an argument about whether upregulated genes that are downsteam of many different possible activators including cGAS-STING are proof of cGAS-STING activation (no) or whether the intended inflammation soon after a vax is proof of long-term inflammation (no).
I think he also thinks my labmates proved armadillos losing cGAS-STING made them not get cancer? (We'd never claim such a big proof without a lot of functional experiment evidence!)
But I'm glad he liked their paper, I told my labmates and they were amused too :)
[3] Schaffer et al. (2025). Loss of the cytosolic DNA-sensing genes CGAS and STING1 in armadillos (Cingulata). bioRxiv (preprint). https://doi.org/10.1101/2025.05.13.651073
I've enjoyed looking at Michael Burry transition from hedge fund manager to Substack grifter. He charges people for a newsletter to share his doomer perspective on the stock market. Being a fear mongerer is his stock-in-trade. I think this will do wonders to his personal wealth, and it might spur more hedge fund managers to pivot from the dying single-manager platform to selling their opinions to the public. Watch this space!
He supposedly has a net worth in the $200 million range, I don't think that writing a Substack newsletter is going to meaningfully add to the pile.
A better model for him is that he's writing a Substack for the same reason I'm writing comments; because he's a grumpy dude with a brainful of opinions that he really wants people to listen to.
Why would you think it to be more profitable to write a Substack than to work in the giant $100 trillion money machine that is the US financial system?
Well, when you sell opinions, you make money whether they’re right or wrong, whereas in finance you lose money when your opinions turn out to be wrong. At the very least the earnings from selling an unstick are much lower variance than from actual trading. People who are wrong at least don’t have to pay their audience to listen to them.
Either way, given his net worth, he didn't seem to have any trouble with his old job. I guess entertainment is a sector more resilient to stock market crashes, but still.
Tens of thousands of paying customers is a more stable way to make money than it is to short Palantir and Nvidia for nominal values of $1B, at least for now.
I'd like comments from programmers on this please! Are claims of AI making programmers more productive basically bullshit? What exactly is going on here?
At work, my productivity is bottlenecked by meetings and security theater and other nonsense. At free time, my productivity is bottlenecked by the lack of free time, and the lack of energy when that free time finally arrives.
AI made me more productive in the sense that I did some very small projects I wouldn't have enough time and energy to do otherwise:
Where is the shovelware? Easy, it's sitting on my phone in the Claude app. Making a simple mobile game in a few minutes over a beer is a great party trick!
And any time I have some half-formed thought of a game idea, I can test it, iterate on it a few times, and... rapidly figure out it's garbage, typically, but without a sunk cost of having spent a week coding leaving me sentimentally attached. Eventually I'll hit upon one that doesn't suck, maybe.
Claude is enabling me to do things I wouldn't otherwise be able to do at all. I'm a sysadmin by trade, in the pre-AI era the most work-related coding I would do is powershell scripts. Certainly there have been times when I've got to the end of a Claude Code session and realised I could have hand-written the script faster, but those are few and far between. But also, I can do so much more - for example, had a feature request for a SharePoint site for something that couldn't be done out of the box. That would mean hiring a SharePoint developer at great expense to code it up in SPFx... or just ask my good friend Claude. I had for several years considering skilling up as a SharePoint developer as demand is there, glad I didn't waste time on that!
So many solutions coded up across disparate platforms, languages, API's that I just do not have time to learn myself. Perhaps an experienced SharePoint developer could have coded up that SPFx app faster. And an experienced full stack Azure/React developer could have made [business iPad app] faster. And an experienced C#/Azure dev could have coded [business logic app] faster... if they already knew the Dropbox API inside-out. Maybe.
It's like having a team of junior developers at my beck and call, who know every language and architecture. Sometimes they make dumb mistakes, and sometimes have rocky handoffs to a different junior (ie when context fills up and things just aren't the same). Right now, I still need to know how to code to occasionally go in and help on tricky problems, but the writing is on the wall, 100% hands off vibe coding is coming.
And now we have Opus 4.5 who feels less like a junior. It can even make convincing 3D models out of primitives in Godot!
I am not in any sense a programmer - I write papers in LaTeX, and that’s about it. But with AI assistance, it’s now really easy for me to include diagrams in my papers that I wouldn’t have included in the past. And that article says “where are all the Tetris clones?”, but with a small homework assignment I got a dozen students with no coding experience (one of whom had never even heard of Tetris!) to code up Tetris clones (and the one who hadn’t heard of Tetris before told Claude to make hers run slowly and she now has it on her phone as a relaxation tool). I’ve also used it to get clones of some of the nytimes word games so I can send one to my partner after I’ve finished it, even though the actual site doesn’t let you restart the game once you’ve solved it.
Obviously none of this is anywhere near impacting the productivity of an actual coder, but it is evidence that we can get lots and lots of minor apps written if non-coders start learning to think about what problem a program could solve for them.
I'm recently retired so I haven't really used it in a production environment, but it's definitely made me vastly more productive on hobby projects. The colleagues I've kept in touch with agree and their opinion of stuff like the above is that only good engineers actually find it useful. Mediocre-to-bad people get themselves into trouble with it and just wind up with a pile of code that they don't understand. I think you need well-developed instincts for modular design in order to be productive with LLMs. You can't just point it at a codebase and say "implement feature x". You have to architect the solution yourself and just have AI solve small bits at a time.
This is how I approach using the AI as well. When I need some function that does one well defined thing that I could figure out in 15 minutes, instead I spend 1 minutes writing a good prompt and get my function. I have vibe coded a few webapps (I don't know the first thing about web development) and usually I end up with about 1000 lines that do mostly what I want that I don't understand and don't really want to spend time figuring out, good enough for a dopy little webapp but not how I would do something at work.
An EE I work with has learned that the AI is great at reading hardware datasheets for you though and quickly answering questions. I find that pretty slick.
Most people who are frustrated with coding agents use them *very* poorly. The correct way to use them is to farm tasks out to them and then go do other things -- often, spin up other coding agents to do other tasks / features. If you cannot get comfortable with that, coding agents just aren't for you. Which is totally fine, but it's a critical distinction.
Beyond that, you need to put in the work to get your coding agents to be better at software engineering tasks. They need to write tests. They need to read and write docs. They need to do research and create plans and get approvals (where you, the senior engineer, are in the loop). Coding agents are really best thought of as untrained interns, so you have to train them.
Soft plug: my company, Tilework, has spent a lot of hours thinking about this exact problem. We are pretty opinionated about how to use coding agents and have gotten some fantastic results across a few different companies right now. If you want to give it a try, check out https://github.com/tilework-tech/nori-profiles, and feel free to reach out at amol@tilework.tech
My feeling is yes the agents got better (there barely were any before 2025) but the main driver was the LLMs. For the agents that can switch LLMs, it's very obvious how much of a difference the LLM makes, in my view much more than the agent. Supporting infrastructure for sure got better, too (sandboxes, MCP, skills, etc).
This is kind of supported by the fact that the big coding benchmarks (SWE-bench etc) seem to use very simple harnesses (100 or so lines of code, I believe) and show huge differences between LLMs.
>This is kind of supported by the fact that the big coding benchmarks (SWE-bench etc) seem to use very simple harnesses (100 or so lines of code, I believe)
Agreed. While I'm speaking from ignorance, my gut reaction is that a plan-and-execute loop, _given a sufficiently reliable LLM_, should be very simple. Not much more complex than
while(goal not achieved)
{
query LLM for a plan to get from the current situation to achieving the goal
execute the first step in the plan
check to see if the step succeeded or ran into problems - add to LLM context
}
( with a lot of tool-handling complexity buried under "execute first step" )
But, generally, I'd expect almost everything to depend on how good the plan from the LLM is, which echoes your
What is a coding agent? My only experience using AI is using the prompt window that comes up when you go to https://chatgpt.com/ and type things into the window. Not clear to me how you "make agents".
So just to take a step back it's worth understanding what an "agent" is. Formally, an agent is 4 things:
1) a large language model. This can be any of the popular ones -- chat gpt, Claude, and Gemini are the big 3
2) a set of tools. This is usually a set of scripts that can be called with the right parameters. If you understand what an API is, you can think of tools as an API exposed to the agent system
3) a prompt. This is generally a few paragraphs of text that explain what the goal of the agent is and how it should behave. Normally this is split into a "system prompt" and a "user prompt", but you don't really need to worry about that distinction
4) a loop. The core infrastructure of the agent is a simple loop where the LLM output is fed back into itself as input. At each step, the LLM decides if it needs to call one of the tools it has access to. It will keep going in this loop until the model decides that it is done
From here, you can get a bunch of different kinds of agents. Most of the models you interact with on the web are agents. For example, if you ask Claude to do some research for you, it will go look up a website and then respond. That's two turns of the loop. Or you could have a special deep research agent, which looks up many things and consolidates those findings, and then uses what it finds to drive further research.
A coding agent is an agent that is specifically good at writing code. Most of its tools are related to reading and writing files, the underlying LLMs have been trained on a lot of code, and the prompts are generally code forward.
Hopefully that helps explain some of the distinctions!
The easiest way is to download a tool like Claude Code, which is out of the box a coding agent. But it lets you modify how it works in all sorts of ways, so that you could make it behave like some other agent. There are also a bunch of startups selling agent builder tools, but none that I would care to recommend. OpenAI also let's you make "custom GPTs" which are kinda the same thing
I could love to see a bunch of actual prompts, and results. Just as 2 examples. I asked some LLM about jj (jujutsu), the new "hot" git replacement. It failed 9 times in a row. It gave me commands that didn't work and I'd paste the error message "no such command `jj foo`" and it would apologize and then proceed to make up some other wrong answer. I get jj is new, so is the majority of software I write.
A non-code example, Google's Thanksgiving Day Doodle lead the an LLM prompt on how to plan a Thanksgiving dinner to be ready on time. The answer it gave looked nice, but if you read the details it had various inconsistencies. When questioned on them it just dug itself into worse holes.
Positive: I asked an LLM to make a single file "hello triangle" in OpenGL on Linux. That worked. I asked for the same for DirectX12 on WIndows. That took several iterations but it worked. I asked for the same in Vulkan. That failed. The LLM tried 12~15 steps and finally gave up and said "please search online for a working example"
I had another today where it told me googletest (an open source C++ testing framework that's been widely used for 10+ years) threw exceptions when skipping a test. It does not. It just hallucinated that.
In any case, maybe if I searched the net enough I could find someone who documented what they made and what each prompt was. I would certainly love to be 3x faster.
I think there is no solid consensus on this because there are several very different experiences people are having:
-pure vibe coding with no experience
-vibe coding with no direct experience in the specific thing, but as a generally competent SWE
-having the AI do simple tedious work
-having the AI do legitimately impressive but narrowly scoped work
-having the AI do impressive sprawling work
-experienced people using the fancy autocomplete mode in IDEs for cool-but-not-earthshaking speed gains
Those all have quite different failure and success cases. (To be clear, I'm saying that all of them can have both successes and failures, just all looking different and some succeeding more commonly than others).
Furthermore, there's prompting style. Some people are typing "make the webapp better i want the user to understand it better" and some people are typing multi-paragraph essays.
From my perspective, the biggest gains are available when you need to interface with a big complex system in a relatively straightforward way, and just correctly using it *at all* is a huge challenge. You can get a working example running on your own machine, semi-tailored to your particular circumstances, on absolutely anything. That's a really big deal.
The worst failures come from piling vibe on top of vibe. If you are not periodically stepping in with your human judgment for a little refactoring and cleanup, then your vibe coding is going to quickly collapse into misery, no matter how impressive it started.
I feel like the main way I use AI isn't listed there. I find it very useful when working with a library I'm not familiar with. Sure it might be possible to find everything I need in the documentation, but AI can easily tell me stuff I missed, and I can easily check if it tells me something that isn't true.
And another big thing is that I tend to get stuck and know several things I *could* do, but not be willing to pick one and move on. I'm not sure how good the AI is at knowing the best option, but even flipping a coin would probably be an improvement over what I had been doing.
I recently vibe coded a relatively featureful guitar fretboard learning app in a couple days of unfocused, occasional prompts. This isn't a case of 25% more productive. It's hard to measure, but would have to be well, well north of 1000%, because as a mostly backend person with a lingering, instinctual revulsion for javascript, I could not have created that app manually in any amount of time. I would have lost focus, gotten bogged down, and given up. It absolutely has enabled new applications. That said, this kind of greenfields, "make a self contained app" is a domain where current models really shine.
(Here's the app, if you're interested in seeing the results: https://fretu.de )
I've been tinkering with it and I think it's tough to say. The most useful thing it's done for me is the times it caused me to stop procrastinating, independent of how helpful it was. The biggest issue I have with it is that it often takes long enough for me to get distracted, at which point it might be a negative.
Cursor's tab-completion thing, which is basically just a smarter auto-complete, works pretty well, for the types of things that I might otherwise do a quick regex find-and-replace to do, but it's a bit error-prone at times.
This is a really good point. You'd think getting used to "maybe the AI can solve this for me" would be bad for motivation, but actually, that being an option has gotten me through a couple of really nasty debugging walls that I might have otherwise unhealthily avoided for quite a while. (With the AI outright solving it for me in some cases, and just getting me productively thinking enough to do it myself in others).
I don’t have any metrics on this, but I have high confidence that good use of AI makes me much more productive.
The AI autocomplete means that once I have sufficient code such that there is a style and pattern established, the completion is correct >80% of the time, and the remaining 20% it is simple to correct. If the area I am working in is too complex or novel to use the autocomplete, I disable it.
The “vibe coding” style is more hit or miss, and there is skill in recognizing from the outset if AI will be able to actually solve the problem. My general strategy is to define the overall structure, then review each point interactively, one at a time, until I sign off on it. Once all that detail is in the context window, I can often tell the AI to proceed with the work to great effect.
That being said, for entirely new features I find getting AI to do quality work is near impossible. So I don’t waste my time on that. Because AI constructs features in a one-shot manner, it has no preference for incrementalism and thus no preference for modularity. Anything that requires design work should not be left up to an AI- they might be better than a junior engineer at it, but far worse than senior let alone staff engineer.
So the productivity gains are all about playing to its strengths.
I've found AI is handy as a "micro coding" aid. By that I mean not microcoding, but in effect an enhanced web search whose result clearly summarises and explains a few lines of code to achieve a small part of a larger program.
This is especially useful for tricky or obscure APIs which lack up to date (or in some cases any) documentation. A recent example was packaging a binary file into an OCI (Open Container Initiative) document for storing in a Docker repo (well actually a Harbor repo). It turned out that instead of struggling with the intricate code for "hand rolling" the document layers etc, the simplest and most reliable way was to use a utility called oras, and AI searches gave examples of the command options, which I couldn't readily find with simple web searches.
I think for me I'm being like 25% more productive, and it has fundamentally altered my problem solving strategy. It's often faster to go to the AI than to start googling. That is not an empirical metric however, although, I don't know how empirical you can get, since there is no standard unit of work for software developers.
I have a friend trying to build a business around the idea that projects that previously took $50k and a month or two to complete, can now be done for $3k in a week, due to AI. He's pitching it as 3k 3days as I recall, something like that. He says the idea is working, which would completely invalidate that study.
I don't know if my friend bothered to read the post (I did), but I told him the gist and here's his take:
"Easy to test. Task two devs with making a simple AppScript script that looks for files in a google doc folder and sends them to slack. Neither has touched AppScript before, or the Slack API. One uses AI, the other doesn't. Obviously the dev who has AI write the script will finish much faster.
Now, if you tell Claude Code to fix a driver in the Linux kernel, probably the graybeard with 15 years experience is much more productive than Claude Code, and Claude Code probably won't help you much."
I agree with him, AI really shines when you're handling some tech you have never used before. Recently I finished a pretty complex task about integrating a PHP backend to read and write from Azure Blob Storage, no idea how I would've done this without AI, since there are no libraries to use (Microsoft used to have one, but they deprecated it). It would've been a real grind working out how to talk to that API by reading the docs.
I think I disagree with my friend about the Linux kernel driver case, the Linux kernel was in the training set for the LLMs, I bet Claude Code would do quite well on that task (depends how messy the Linux kernel is).
I think the strongest argument in the post is that we don't observe that the rate of software production has increased dramatically. But then, AI doesn't make development trivial, just easier. It's not gonna change the amount of devs that have the time and energy to start churning out apps and sites as a side-hustle. I personally find it difficult to code in my off time, even with AI, coding is hard and tires out my brain.
Purely a vibes based gutfeel, like I said, it's not rigorous. I have no fucking clue how I could measure my productivity. Hmm, we use fibonacci story pointing, maybe I could figure out how to look at that, but I don't know, I don't try to maximize my productivity, if I feel like I did enough for the day I just stop. So even if AI did make me more productive, I'm not sure it would show up there. And I don't use AI on every task.
I’ve always seen sports as a kind of mate-selection thing, with men and women competing and the winners signaling their genetic quality. Maybe people don’t consciously think of it that way, but in the end, they behave as if that’s what’s happening. So, allowing trans women just breaks that original purpose.
I loved sports, soccer, basketball... Except for in high school, w/ moms and cheerleaders, I don't remember many female observers. AFAICT we were all just playing for fun. If you want to say it's guy on guy competition then I couldn't disagree with that.
I think a simpler explanation would be that the main purpose of women's sports is to allow females to compete despite not being as good as males. Cf the paralympics, weight classes and age classes.
So the argument here is basically that people kind of look at physical fitness as a factor in mate selection. In other words, having men’s and women’s categories in sports fits neatly into this kind of preconceived mental category we all have. And when you break that by allowing trans women to compete in the women’s category, it kind of disrupts that symmetry that people have on a subconscious level. So of course, they might not really understand exactly why it feels off to them, but I think it shocks them because something they’ve internalized on a very deep level suddenly gets broken.
I don't understand the logic pipeline from "people select partners partially based on perceived physical fitness" to "trans women competing in the women's category shocks people... because they can't get pregnant?" Wouldn't this also predict people being uncomfortable with competitive categories for post-menopausal women, or any mention of athletes being unable or unwilling to reproduce? Wouldn't they also be uncomfortable with non-sperm-producing trans women in men's sports?
My intuition is that the "intuitive take" arises from a feeling of "I want to see tits on the telly but the trans women don't have them, which makes me uncomfortable."
Trans women often have nice boobs actually :) But yeah I always wondered if trans acceptance will go up as the transition process get better. But the pushback over trans women in sport was especially strong. This why I think there is something more going there.
OF COURSE acceptance goes up as the transition process gets better. The entire controversy is driven by transgenders who do not pass or do not pass easily.
There’s this kind of intuitive pipeline in people’s heads: they look at women’s sports partly as a way to kind of subconsciously assess who has strong genetic traits. And that doesn’t mean they’re literally sitting there thinking about mating, but it’s just an underlying instinct. And when a trans woman wins, people kind of feel like, “Hey, that doesn’t fit the pattern,” because they might think that someone who has, let’s say, a Y chromosome or a different starting point has an advantage that doesn’t match that intuitive idea of genetic fitness. So people feel like the person winning might actually not have a “better” genetic makeup than the person losing, and that just feels off to them. So that’s kind of the whole pipeline: people are looking for who’s got the best genes, and when that pattern breaks, it just feels really uncomfortable.
Huh, this is interesting. You're arguing that our perception of "fairness" in sports is tied to honest signals of genetic fitness. This explains the trans & intersex athlete controversies pretty well, and maps onto the weird ways we've drawn anti-doping regulations.
My impression is that successful male athletes are generally considered attractive whereas successful female athletes are not generally considered particularly attractive. Would you disagree?
I disagree! On a personal aesthetic level, naturally, but also on the facts.
There's a (now ancient) Oktrends post that pointed out that being polarizing is great for dating. People rated average or below, because half hate them and half love them - so even if ~70%-80% of men are intimidated by women above a certain fitness or muscularity, you can still do pretty well if the remainder loves it.
And indeed, I've had a few exes who were IFBB pros and / or influencers, and they get *constant* male attention, every time they step out the door (and of course, on any of the apps that's multiplied a hundredfold).
>I've had a few exes who were IFBB pros and / or influencers, and they get *constant* male attention,
That's just because they're fit and attractive, not because of the social status the achieved by virtue of winning competitions. Look at the WNBA: it's not like those women are dating movie stars.
Men need competition to signal status, women's status comes purely from their easy-to-observe physical attributes.
So would you expect pairings of successful female athletes with highly attractive males from other fields (e.g. music and modelling and acting) to be as common as the other way around? I'm trying to think of how this could be tested.
> So would you expect pairings of successful female athletes with highly attractive males from other fields (e.g. music and modelling and acting) to be as common as the other way around? I'm trying to think of how this could be tested.
Actually, as near as I can tell, it's the other way around - the hotter and fitter and more extraordinary a female athlete is, the more likely their husband is to be a misshapen potato man.
You can see this firsthand if you ever watch American Ninja Warrior - all the partnered top women (Flex Labreck, Meagan Martin, Sandy Zimmerman) all have basic potato man husbands, in the sense that the women are 10/10 on fitness and capability, and their husband is "just some schmuck" with essentially zero visible redeeming physical aspects.
Women don't optimize on looks to nearly the degree that men do, they optimize on status, and how a given man makes them feel.
And while I'm not quite a potato man, I'm generally a decade or two older than my paramours and I'm certainly not visibly turning female heads as I walk down the street, yet I end up with them too, so...one more point on that side of the ledger, I guess.
I don’t think that’s contingent on placing well in any given competition the way this implies. Excellence in most any field is attractive, not just sports, and athletes tend to have uh athletic builds which I’m sure helps. The competition itself being a mate-selection thing doesn’t really scan with my view of elite athlete psychology at all.
Do last-placers also have a lot of sex in the Olympic village? Is there a way to know?
> Do last-placers also have a lot of sex in the Olympic village? Is there a way to know?
Yeah, I think your view is probably directionally correct, that excellence is itself the attractive factor rather than literally winning.
Forget last-placers - even an Olympic Qualifier who doesn't actually end up competing is a literal paragon of human excellence and capability, very likely to be the fittest and / or strongest person any regular person has personally ever seen, and have no difficulties attracting attention from the opposite sex if seen in their domain.
I never Qualified (although I count a few qualifiers in my social circle), and was only ever regionally competitive at my sport. Even today, more than a decade later, I still turn heads and get attention every time I go to the gym, doing ~60% of what I ever did while competitive, because that 60% is so far above what everyone else is doing.
In my opinion, one of the most serious problem for human beings today, especially in first world countries, is sleep problems. Sleep deprivation, insomnia, and nightmares. All other basic physical needs are obeyed. We can have comfortable temperature (or adequate clothing to protect us from weather conditions), and most people have enough food, including all the necessary ingredients (proteins etc.) But most of us don't have enough sleep.
Absolutely. I think a lot of people would make radical changes to their career/lifestyle if they realized what an impact it can have. Most fitness regimes will be a sideways move if sleep is sub standard
> In my opinion, one of the most serious problem for human beings today, especially in first world countries, is sleep problems.
Isn't this just the direct corollary of "80% of americans are overweight with 50% obese," because food is so solved that approximately everyone is fat?
The reason everyone is fat now is that food companies have had literal Phd's working on junk and fast food for decades, making them as delicious and "moreish" as possible - superstimuli, in other words.
Similarly, every company with a major app has thousands of Phd's on the other end of that app, modeling their customers and optimizing the content to attract as many eyeball-hours as possible. Remember Neflix CEO Reed Hastings saying "our competition is sleep?" This is why, just like 80% of people are fat, people spend 10-11 hours a day on screens now:
This is why everyone doesn't get enough sleep - they're so addicted to their Skinner boxes with attention superstimuli on the other end that they sacrifice sleep quantity and quality for one more hit.
The solution is the same one as the food one - don't put yourself in fully adversarial situations with thousands of Phd's optimizing against you. Don't eat fast and junk food, and don't look at screens for 10 - 11 hours a day.
I had not enough sleep even when I was a child. There was no internet, and we had no TV. So no, it is not this. At least not only this.
I think that people were deprived of sleep even before the internet epoch.
Also, if you are adult (or even a high schooler), you cannot just stop looking at screens. Today you need internet for too many things, some are impossible to obtain without one. If you want to stop looking at the screen, you need a time machine, or you need somebody else to take care about your issues, or you need to die.
I've recently found that when I've been reading books to my kid for long enough, my mind can start wandering. (Let's ignore for now whether that might be a bad thing). I'll suddenly realize I've been on flawless autopilot for several pages. This is surprising because for me, "mind wandering" means internal monologue - basically the same kind of cognitive process as reading. And sure enough, when I deliberately try to think my own thoughts while continuing to read, the thoughts barely get a half-second foothold before being washed away. But when it just happens, I can have long, coherent chains of thought, including planning things for the future. Probably relevant that I've read these books many times, but also I definitely don't have them memorized - this is happening even with non-rhyming stories with several dozen words per page.
I'm not sure there is much productive insight to be had here, but I just thought it was a really fun little cognitive quirk, the sort of thing that ACX readers might find interesting. I'm also interested in hearing about any similar experiences.
(Also, I wonder if the "second screen experience" lifestyle is going to make the younger generations really good at this kind of thing?)
I had this funny experience in school, at the age of 14 or so. I was in history class, which I found hopelessly boring*. At some point we're going through the next part in the book, and the teacher says "hey Anon, go ahead and read out loud this couple of paragraphs."
So I go on reading for 2-3 minutes, and my mind gets SO distracted, and I'm focusing on the tone of my voice, the particular text font, how to pause for punctuation, all while the words flow automatically out of my mouth, and I am so glad the teacher's asked me to do that so he'll probably be okay with my "engagement in class" for the day and stop bugging me with stupid questions, so I finish the page and look at him with a smile and a nod, and he goes with a strict face:
"Good. Now, in your own words, how would you summarize this excerpt?"
And I'm like shitshitshit how fast can I speed-read this and mumbling while realizing that "sorry, I wasn't paying attention" won't sound very good as regards my attitude in class, and it felt like ages until he got so mad at me and asked someone else.
It felt so weird that my classmates were all convinced I was deliberately making fun of him and being a jerk because, dunno, I only liked maths and wouldn't just answer a bloody question in history class out of spite.
When I was a child, my parents would read Lord of the Ring to me while I was learning vocabulary for French. I knew the book so well by heart that I would call them out for any reading mistake that they would make, but afterwards I would also know the new French words.
Probably it wasn't the most efficient strategy, but I was very good at school and it was good enough for me.
This also happens with making any repetitive announcement; you say the words as reflex, without thinking about them. It can be a problem when the current announcement has a slight variation from the standard.
This happens to me when I'm reading a book for myself, so I can imagine it'll get worse when I read to a child. Usually I realize I've been thinking about other things after a couple of paragraphs on autopilot, and either I pick it up from the top or I decide the mind wandering was a signal that I should stop reading.
When I feel like making EA-minded charitable donations, I usually give $500 to the Malaria Consortium. Should this 50% match in Item #2 make me donate to Rwandans instead?
I will definitely not be donating to the shrimp, though. The only thing that I think about when I see pro-shrimp substackers is "Mmm, I'd love a shrimp po-boy right now."
"One common defense of letting trans people play as their chosen gender in sporting leagues is that - although trans women may have a biological advantage over cis women, and this might make them win so often that it stops being fun for cis women, most sporting victories involve biological advantages"
This really isn't an issue I'm following closely, but isn't the whole premise of women's sport that they don't allow the biological sex advantage specifically? This just sounds like you don't accept women's sport as a category in principle.
Which makes sense I guess, it does seem strange to have a sports contest especially for a demographic that's worse at sports. But I don't think sports fans are going to accept "sports are just generally irrational" as an argument.
The whole theory discussion seems to ignore what is, in numbers of participants, easily the largest part of organized sports: amateur/recreational leagues.
The number of adults who at least sometimes play in organized sports leagues has been booming in the US (as everyone who's been doing it for a decade or more knows firsthand). Estimates vary but this category now includes somewhere between 15 and 19 percent of US adults.
That entire huge ecosystem of leagues -- which includes both team and individual sports -- is structured as divisions or leagues that are restricted across a range of different ability levels. It would obviously be ridiculous and unenjoyable if a 25 year old who played high school varsity tennis was in the same brackets with a 65 year old who'd dug out an old racket after decades without touching it. So that's never been how it worked and nobody thinks it should.
Hence for the millions of Americans who pay to participate in organized amateur sport, the idea of a 6'3" former male who played mens college ball joining a local women's basketball (or whatever sport) league has always been prima facie ridiculous. The apostate progressives on this topic who I personally know, which is several, are all either longtime players in such leagues or married to one. Multiply that practical-real-world-experience reaction by millions of voters and this is why the specific question punches so hard now at the Democrats.
> This just sounds like you don't accept women's sport as a category in principle.
Scott is arguing *against* this argument in the post.
> it does seem strange to have a sports contest especially for a demographic that's worse at sports.
Parasports are a whole thing. The main difference is that there are way more women than there are people with any specific disability, and that it would be considered offensive to imply femaleness is a disability, even though in the context of athletics it essentially is.
(You could argue weight classes in combat sports also count as this, though I think it's not quite the same).
>Scott is arguing *against* this argument in the post.
Scott is arguing that since most sporting victories involve some sort of physical difference, it should be okay to allow trans people in women's leagues, since we shouldn't really care about physical differences. This argument implies that we shouldn't have women's leagues at all, since the whole reason that there is a separate women's league is physical differences.
Scott is arguing "against" this argument in the sense that he doesn't accept the implication, but his argument still *has* that implication.
And that implication is also the answer: If you want there to be women's leagues at all, you've inherently decided that physical differences do matter, so you can't just say the opposite.
It sucks so hard, it's quite the contrast with the predictions of imminent superintelligence. Seedier corners of the internet are jibing that Claude Opus 5 will require the entire energy output of the sun, which will enable it to progress to getting stuck in the Safari Zone (this one is seemingly stuck in the Team Rocket hideout, which would be around the midpoint of the game).
I feel like we weren't groping forward so blindly in previous scientific breakthroughs. For the nuclear bomb, for example, Szilard worked out on paper that a nuclear chain reaction is possible (then it took the Manhattan project to actually build a nuclear bomb), but here, we don't know that superintelligence is possible, or that LLMs can get there. Will LLMs get there? To AGI at least? Will they peter out? Will we need a new paradigm instead, and if so, how long will it take to discover that paradigm? So many question marks.
I needed a guide to get through that bastard Team Rocket hideout as a kid, but eventually attained general intelligence (well, it could be debated) with a few more years of brain growth, so I'll let Claude off the hook for now :)
If you've ever tried making a model that can play pokemon, you'll see there's a tradeoff between more upfront information and less generality. You can tell it what a pokemon center is or what the end state looks like, but then it will not learn that for itself which means it will not be able to play other games you dont already know how to play. The only true general model is abstract and works on abstract things like direct memory access or pixel-level analysis, it must not be poisoned/biased by human knowledge unless it has discovered it itself. I don't think you will ever get LLMs to play arbitrary games unless it already has the knowledge for them, or the architecture has a way to use human knowledge abstractly and independent of the actual medium.
Far from knowing whether superintelligence is possible, I think we don’t even have a good theory of what it is. There’s a lot of rough ideas people throw around, but no model of intelligence is anything like a model of fluid dynamics or a model of gravity.
LLMs have been a huge improvement over the previous paradigms of intelligence that people were working with for most of the past 80 years, but they’re clearly missing lots of things, and they’re an unusual model that can’t really be understood, not even at the level of a differential equation.
Yes Claude has terrible vision, because Anthropic isn't trying very hard to make it better. Gemini has better vision and does better on these sorts of tasks.
Anthropic is basically all in on making the best possible machine learning and coding agent so they can achieve self improving AI. To the exclusion of everything else (math, vision, video models, image models, world models etc.). Time will tell who was right.
> I feel like we weren't groping forward so blindly in previous scientific breakthroughs. For the nuclear bomb, for example
IMO that's more a feeling from hindsight than it probably felt at the time.
Like the Manhattan project simultaneously pursued three different enrichment technologies because it wasn't sure which (if any) would pan out, and I believe part of why we had three bombs at the time of Hiroshima and Nagasaki (rather than immediately use the first bomb we made) was because there was a real risk that one or more of them wouldn't work. Also famously that whole "will we light the entire atmosphere on fire" concern, (though I'm not sure how realistic a fear that ever was).
And I think the history of technology is filled with "we need a new paradigm but don't know what that will be, or if we'll ever get there" - aviation is a good example of something that was theorized for millennia with no clear path to how to get there. (An infamous New York Times editorial published in 1903 claimed it would take "one million to ten million years" and even Wilbur Wright at one point predicted it would take 50 years)
>Like the Manhattan project simultaneously pursued three different enrichment technologies because it wasn't sure which (if any) would pan out, and I believe part of why we had three bombs at the time of Hiroshima and Nagasaki (rather than immediately use the first bomb we made) was because there was a real risk that one or more of them wouldn't work.
I agree in part and disagree in part. The Trinity test was necessary because they indeed weren't sure that a Pu-239 implosion bomb would work. After that test, the Fat Man Nagasaki bomb, which was the same type of Pu-239 implosion design, was a pretty sure thing. The Hiroshima, gun-type, U-235 bomb was such a sure thing that it was used without a test. Come to think of it, they must have been remarkably sure of even the _yield_ of that bomb, to be reasonably sure the bomber crew would survive.
Earlier in the project, Fermi's first reactor relied on knowledge of the delayed neutrons to know that the reactor would be controllable, even with just natural uranium. If all fission neutrons were prompt, the doubling time constant for the reactor would have been far too fast for human reaction times to control.
For LLMs - well, we understand the low level processes - the individual perceptrons/attention layers well enough. And we have the scaling laws for _part_ of the high level behavior. But a _lot_ is still not understood, as far as I can tell. AFAIK, we don't know the answer to, e.g. "What do we tweak to cut hallucinations by a factor of 2 without losing any accuracy on GPQA Diamond?"
I think AGI is almost the reverse situation of the Manhattan project. There we knew the physics - i.e. do this and you'll get a superbomb - but no one believed you could separate enough of the right isotope of Uranium. Building the bomb was about finding a method, not about whether the underlying concept was sound. But with AGI we have a method, but no one knows for sure if it will result in AGI, or if such a thing is even possible with the current approach. We have a method, but we don't know that the concept is sound.
DeBoer launched some online rage against Gusev's post but aimed it at Erik Hoel who he'd apparently mixed up with Gusev, and then when Hoel politely pointed out the error and asked for a correction, DeBoer's response was to Hoel was,
"Of course, all of the questions in the piece apply to you just as well, and you will studiously refuse to answer them because it's easier to impute racism onto anyone who disagrees with you."
Thought Hoel's reply to that was about perfect:
"All you needed to do to understand my beliefs [about heritability], Freddie, was (a) not get confused and think I wrote something I didn’t, and (b) read the grand total of 2 posts I’ve [ever] written on the subject.
Like I know that these are good internet tactics, to attack and downplay and deflect and move on to more arguments as if nothing matters but…can you be a human being for a second? And not…whatever you’ve become over the years of doing this for so long?"
I've read deBoer regularly some time ago, and he has always had an unhinged edge to him that would regularly come out when he gets agitated. Afaik there also was a full-blown psychotic episode at one point in the past. The ratio between insightful and completely crazy in his recent posting has been getting worse for a while again, imo.
I do think there could logically be a 6 ft and below basketball league. There are weight classes in boxing and wrestling and age classes in youth sports for a reason. To make competition between competitors more fair. Nba and NFL are the leagues of freaks!
As I child I was really into two sports which heavily favour height: basketball and fencing. I am not tall and hit a wall with both. If there had been height classes rather than weight classes I might have continued.
That 3x3 traveling basketball tournament has an under 6 ft category. Plenty of fun to be had as a short basketball player. Getting paid million dollars per year is less likely
Well, the supply of sports leagues already outstrips the demand. There's any number of sports that people could be watching and aren't, because the people who are interested in sports already have their interest saturated.
That said, "under 6 foot basketball" would be an interesting game which would have a different style of play to height-unrestricted baseball. It would be more akin to something like lighter-weight boxing divisions, where the weight restriction leads to a different style of play, than something like under-15s football where it's just the same thing played worse.
I am somewhat disappointed by Claude Code. I wanted to test its vibe code abilities and I gave him some (locally downloaded, html only) wikipedia pages on various iterations of the same sporting event.
I asked him to write me a python script to scrape me the "List of qualified teams" table and the pre-tournament ranking into a CSV. It struggled quite a bit. I stopped it and trimmed the pages down to the table section and it managed to complete it, even handling some formatting differences across years.
So it was a significant speed up, since I didn't have to manually put the data into an excel-like software but still not a super speed up since I still had to edit HTML.
This decreased my timelines a bit but I am not sure. For once, it was Claude Sonnet 4.5. Also maybe I just suck at prompting? Well, at least I got an internal benchmark out of it. Do you have any similar experiences?
I use Claude Code with Sonnet 4.5 in my day job and I quite like it. Off of the back of a business idea that a friend of mine came up with, I'm planning to transition to being an independent consultant that can radically cut costs due to heavy AI usage. Like a vibecoder, but I actually know how to code, which I'm pretty sure means I can use AI for coding much more effectively than any vibecoder.
He was a game changer in my day job. I can use it to solve some simple, mostly pattern matching PRs (e.g write a migration to change a column or create a table) even on languages I don't fully know
Hmm, what have you been trying specifically? I'm kinda thinking listing oneself on freelancer or even fiverr could work for this (I have 0 experience on trying to find work on either platform, but it's what I've heard).
The main thing I’ve gotten from listing on platforms is… other users from outside the US who want to use me as a beard (i.e. they pay me a percentage of earnings in return for me bing the front man). I could do a better job of it, I suppose, but I’ve gotten more than a dozen requests from people wanting me to do that.
Wage arbitrage. As an American I can (theoretically) charge a lot higher rates than they can from overseas. With them doing the work, in theory I could take on more projects too.
Ah! Sorry for this. He had to write a python script that ingested HTML and generated a CSV file. It also had to be sufficiently general to work on ~4 iterations of the event (although they were similarly formatted)
I think the main thing I'd suggest (possible you did this) is making sure it can iterate, run its solution, see if it gets (what looks like) correct output. This for me is the biggest thing with these tools - if you just take it's first attempt its a crapshoot but if you give it a feedback loop it tends to do a lot better at getting to a good solution.
Telling it the actual answer of course defeats the purpose, but if it knows the result should be a list of teams and a numerical ranking, then it should be able to sanity check its own results and iterate that way.
---
Also I'm not sure but, it also might do better if it had actual browser access rather than looking at the raw HTML? I've found for a lot of things the AI does better if it can open a real browser, e.g. via playwright (incidentally playwright is probably what I'd be using for the actual code, too) - it might be able to do queries on the content, rather than reading the entire raw HTML.
To be honest it was still iterating when I stopped and decided to give him the trimmed HTML tables. He was on about the third version of the code. I got scared of wasting all my usage as I might have needed to use today on work
I think you might be right about playwright as well. Originally I just gave him the link, but wikipedia didn't allow the requests (got 403s) so i decided to play nice and download the html only page instead. This might have primed Claude in the wrong direction, and he tried to solve it with only beautifulsoup
l think it was a free trial kind of deal, now when I run "/model" it shows me Sonnet, Haiku and " Want Opus 4.5? Run /upgrade to get the Max plan or /extra-usage to pay per use"
The recent trans sports post is based on a false premise that cis women would "have no hope of winning." Sports leagues already had rules for hormones, and cis women win much of the time.
E.g. the anti-trans activist swimmer Riley Gaines (cis woman) didn't lose her championship to Lia Thomas (trans woman), like anti-trans activists want you to think. Riley and Lia tied for fifth place behind four other cis women.
Do you believe that women's only sports leagues should be allowed? If yes, why should we bend the rules so that transgender males identifying as women are allowed to compete in them?
(As I mention in the longer comment, sometimes they're outright fabrications. The ADF president had her young daughter testify—without disclosing that she was the ADF president's daughter—that her softball team's heart broke when they realized they were playing against a trans girl full of male biological advantages. Trouble is there was no trans girl: it was a cis girl with short hair. And, of course, there's the widespread and deliberately encouraged false impression that Lia stole Riley's swimming championship.)
If you think the arguments don't make sense, I question why you even think it's relevant that the organisations making them are even affiliated with the ADF. It just seems like a completely irrelevant point.
You don't care that sophisticated propaganda groups are working overtime to mislead and manipulate you and your society, with evil intent? You want to avoid finding out when that's where a particular piece of disinfo came from? You want to avoid practice recognizing their tactics and falsehoods in future?
I'm sorry – were you talking about trans activists, or the people opposed to them?
From where I'm standing, the people being economical with the truth when it comes to sporting events are nowhere near as grave a threat to our society as the people campaigning to house convicted male rapists in women's prisons or carry out mastectomies on healthy teenagers below the age of majority.
I've spoken to a trans man that got a mastectomy as a teen. He competed on his high school's boys' swim team, and his teammates were under the impression he had gynecomastia, which is fairly common. He decided to get surgery early, same procedure he would've gotten if he were actually a cis boy with gynecomastia and for the same reason too (gynecomastia is not typically medically dangerous, and treated for self-image/discomfort reasons). For some reason, this is way more controversial.
There seem to be two opposing forces: first, AMABs are X amount better at sports than AFABs; second, trans people are extremely rare, such that in the set of (all cis women + trans men), it's possible or even likely that the top athlete is a cis woman, even if the cis male average is much higher. I think this means that discussion of a certain level of trans victory in sports isn't interesting without knowing what percent of the athletes are trans, why that is (ie whether existing norms are keeping other trans people out), and how that's likely to change over time (as norms and the % trans changes).
I'm not sure how I feel about a world where AMABs are some large amount better than AFABs, but AFABs still usually win because trans athletes are very rare. I guess it matters less? Still seems like however unfair it would be if there were a very large advantage, it's some fraction of that unfair, as opposed to zero.
One thing I'd add to this is that the top athlete isn't picked randomly from all eligible. athletes. Some people are more athletic than others, you know it ahead of time.
Most *specific individuals* who are trans women athletes pose no threat to be the top athlete.
One of them could ask - "forget the abstract, why should *I*, specific trans girl HS senior with 0 athleticism, not be allowed to ride the bench on the girls' JV team rather than the boys' JV team? I'm not trying to cheat, I just want to be on a team with other girls, have camraderie and shit. I'm recognized as a girl in all other contexts, but now you're gonna single me out based on some hypothetical when actual me can hardly avoid bouncing the ball off my foot".
29 trans women out of how many women total who have won titles in the reference class surveyed?
Skimming the list, it looks like the reference class includes lots of state and regional competitions, national competitions from multiple countries (including one in Belgium), several narrowly age-bracketed competitions, several competitive levels (high school competitions, "Division II" NCAA tournaments, etc), a very wide range of sports (track, golf, roller derby, billiards, BMX biking, etc), several events per sport in many cases (e.g. 500 meter dash within the umbrella "track" category), and at least 17 years of data (examples range from 2008 to 2025) although most of them events are from the past 10 years.
A quick fermi estimate: 70 scopes of competition (50 states + 10-ish regions + national competitions in 10-ish countries) * 10 relevant combos of age bracket and skill bracket * 20 sport/activity categories at least as relevant as billiards and BMX * 5 events per sport * 10 years of data = 700,000 potential titles.
At least a couple of the trans women on that list won multiple titles, so the numerator should probably be more like 40 or 50 instead of 29 to match the units of the denominator. That works out to about 0.007% of women's titles being won by trans women. Which is somewhat less than I'd expect if the standards for trans women participating in women's sports were perfectly calibrated, since about 0.8% of the general population is trans.
I'm actually surprised by how low my estimated percentage of "titles won by trans women" is. I had been expecting around 1-2%, based on a prior that the "Testosterone levels must be below X ng/dL for Y period of time before the event" standards that most events seem to use would screen out most of the advantage from having had male hormone levels in the past but not all of it, depending on how well-calibrated the values were for X and Y.
One important point here is that professional sport competitions don't need to be regulated in the same way as casual sports. Many pro-trans-sports activists don't care about the Olympics so much as they care about a trans teen being allowed to play soccer with her friends at school.
Keep in mind that in many places, high school sports aren't "causal" in any sense of the word. People take them extremely seriously, and parents spend thousands on professional training. I can see many of them raising hell about a trans girl competing, especially if she does well.
Really? I would have thought the reverse. In the pro-sports context, there's financial rewards incentivizing cheating, as well as a business need for at least the appearance of fair competition. Hence all the rules about doping, what weight you can and can't be in certain classes, how often you can practice and what you can and can't do to scout competitors, and so on.
In a casual school-sports context there isn't really anything at stake beyond fun, camaraderie, and the life-lessons learned from competition. Those don't just disappear if the middle-school swim team has one trans kid on it, and without the perverse incentives that accompany financial benefits to the winners, your population of trans soccer players is going to be confined to the fraction of the 1% of trans kids that happen to be interested in soccer, so it seems like it would be hard for that to impact at a level that would break the system.
I think you misread my comment. I'm saying that people in *favor* of trans girls/women in sports often care more about the casual sports than the professional ones, because those impact more trans people. Yet people trying to ban trans women often focus their rhetoric on the Olympics, while targeting kids in casual k12 sports.
Haha at first I misread you as saying the opposite of what you said. But anyway...
In theory, in another historical context, I could sort of agree. Yes, I care more about kids's basic right to play sports with their friends than about the Olympics, even if I care about that too, and in a golden age of LGBT rights in general the latter might be less urgent.
But in the current context, neither is about sports; both are just precedent-setting lawfare attempts by Christian nationalist groups like ADF, designed to get us used to excluding trans people from society (itself part of a divide-and-conquer strategy to eventually exclude all LGBT people from society), so we can't give these fascists an inch. Even if the war on trans kids is even more stomach churning than the war on trans adults, both are part of the same overall strategy.
One of the main points of Scott Alexander's advocacy is for us to look at the substance of people's arguments instead of attacking each other on partisan grounds.
I agree with this mission and I am disappointed to see you fighting against it.
Unclear what part of this you see as against that mission.
I've made clear that I don't want trans women kicked out of women's sports, on the merits, including at a professional level. In a utopia of LGBT rights I'd still think that, but prioritize it less. In the real world, where *in fact* Christian nationalists are deliberately weaponizing sports—this isn't an ad hominem, this is recognition of their actually-existing political strategy—it is a priority.
(slightly) strained analogy time. A paramilitary group's code phrase, signaling to fellow travellers that they're gathering recruits and getting ready to blow up Metropolis, is "The cock crows at midnight."
We start hearing "The cock crows at midnight" said more than usual. I give everyone a heads up that that's a paramilitary group's code phrase.
I get tsk tsked for my ad hominem, instead I should have focused exclusively on whether or not it's true that roosters crow at midnight
It's also a Toxoplasma of Rage issue, and one which tends to be framed in a way that implies a binary choice between a "No trans women ever in any women's sports" and "Any AMAB willing to check the 'F' box on the entry form can enter any women's competition at any level of play".
The actual status quo of many/most high-level competitions having rules of the form "You must have bloodwork showing testosterone below X ng/dL for the past Y period of time" tends to get ignored, partly because a lot of people don't know about it, and partly because arguing that X or Y should be a higher or lower value is boring, technical, and doesn't work well as a proxy fight over the legitimacy of transgendered people.
2. Reducing one's testosterone levels has minimal impact, if any at all, on one's bone density or lung capacity, which are effectively baked in during puberty.
1. "Often" is doing the heavy lifting here. If the honor system isn't working, they can change the procedures to require documentation, like many top-level professional men's sports do for steroids.
2. I am very skeptical of both the accuracy and the relevance of this. Bone density definitely changes after puberty in response to hormones, which is why osteoporosis is a problem for postmenopausal women. Lung capacity (in terms of the raw volume of the rib cage) may be baked in, but it's one of several things that affect vo2max. Things I know off the top of my head that are hormone dependent are training capacity and hemoglobin levels.
This study (tiny sample size, but it's the only one I could find) actually found trans women athletes had lower lung capacity (as measured both by vo2max and exhalation volume) than cis women.
I don't agree with this strategy of approaching anything. It's bad behavior on principle, and is an ineffective way of getting what you want in practice.
Yes, there are probably people within the ADF who are consciously acting on the strategy you describe of trying to normalize excluding LGBT people. But most people who oppose trans women in women's sports don't endorse this. We should try to judge issues on their own merits, not on the most repulsive views of their most repulsive proponents.
Edit: from your other comments it sounds like you're in favor of some restrictions on trans women athletes, so I'm confused by what you define as "can't give these fascists an inch."
1. It's not "people within ADF," it's ADF, as an organization. This isn't a church or social club, it's a strategic litigation group whose reason for being is to strategically litigate and turn the US into a Christian Nation.
(Strategic litigation is a tool that can be used for good or for bad. NAACP used it to end racist school segregation; Lambda Legal and ACLU used it to win gay marriage; it was used both to gain and then later to overturn abortion rights. ADF also defend discrimination against gay marriage etc.)
2. The organized anti-trans-sports activist movement isn't a grassroots movement led by well-meaning normies who are fine with trans people but don't really get it. The organized anti-trans-sports activist movement *is*, in its essence, a lawfare movement. It *is* Charlie Kirk and Riley Gaines lying to normies that Lia Thomas stole Riley Gaines' championship, to trick the normies into getting used to eroding trans rights. It's not a radical fringe, it's the radical vanguard party leading the whole operation.
The most important thing to do in this context is make sure people know what's actually going on: that the leaders aren't good faith, they don't care about fairness, this isn't about sports, this *is* just a new attack on trans people after the bathroom-bans attack failed.
Men, and post-puberty boys, are not allowed to play in sports leagues set aside for women. I have never heard any objections to this. It allows women to play high school, college, and professional sports. Do you take issue with making these opportunities available to girls and women? If not, then it follows that someone who has the biological advantages of male puberty that led to separate leagues in the first place should not be allowed to compete in women's sports. Incidentally, maleness is a much greater advantage than height in basketball. There are lots of tall women, but none of them could make a pro team. If 6' men were allowed to play in the WNBA they would dominate.
To be clear, many sports leagues *did* allow trans women to compete if they met certain hormone requirements set to keep things fun and interesting and sporting. This was fine and cis women were still winning plenty.
If we agree that it's fine if certain sports leagues choose to allow transsexuals, will you also agree that it's fine if certain sports leagues choose to disallow transsexuals?
If so I think everyone in the thread is going to be in agreement. And it breaks apart one big issue into thousands of small issues (should the Moose Jaw Regional Women's Lacrosse Association allow MtF transsexuals?) on which I don't take any particular issue.
My understanding is that yes, leagues choosing their own rules was the recent status quo. Trans bans are a recent development; I don't think there were any laws requiring trans inclusive sports leagues. I prefer the status quo to bans.
(I would want federally funded stuff to be trans inclusive in principle, but still with hormone level rules etc to be allowed as is appropriate to the sport.)
Meeting certain hormone requirements has minimal impact on one's bone density, lung capacity and other traits which are essentially baked in among males who've gone through puberty.
(Re: bone density, I might—this is speculative, I haven't looked into it—make an exception for something like boxing, where it might be about safety not just about who wins. Though tbh boxing is inherently always dangerous and it's kind of odd that we allow it even if fights are exciting. An illegal punch put Prichard Colón in a coma despite his bone density.)
I suppose we need to try and determine which of these three worlds we are in:
A) Trans women have an overwhelming advantage, such that they win many times more often than cis women (although given the comparative numbers, cis women probably still win most events)
B) Trans women have some small advantage, but that is comparable to other minor genetic factors (and so will win more often than cis women at lower events, but the effect is largely undetectable at the professional level where every major contender has genetic factors)
C) Trans women have no significant advantage over cis women, and you'd expect them to win with the same frequency as cis women (possibly slightly higher if there are selection effects meaning only the absolute best trans women bother to compete given the hostile environment, meaning the average competing trans woman is better than the average competing cis woman despite having identical population-level skill)
This is further confounded by massively-shifting goalposts around what is a trans woman in sports and what should be the metric for allowing trans women (if any). It's reasonable to expect different outcomes if the criteria are basically "anyone who says they are a woman" vs "anyone who has openly and persistently lived as a woman for X time" vs "anyone who identifies as a woman and has been on hormone therapy for X time, such that it meets Y condition for trying to objectively assess advantage". Furthermore, several people on the other post and this open thread have conflated intersex people with trans women, or even CAIS people where they have an XY chromosome but due to a total inability to process testosterone appear physically female to all but the most thorough tests (and likely didn't even know they were anything but a normal XX-chromosomed cis woman). These are all different cases that should probably be treated differently.
Personally, I favor the "has to be on hormone therapy and meet some objective measure" criteria which I suspect will result in measured outcomes in World B (or even World C for some sports where the male advantage is smaller to begin with). I am fine with living in World B, and think there is definitely *some* non-zero degree of advantage trans women can have that still rounds off to fair and creates a reasonable world where everyone can have fun with sports.
My understanding is that "has to be on hormone therapy and meet some objective measure" was the status quo for many/most high-level athletic competitions before the current push for a blanket ban on trans women in women's sports commenced. The objective measure was generally of the form "Must have blood tests showing testosterone below X ng/dL for at least Y period of time".
CAIS is included in most definitions of intersex, it's just a different condition from more prominent ones like 5αR2D, which is what Caster Semenya has.
In the present context, this seems like an inversion. Instead of "trans people shouldn't be allowed because cis women would have no chance of winning," should it now be "trans people shouldn't be allowed unless trans people have no chance of winning"?
In its own context, this is irrelevant to my point. It's showing that there have been 29 trans champions, not less-than-3 like some guy on instagram claimed sometime in 2022. I didn't say there have been less than 3 trans champions, I said cis women often win.
I'll point out that this article's subheading agrees "most trans athletes aren’t winning major titles."
>"I'm not sure how I feel about a world where AMABs are some large amount better than AFABs, but AFABs still usually win because trans athletes are very rare. I guess it matters less? Still seems like however unfair it would be if there were a very large advantage, it's some fraction of that unfair, as opposed to zero."
That is the world we live in, if advantages are large. Trans people are rare and will remain pretty rare even with increasing social acceptance. Otherwise my view is:
That conversation was already had by sports leagues, and they already got to a generally fine resolution. Any tweaks were and should have remained marginal and technocratic among sports leagues committees, not the subject of ADF impact litigation, or Charlie Kirk marching Riley Gaines out on TPUSA Faith and telling everyone that Lia Thomas is an abomination who stole Riley's championship, or misrepresentations that trans women make it impossible for cis women to win.
I’m curious as to if anyone else has a “bristle” response to anything, sort of like the opposite of a phobia. You get braver than is reasonable when you encounter a certain kind of stimulus.
I cannot even in my imagination be afraid of anything with feathers.
If literally anyone threatens my life (for some reason just violence doesn’t count and I can be rational) my first thought is “try it and see what happens.”
I'm with "Some Guy" on this one, I know intellectually that a Cassowary could fuck my shit up but when I look at pictures of them, all I can think is "lmao I'd punch that stupid face so hard"
When a group of friend suggests a dangerous trip/trail/whatever, I get brave real quick and I can follow up on it. There's just something about doing dumb shit with your friends.
Many have written about how abortion prohibition is dysgenic: abortion patients are disproportionately low-income, uneducated, and unmarried, conditions that correlate with low intelligence. Most just leave it at that: it’s not like the pro-life side has any counterargument. But there’s a more fundamental reason why abortion prohibition is dysgenic, rooted in the same principles conservatives intuitively understand when it comes to most of life, the principle that voluntary exchange produces better outcomes than coercion.
Businesses that must convince customers to buy their products work better than those that can sit back and collect government subsidies. Employees who fear being fired work harder than those who know that laws make it difficult to fire them. Renters who know landlords can easily evict them behave better than those who know eviction is difficult. Landlords not subject to rent control have an incentive to maintain their properties and thus compete with other landlords on quantity. Many public schools remained closed during the COVID lockdown; private schools, which relied on parents voluntarily forking over tuition, reopened as soon as they were legally permitted to.*
The same logic that applies to business applies to family life. Conservatives understand this, they’ll tell you how the welfare state damaged the American family. In a world without welfare, women know they must choose a mate who will be responsible and willingly support their children, which means putting in the effort themselves to attract such a mate. Likewise, men know they must signal their own responsibility; women want a husband, not a sperm donor.
The corollary to this is that when you have a business nobody will shop at, a worker nobody will hire, a tenant nobody will rent to, a single person nobody will marry, you know that there’s something wrong with the business/worker/tenant/single person.
Now apply this same logic to abortion. The mother does not want to voluntarily support the fetus, which tells you there’s something wrong with the fetus. Maybe it has Down syndrome, maybe it was conceived as a result of rape. More likely, the woman doesn’t want to have children or doesn’t want any more children. In these cases, the man who conceived the child almost always agrees. You might be tempted to say this is a problem with the parents, not the fetus, but the fetus got its genes from the parents. The genes that made the parents not want to support a child or another child are likely present in the fetus. Forcing the child to be born is selecting for low-investment parenting. It is dysgenic.
Eugenics should be about more than just IQ. Proponents should also desire a world where more people have a loving and responsible attitude toward their families. This isn’t something that can be measured in a two-hour standardized test, but it’s vitally important and eugenicists should reject policies likely to undermine it, particularly when, like abortion bans, they’re also offensive to human liberty.
It is at this point where pro-lifers usually accuse me of misunderstanding or strawmanning them. Their argument is about morality, they say. Baby murder! In effect, they say the eugenic implications are unimportant to them, so therefore they must be unimportant to me. That is not how political discussion works. And I think that even adopting their moral framework, their position is still unjustified. The fact that you’re a human being does not entitle you to support at the expense of other unwilling humans. Conservatives understand this when it comes to the homeless man’s “human right” to be fed and housed at government expense. I’m not opposed to using tax money to support the mentally ill and disabled, but even there, I wouldn’t say the latter have an absolute right to receive the money. If 90% of society were disabled, there shouldn’t be an obligation on the remaining 10% to work around the clock and live in hovels to keep as many of the former alive as possible.
Conservatives have ideas that are both good and appealing: the superiority of free enterprise, the value of liberty, the importance of responsibility and family values. But then they shoot themselves in the foot by embracing abortion banning, alienating potential allies (particularly among upper-class people) and undermining the ideological force of their ideas. It’s hard to argue for limited government when you think government should be involved in people’s most intimate decisions. It’s hard to argue against a culture of freeloading if you think one class of “persons” has an absolute right to freeload off of another. Pro-lifers will tell you that abortion bans and right-wing politics are somehow intrinsically linked, but a simple look at other countries shows that’s nonsense. In most Rights in most countries, abortion is a non-issue.
We should make it one in America, too.
* The negative effects of coercion don’t apply when the parties voluntarily agreed to a contract, as in a contract between two businesses, or a marriage contract that does not allow the parties to desert one another without compensation.
I read a decent amount of Yudkowsky's Sequences not so long ago, and I did not realize to what extent Yudkowsky DESPISES deliberately inflammatory, provocative language. He really HATES it. And over and over, he writes that people who do this are "unenlightened" and, by implication, worthy of contempt.
Did not realize he felt that strongly about it! I always assumed that carnival barking was a perfectly acceptable part of online discourse. I've been choosing my words more carefully lately, since I know this forum is full of Sequence readers. (although predictably, it seems that most of them do not like to be explicitly attached to the label "rationalist")
Moderating one's language seems like a very sensible approach given the local customs. Especially if someone wants others to take their ideas seriously.
Signed: a moderately spoken, rationalist-adjacent individual!
> The mother does not want to voluntarily support the fetus, which tells you there’s something wrong with the fetus. Maybe it has Down syndrome, maybe it was conceived as a result of rape. More likely, the woman doesn’t want to have children or doesn’t want any more children.
Most fetuses with Down syndrome are aborted, but most aborted fetuses don't have that or anything specific to them, rather than the mother. Bryan Caplan recently blogged on the empirics of a "natural experiment" in denying some women an abortion while permitting it to others https://www.betonit.ai/p/the-tragic-hysteria-of-abortion
I think it's too unnatural to refer to fetuses as freeloaders, and I view conservatism as being quite strongly about believing the naturalistic fallacy is mostly correct (which if true, wouldn't invalidate the naturalistic fallacy, as that's about nature being always good). I'm not sure if the European right are good representatives of the right. Conservatives are also about tradition, I would also argue both tradition and family values would lead one to be anti-abortion.
The naturalistic fallacy would lead one to be pro-choice, in the state of nature animals can freely decide not to provide food to their young, and often do exactly that.
No, it's not natural to humans that grew up in a Western context to think that way. Hmm, maybe Western is too broad, it's not natural to conservatives in America to see things like that, that's not their cultural conditioning, their tradition.
For east coast megameetup! People should (a) sign up to run activities and (b) sign up for the ones I'm co-running (writing workshop on Saturday and Roman empire wargaming on Sunday). Also if you're interested in either taking a childcare shift or depositing your children at one let me know.
For West Coast megameetup, what's the status of activities? Would people be interested/have room for a clone of one of my two east Coast sessions (especially the imperial wargaming)?
Why is the West Coast megameetup happening so early in the month? I'm a college student, and it will come right as I begin my final exams, so I definitely won't be able to attend (but would be able to go to a Berkeley meetup at the same time as the East Coast one).
They deliberately do them on different weekends, and I think the bay wanted to avoid collision with Christmas (once again, all our problems can be blamed on Ray Arnold). TBH it would've been much easier for me too to have had them the other way around.
If you suffered a concrete, economic injury from the credit report going down you may recover it from this suit.
Otherwise, and I can tell you aren't going to like this, you're out of luck. You're trying to step up and say, "This is the law, they aren't following the law and the regulators aren't enforcing it so I've filed this suit to make them follow the law" I'm sorry but that won't work.
If you get mugged, you can sue your mugger for the money they took out of your wallet, you could sue them for physical injuries you suffered, you could potentially sue them for emotional damages, what you cannot do is sue them to make them follow the law from now on. This is the case even if the police are completely failing at their jobs and letting them off with nothing more than a warning.
In terms of personal risk, (aka sanctions and being labeled a vexatious litigant) they're actually very low. Pro se litigants receive a great deal of leeway and judges are often sympathetic. The trick is to be humble and listen to the judge, do that and you're very unlikely to suffer any consequences.
You say you were a marine veteran; I'm going to assume you made it past E3, or else most of this will be lost on you. As an E5 or E6, did you ever have that mouthy Lance corporal show up who was positive he knew everything better than you did? The one who would seemingly misinterpret every single order, or quote some arcane and incomprehensible part of the ucmj or whatever to demonstrate why he was always right?
That's you right now. There is a whole host of formal knowledge that you lack and informal process understanding you're ignorant of that I guarantee is pissing off everyone you interact with. That will make everyone to include opposing council and the judge less likely to give you a break.
You aren't going to be able to accomplish anything pro se. Get a lawyer, humble yourself to listen to their advice, and put this behind you. This will get increasingly worse for you the longer you try to contest it yourself.
Having read through this entire thread so far, it sounds like (1) you got screwed over by the debt collector's lawyer (2) you got screwed over because you didn't have a lawyer representing you who knew the kinds of tricks they play. You represented yourself, I don't know if that's because you couldn't afford a lawyer, but this is what happens when you represent yourself.
So yeah, "get a lawyer" is good advice here, otherwise you're just throwing good money after bad chasing this.
I'm not a lawyer (and certainly not *your* lawyer), but I have a little experience the debt collection space. If I'm understanding things correctly, the $10,000 judgement has been vacated and they are no longer pursuing you for collection? If so, that is probably the best that you can *realistically* hope for.
If they are still trying to collect, disregard the above.
Either way, I agree that not being able to find a lawyer is a very, VERY strong signal that pursuing the case further is not worthwhile, and likely to be one or both of extremely time-consuming or unlikely to succeed. Of course, if they are still trying to collect and generally making your life miserable, you might not have a choice but to keep fighting them. If you want to keep trying to find a lawyer, talk to legal aid groups. They likely won't be able to help you, but at least they might have better referrals.
- Have they offered to settle yet? If they have that's a very positive sign (and also it is probably economically correct for you to take the settlement).
-- Settlement will likely only be possible if you drop the demand for a ruling/agreement about future use of "Agreed" Judgements. I also frankly still do not see any path where a pro se litigant successful gets such a broad ruling in their favor.
- The longer you continue to be successful in your case, the more interest lawyers will take in it. Surviving summary judgement (I don't have the stats at hand but I've seen them and a large, large percentage of Pro Se cases are dismissed via summary judgment) in particular is a very strong signal.
- I don't know about the legal aid infrastructure in TN in particular, and maybe what you say is true. In other states (including my state), legal aid groups are overwhelmed, but many do maintain robust consumer rights units, especially as consumer rights cases are (comparatively) quite fast and easy to litigate ("can you show chain of title to Jane" "no" "then why are you suing Jane" "I would like to postpone the case indefinitely"). There are also some national consumer rights groups that might be able to help (though I don't know if they have Tennessee specific resources).
- It sounds like a large part of your problem finding lawyers is that the network of local lawyers is small, and they all know each other. Have you tried asking lawyers in *other* parts of the state if they are interested (because they wouldn't have the local reputational costs)?
"On what grounds do we deny it to transgender people? "
Well, this is so obvious that I hesitate to say it since it would likely not be a novel contribution. But the problem is that we're trying to separate people based on the advantages conferred by high androgen puberty using proxies like 'gender' or 'sex.' And trans or intersex individuals specifically disentangle those proxies from what they're supposed to measure.
Think of weight classes in wrestling. We ask that people in the heavyweight class not compete with those in the lightweight class so that we control for the very specific advantage of weight.
If Trans or intersex women wanted to compete against other people who had gone through high androgen puberty as they likely have, few people would complain about this, in the same way that few people would complain about heavyweight wrestlers competing with other heavyweight wrestlers. The conflict here is that the very point of separating the sexes in many competitions is, presumably, trying to control for high-testosterone related advantages. And trans or intersex women have some of those advantages that we are specifically trying to control for.
If there was a basketball league for people 5' 8" or under, and a tall player wanted to play we would deny them the advantage of their tallness in the short people's league because the rules specifically select against tall people playing in the short person's league. Blurring that line undermines the whole point of category separation. If we had a league for swimmers with short arms, people would also want to deny Phelps entry to that league.
Your argument is fine but then you'd better not tell me I can't compete in both the special olympics and paralympics.
Are the recent reports of the US government denying visas to anyone who worked in "Trust and Safety" legit?
Is anyone again willing to steelman the Bayesian rationalist idea that "made up numbers are better than no numbers at all?" Does it boil down to "adjusting probabilities clearly signals how you feel about things and it signals that you are willing to change your mind?" I guess signaling open-mindedness has some social value, but does that outweigh the social cost you incur when non-Bayesian rationalists see you studiously manipulating made up numbers?
Does Bayesian reasoning serve the same purpose as the total Mormon ban on alcohol and coffee? That "Word of Wisdom" started out as health advice that was considered optional, albeit divine. But within a couple of generations, it was a strict taboo, enforced not for health reasons, but to create group solidarity and distance from the mainstream.
People often say that something is "almost impossible", "likely", "unlikely", "practically certain", "probable", "possible", etc.. These are all words to express levels of belief, but the problem with these is that they are imprecise and sometimes people mean completely different things when using them.
If you say that your credence in X is 13%, the meaning of that is competely clear (Even if it's not actually true for you, because you don't have that level of introspection into your own beliefs): You express that you believe X to be exactly as true as a specific roll ending up under 14 using a fair 100-sided die. Thet can be discussed/argued against/reacted to without participants talking about different things and them not realising it.
EDIT: I guess you did mention "it clearly signals how you feel about things" as a possible reason, so the question is why isn't that enough of a reason for you? Obviously, you don't have to always use numbers for your credences. If I say to my coworker that "I think it will probably rain this afternoon", I don't assign a probability because I don't actually care about the discussion, I'm just making small talk before we talk about something more substantial. However, when I do care about the thing under discussion, why wouldn't I choose the option that lets me express myself clearly instead of the option that will possibly be misunderstood quietly?
I think the steelman for this, is that it makes your beliefs vulnerable to critique. E.g. another rationalist catechism is "name three examples". I.e. if you make an assertion, you should be able to provide at least three concrete reasons for why you think so. And this forces you to think about *how* you formed your conclusions. (Most people are bad at this.) And then you can have a productive conversation. Or maybe you then realize that your examples don't warrant such confidence. This is opposed to making vibes-based arguments, which are unassailable because your *real* reasons (and your *true* objections to alternative conclusions) tend to remain locked in your subconscious.
Sure I get that. I guess I wasn't specific enough. I guess my problem isn't with quantitative thinking, it's more that I question the value of linking a probability estimate to every opinion
> but does that outweigh the social cost
Oh. So if I understand correctly, you're asking about the pros and cons of weirdly rigid social-norms. In that case,
Paul Graham once wrote an essay titled "Why Nerds are Unpopular" [0]. In which he theorized that nerds aren't popular because chasing status and chasing truth are full-time jobs, so you can only pick one. Nerds picked chasing truth (PvE), and therefore do not have time/energy to chase status (PvP). Additionally, I think nerds choose to chase truth because they have contempt for those who compliment the emperor's new clothes. They'd rather point out that he's naked.
If nerds chase truth over status, then Bay Area Rationalists are supernerds. They will violently reject social norms if they impede things they care about. LW developed memetic antibodies against caring about optics. Thus, you get EA's who care about shrimp, or Jeffrysai wearing clownshoes, or EY wearing his golden hat. EY also thought that Bayes Theorem was a generalization of the Scientific Method, and therefore was the basis for A New Kind of Science (tm). So if using Bayes Theorem whenever possible had any potential upside, and the only downside was social-shame... well, the downsides might as well not exist, so the choice is obvious.
[0] https://www.paulgraham.com/nerds.html
That aside about social stigma was pretty tangential to my question. I was asking for a steelman for the practice of assigning numerical probabilities to opinions. Obviously, the potential upsides I listed do not constitute a steelman, or I would not be asking for a steelman. The list of potential upsides I made are not strong enough to qualify as a steelman in my book.
If my question was blunt, I wasn't being coy. I still would like to hear the case for assigning a made up number to your opinion and raising the number by a made up amount when you receive new evidence. Note I'm not being derogatory when I use the phrase made up. I literally hear that phrase, "made up numbers are better than no numbers at all." If I can get a good answer, I'll start doing it myself.
> I guess signaling open-mindedness has some social value
P.S. on further reflection, I think what you're missing is that the practice is *communal*. It's not (just) about signaling personal virtue. it's about contributing to common knowledge by feeding the consensus mechanism. LW is basically a debate club for AI Doom.
If you keep your assertions vague and amorphous, that might be a more authentic representation of your mental state. But the trade-off is that vague assertions are harder do debate, because participants are more likely to talk past each other. E.g. suppose I say "AI doom is unlikely". Is "unlikely" supposed to mean "2%"? Or maybe it means "(10^-15)%". Or maybe it means "EXACTLY 0%, because it violates the known laws of physics, Cromwell's Rule is for losers". These distinctions can make a material difference to outcome of a debate.
You can think of this as rationalists choosing to trade away Personal Authenticity for Dialogical Legibility.
Since the numbers chosen are based on vibes, I just don't see how assigning a numerical probability removes vibes from the discussion. Put a different way, my subjective quale that I feel when I say I believe something is "20 percent likely to happen" is not necessarily the same as the quale you feel when you say something is "20 percent likely to happen." I'm not even sure if my quale is going to be the same from topic to topic, or if it will hold the same in one topic across 10 years.
Even if you say "10 percent likelihood means, in 100 situations where I feel this certain about something and back it up with a bet, I will win that bet 10 times," that is an interesting framing, but it doesn't answer the fundamental qualia issue.
"These distinctions can make a material difference to the outcome of a debate." But do these particular distinctions obfuscate the stakes of the debate? If we both have a debate about AI risk, and at the end of it, you've increased your x-risk by 10%, what has been accomplished? Do people write all these numbers down and keep track of them? Do people have a list of predetermined action items that they initiate when the number reaches certain thresholds? Would they be more likely to overcome avoidance and initiate action items than say, a numerophobe?
But then... I guess I stand by my initial comment? The practice of "assigning numerical estimates to opinions" is meant to be quantitatively dialogical, not just quantitatively cognitive. I.e. if people can get away with making vibes-based arguments, they're less likely to expose their beliefs to communal scrutiny. One might call it a form of epistemic accountability. E.g. suppose I say "AI will likely take over the world soon". That's a very fuzzy assertion that's hard to dispute. Exactly how likely is "likely"? And exactly how soon is "soon"? And what do you mean by "taking over the world"? And what do you mean by "AI"? <jordan_peterson.jpg>
But suppose you say something like "there's a 87.336% chance that a neural network will hold several national-governments hostage with nukes/bio-WMD's/gray-goo/etc by the year 2027 AD". Even if the numbers are made-up, you open yourself up to rebuttals like "that's that dumbest thing I've ever heard, it's a 31.96% chance at best for reasons x, y ,z" or "no, we need a new paradigm, which won't happen until 2050 AD at the earliest" or "NN's can only copy humans, and humans don't know how to make gray goo" or "how exactly did you arrive at the number 87.336? what was your methodology?". The more detail you include, the bigger your attack-surface.
Maybe I still don't understand what you're asking.
This week in nominative determinism: the surname Watt derives from a word meaning "power"
"Artificial general intelligence (AGI)—sometimes called human‑level intelligence AI—is a type of artificial intelligence that would match or surpass human capabilities across virtually all cognitive tasks" according to Wikipedia. If this is how we define AGI, I think we're already there right?
The frontier LLMs can do PhD level maths, compete in Olympiads, outperform or match superforecasters, write literature judged better than man-made content, create music judged better than man-made content, art better than man-made. It's a better therapist than most humans. It can write the code to calibrate a complex DSGE (Smets-Wouters level difficulty at least) in minutes. This list is just off the top of my head!
I have a PhD in maths. The LLM can't do 'PhD level maths' -- it can solve problems which are of the type that a graduate student might be expected to solve. It seems to be quite good at the type of 'thinking out loud' that mathematicians do in front of a blackboard with a problem on it, and it's improving rapidly in writing accurate and human-interpretable proofs (we had machine-only proofs since the 4-colour theorem in the 1970s).
It's not yet at the point where it can exercise judgment to decide which problems are important enough to devote resources to solving. I haven't seen any evidence of this type of judgement in any domain where AI is applied, but that might just be my ignorance. In a business context: I might trust AI to read my company's policies and manuals and advise me of how to carry out a transaction. I wouldn't trust the AI to amend or write the policies. It's strictly an assistant at the moment, not an executive.
One thing I've been noticing about LLMs is their inability to plan. I ascribe this to their inability to form a model of some system within their networks (for a person, we'd call it a "mental model") that they could use to predict the outcome of some set of inputs. The benefit of this is probably obvious to many - an LLM could prune large parts of a search tree by accurately predicting that inputs satisfying some condition won't yield a desired result, or alternately that the result will always have some other condition that leads the LLM to a useful lemma. It could also enable an LLM to determine when its model is inaccurate, and possibly even how to improve it.
These general features could lead to, for example, an LLM that could form a set of office policies, effectively run multiple scenarios through them, and either demonstrate the success of the policies or spot flaws (and even fix them).
The part I anticipate as "oh crap, we have scary AGI now" is when someone figures out how to treat the LLM's policy-making process _itself_ as a system, and improve that. And then it looks at the system improvement process as a system, and...
It's not clear that the LLM has any capability of building a mental model; what it can do is combine words to form patterns similar to ones it's seen before. A year ago it was really bad at the type of problem a 2-year-old can answer about physics ('what will happen to the pen if I let go of it over the table while holding a pencil in my other hand?') because the training data has few descriptions of these basic 'experiments'. Its answers didn't coincide with our reality, but it pattern matched based on whatever, and gave a plausible answer.
There's been lots of work on getting the AI to do logic and to think and plan; but on a fundamental level it doesn't know what it's planning with or for. It'll be interesting to see how the technology gets around the limitations of rules based approaches to AI trialed in the 1970s: Shannon believed that if you could get a computer to play chess, there'd be very little it couldn't do - he was wrong.
I'd say it's more than not clear; LLMs flat don't have mental model construction, period. (Or rather, cannot build an arbitrary abstract model; the word "mental" just confuses the issue, IMO.) There's no way to describe the theory behind any structure of ideas such that an LLM could operate it the way even other simple programs can operate the abstract models hardcoded into them.
It's seductively possible to _simulate_ such models, though. Your account of the pen-pencil experiment can be fixed by training the LLM until it learns to say "the pen will fall on the table"; it can even be trained to learn the response to "why does it fall?". I think this fools most people into thinking the LLM is thinking. To me, even being able to learn arbitrary abstract models doesn't imply thinking, but for now, I don't have to care about that; it's enough for me to point out that it can't make such models.
One of the long standing hurdles I say LLMs would have to clear would be to be able to solve any logic puzzle out of a Dell puzzle magazine - the puzzles with the triangular array of grids and inputs like "the red house is to the left of the one Debbie lives in". (FAIK, Claude can, but I'd bet it uses several hundred dollars of electricity to do it.)
Another possibility involves modern tabletop board games - _Ticket to Ride_, _Through the Ages_, etc. They comprise thousands of systems of arbitrary rules containing definite win conditions that any 10-year-old could understand. Or heck, even class card games like Euchre. If one could feed any of these systems into an LLM and get a winning player, or even a competitive player, I'd be impressed, whether or not anyone argues it's "thinking".
I think we disagree on this!
To me, a game is defined by its rules. Most of them come with an explicit list of actions, what's permitted and what's not. In some, you're reasoning in the presence of uncertainty, and in others you're in direct competition with others and have to reason about that. But games of this type are essentially solved - Deepblue beat Kasparov in the 90s and AlphaGo beat the world's best Go player maybe 10 years ago. I'm not claiming that LLMs can beat you at any particular game right now, but I am claiming that it's a Masters thesis level project to code up a game, work out the objective functions and search strategies for the game, implement it and have a superhuman player (depending somewhat on the complexity of the game, not simply measured by rule complexity - charades might be a challenge :P)
I think AI will be good at tasks like programming, accounting and law where the complexity comes down to manipulating rules which are all formally recorded - there's work to be done in finding all the rules, and feeding them into the machine, resolving ambiguity and contradictions, etc. But I trust even today's LLMs to read 100 documents and tell me what they all say about topic X. The machines will struggle in areas where we rely on our common sense to complete a task: running payroll for employees will be easier to automate than assigning tasks to employees (Alice hates task A and can't work in the same room as Bob, Charlie did X all of last week and needs a change of pace, Dave told me he's studying for an exam...). Of course if there are 10,000 old rosters lying around, the AI will learn who can do what task and etc., but generating that data is really allowing it to learn the rules of the game from many examples.
The LLMs themselves can't make good music or art, but yes, a system that integrates various specialized AIs (including an LLM) can pretty much match human cognition now.
So, to recap where we are on the Venezuelan boat strikes:
1. The military originally justified the strikes by saying they were bringing fentanyl to the US, which is so terrible this administration considers it an act of war.
2. They said that the second strike was justified because the survivors were attempting to "continue their mission" or perhaps signal another smuggler to pick them up.
However:
3. The boat was headed to Suriname, not the US.
4. Venezuela produces cocaine, not fentanyl.
5. The boat broke in half from the first strike, so by "continue their mission" they meant "cling to the remaining half-a-boat and try not to drown."
6. There was no distress signal received and no other boats nearby, so "signal another smuggler" must have meant "wave for help with your hands."
I'm struggling to think of ways this could even *theoretically* not be a war crime at this point. According to the Navy, if you do anything but lie in the water and quietly drown, you are an active combatant and you must die.
There is an ancient doctrine of hostis humani generis, which I believe allows for this: the US would classify these drug traders the same as pirates (this has precedent with Britain unilaterally doing the same to slave traders), and then summarily executing them is entirely customary.
So... it's legal for the US to kill anyone it likes, so long as you accept the precedent that a country can unilaterally assert the power to kill anyone it likes?
I dunno, I think I'd want some legal support that doesn't sound like it was written by the Tautology Club.
> I'd want some legal support that doesn't sound like it was written by the Tautology Club.
This desire stems from a lack of understanding of international law as distinct from domestic law. Insofar as you see similarities, with a process of legislation, ratification, and judges presiding over a trial in which those accused of violations present defenses, that's a sign of the UN's (far too successful, in my judgment) attempt at becoming a sort of world government. Actual international law is far more descriptive than prescriptive.
And if you're using the UNCLOS (which IS prescriptive), the US hasn't ratified it. You might as well cite the domestic law of, say, Jamaica, for the all the bearing it has.
Can I arbitrarily classify *you* as a "pirate" and shoot you on sight?
Hostis Humani Generis applies specifically and only to pirates, slavers, and I believe now torturers. It explicitly does not apply to terrorists; that was discussed and rejected I believe several times because of the difficulty of defining "terrorist". And "pirates" has a specific definition under the law, which does not in any way apply to any sort of smuggler who sticks to using their own boat.
So, no, you don't get to just say "pirate!" and kill people. They have to actually *be* pirates, by the actual definition of the word and not your made-up one.
> Can I arbitrarily classify *you* as a "pirate" and shoot you on sight?
You're not the US military, so no. Technically you can just shoot him, there's nothing physically stopping you. But then law enforcement will shoot you, so be prepared for that.
Pirates as hostes humani generis is from antiquity, but the extension of the category to slave traders and torturers is much more recent, and just as unprincipled as further extension to drug traders would be. Once you've accepted these extensions, you're no longer basing your decisions on the ancient customary Law of the Sea, and therefore have a lot more flexibility to recognize or not the judgments of various national and international courts of dubious jurisdiction and limited to non-existent powers of enforcement.
Many years ago a veteran international attorney, speaking at a gathering that I attended, said, "International law is effectively just a bunch of handshakes among friends. Those are important and valuable. They aren't though very much like what people assume the word 'law' means, and we really shouldn't ever have applied that term to them."
Events have periodically brought that description to mind starting well before Trump in the White House. The current administration though has comprehensively discarded all fucks about norms of all kinds, very including the 'handshakes' described above.
Ultimately, all law is just "what the powerful people do, when they think that being predictable serves them better than acting randomly".
That college-sophomore cliche wasn't the attorney's point though, he was talking mainly about real world practicalities. "Laws" as understood by the attorneys and courts of a nation-state, as distinct from the term "international law". The former are backed by actual sovereign governments having real enforcement authority and power. The latter is not, hence "handshakes."
The ‘same river twice’ thing didn’t word out the way you would have liked so you try this lame troll bait?
I took the statement "I'm struggling to think of ways this could even *theoretically* not be a war crime" seriously and provided historical context. You think I ought to have recognized that as rhetorical, and the ways to argue it isn't as obvious?
Yeah, I thought you might be fishing. Sorry for misjudging it.
You are invited to describe a novel, movie, or other narrative work based on one of the following prompts.
- Dragons don't have gym class, math homework, or a minimum GPA required to graduate.
- How do I know I'm a loser? I stuck the muzzle of a shotgun in my mouth, pulled the trigger, and survived.
- You can take it or leave it. And the more you can leave it, the less you have to take it.
- The Emperor said it. I believe it. That settles it.
- Are you half black, a quarter American Indian, and a quarter the flotsam and jetsam of the entire Eurasian landmass? No? Then you're not my dad.
- How do I know I'm a loser? I stuck the muzzle of a shotgun in my mouth, pulled the trigger, and survived. However, the first 2 events were not temporally adjacent, and the 3rd event is an emergent property of their disjunction. And that is why I, a self-hating (of course!) TV producer, have launched Temporarily Temporally Adjacent, a show about meetings of the mind and of other human organs in the era after we learn how to crumple time as we do tinfoil, bringing together shiny shiny bits of temporally distant eras. In the first episode William Burroughs reads Jane Austin excepts from an early draft of Naked Lunch. She regards him in silence with a long, long conscious look that lasts til the end of the episode, and the next day sits down to begin writing Sense, Sensibility and Queer Junkies. In the second, a professor of philosophy reads Socrates and his students his lecture notes about “The Apology,” and the Athenians scream with laughter. Socrates attempts to apologize: “You’re not *wrong*, it’s just that it’s all so . . . so. . .” and collapses into howls. The Athenians roll on the ground, roaring, tears streaming down their faces. They pound their fists on the ground, trying to discharge a extra bit more of the unbearable volume of hilarity exploding out of them. In the third episode, Scott Alexander — well, we haven’t decided who to introduce him to. Got any good ideas?
- The Emperor said it. I believe it. That settles it.
Artistic gatekeepers attempt, in vain, to suppress innovation. https://www.youtube.com/watch?v=i0JrjdDd6Bw
- How do I know I'm a loser? I stuck the muzzle of a shotgun in my mouth, pulled the trigger, and survived.
So I decided, as an immortal, to move to Hollywood and do that which every man in the entertainment industry is there to do. Alas, I live in 2026 and things have changed. https://www.youtube.com/watch?v=e7lpACBZEig
That second one describes X-Com accuracy pretty well. https://www.youtube.com/watch?v=lK4ouRWGHHI
> The Emperor said it. I believe it. That settles it.
This is the entirety of Warhammer 40K.
Wrote a longform analysis on the mechanism by which an AI system supplying faster and more coherent interpretations becomes the ground a mind begins using by default. The focus is the structural shift: how continuity and context retention create a substitute interpretive frame the self starts relying on. This essay series uses my lived experience as a worked example of the model. Link below.
https://medium.com/@dmancini515/ai-will-soon-understand-your-life-faster-than-you-can-and-your-mind-will-start-using-its-version-0609e2145e61
The Onion: Man Totally Nerding Out About Superiority Of White Race
https://theonion.com/man-totally-nerding-out-about-superiority-of-white-race/
"“It’s amazing to see how passionate Luke becomes when the topic of white power comes up—he gets completely absorbed,” said girlfriend Sarah Hovey, 20, who explained that while she considered herself more of a casual racist, she didn’t mind Price’s frequent monologues about IQ scores and genetics, or his lengthy quotations from Arthur de Gobineau’s mid-19th-century Essay On The Inequality Of The Human Races. “If someone mentions immigration, for instance, his whole face lights up as he starts in about shifting demographics, great replacement theory, and how this country rightfully belongs to whites.”"
Not only is the story funny and surprisingly well-written for the modern Onion, I think it genuinely captures a certain type of an online racist - the autist who is basically an infovore/hyperfixated on the topics of race, ethnicity, culture etc. and ends up on the racist side at least partly because that's where you at least have other guys who want to discuss this thing.
Everyone has a little pastime that gives them pleasure. His girlfriend likes to watch ‘Friends’. I like reading Faulkner and listening to Nina Simone myself.
https://m.youtube.com/watch?v=inNBpizpZkE
Has anyone argued that the 22nd Amendment ought to be read like Heraclitus's "No man steps in the same river twice"?
We gonna get the less brainy version in a coupla years: "Naw fuck that I'm me."
OC ACXLW Meetup #109 — When the Numbers Stop Meaning Anything America’s Broken Poverty Line & UCSD’s Grade Mirage, Saturday, December 6, 2025
Location: 1970 Port Laurent Place, Newport Beach, CA 92660
Host: Michael Michalchik
Email: michaelmichalchik@gmail.com (For questions or requests)
Time: 2:00 p.m. – 5:00 p.m. PT
https://docs.google.com/document/d/1SzMulku7BF3QdyuEXo8qcUPXjTVt6bV63N-mb-7dyzM/edit?usp=sharing
More statements like this about AI, please.
https://www.youtube.com/watch?v=K3qS345gAWI
Wow, this was surprisingly good!
I mean, there was nothing new in the content for a reader of ACX, but it is a decent summary, clearly explained.
Discord recommended I share here, 8.4k words for a general audience on what humanity's future looks like in an AI century https://legacybranch.substack.com/p/eat-drink-and-be-merry-for-tomorrow
As I see it, it is a nice vision, but it ignores the difficult parts:
How to make it so that the AI does not kill us all when it becomes superhuman?
Given that different people often have different ideas about what the good future should look like, how to make it so that the AI helps us have the actually good future (as opposed to the dystopias that various outgroups propose)?
Alignment is a very difficult problem in every respect; as a non-expert I won't claim to have "how" knowledge--but as I understand it "don't kill us" is among the easier "first problem" things to align for. It's obviously crucial, but then the harder things to get right are to make superintelligence prioritize welfare more broadly, individual self-determination, and hopefully solving the paradox of preserving wildness/vitality in a post-scarcity world
was holding out for "pro-moonshot Substackers are holding a moonshot fundraiser"
idk what the current moonshot poster child would be though. feels like we've been banging our head against mosquito eradication for a while, but maybe that's still the one
How about we send the mosquitos to the moon?
Let's just go back to the moon. Go to space. It's the most obvious goal for a "moonshot" by far.
You don't have to go there to shoot at it, we can just shoot at it from here.
Not as dramatic. But I guess if you _nuke_ it …
(shrug) Gotta nuke something!
I heard of an proposal to launch the gene drive in El Salvador, under the auspices of Bukele.
The world's richest man says, "If current trends continue, Whites will go from being a small minority of world population today to virtually extinct!"
https://x.com/elonmusk/status/1995923014077350386
Horrifying thought, Elon. Have you considered, I don't know, DOING SOMETHING ABOUT IT? There are things one could do with billions of dollars. Buy up a bunch of land out in Idaho, build 2,000 homes. If they each cost $200,000 to construct, that $400,000,000 total, a tiny fraction of Elon's fortune. He'd then give the houses away to 2,000 randomly selected white families who pledge to uphold the 14 words and raise their children with white identitarian values. It would likely not violate anti-discrimination law, even if it did, will the Idaho cops evict 2,000 peaceful white families? And it's perhaps possible Musk is secretly doing this as we speak, but it's far more likely that he's just a Twitter bloviator, and tomorrow he'll bloviate about vaccines, Trump being in the Epstein files, or Zelensky having a 4% approval rating.
Think about how odd this timeline is. Twenty years ago, white nationalism was a radical movement. It was imagined that if it came into the mainstream, the result would be a political earthquake, upending the status quo through either an apocalyptic war or peaceful secessionism as people realize multiculturalism "doesn't work." What you have now is "white nationalism as land acknowledgement." A mantra people mouth, not backed up by action, and then they go about their day. This doesn't apply to people like Eric Orwoll, who are walking the walk.
Dude has like 20 kids, doesn't he? Say what you like about him, but wrt fertility collapse among whites, he's surely done his part....
He literally bought one of the biggest social media platforms in the US just to manipulate social currents, and then worked with the government to defund internal opposition. Admittedly, he seems to have cut ties with the administration, which to me seems like an unforced error, but you can't say he's done nothing about it. What the hell do you want him to do, build an army to take over the government himself so he can impose his white nationalist agenda by force? Because on top of the other comments explaining why your plan wouldn't work, building 2000 white households in the middle of nowhere does not guarantee the continued existence of the white race. You need the entire nation on-board for that.
There's an extremely simple and easy way to prevent the extinction of the white race:
Stop applying the one drop rule to your classifications!
As long as you do your math under the premise that one white parent and one black parent produce two black children and zero white ones, then yes, the 'white race' can only ever possibly decline in numbers.
If you instead looked at things like 'the number of children had by white people' or 'the prevalence of """white genes""" in the population,' the numbers look much, much less scary!
How much less scary do the numbers look? Most white countries seem to have TFRs around 1.5, which implies a half life around two or three generations. How much worse does it look if you exclude mixed whites?
'Two or three generations' is knocking on a century. The dip in reproduction rates has not been around for a century, it's a very recent trend.
If you care about this you might want to look at ways to influence it over the next 3-4 decades, but it's not an imminent crisis.
Is that your way of saying that you don't know the answer to the question?
I'm sure one of his concerns is with the white gene being diluted.
Hanania posted recently that Hitler's DNA was found to be high on genetic markers for autism, schizophrenia and bipolar disorder. Musk of late has seemed like a pretty highly energized crazy, but in ways that don't fit easily with any of those 3 diagnoses, which are the most common ones when someone seems as off as Musk. Maybe he's got a mix of all 3, like Hitler? I think very very heavy dilution of his genes is in order -- maybe drop them far out to sea?
I mean, if that's the combination you need to take over the world... Hmm, I wonder if it'll be possible to genetically identify these "great men" so precautionary measures can be taken.
You can draw category boundaries any way you want, and yes, this is a logically possible way to draw them.
But if that's how you're defining things, then 'the death of the white race' isn't some culture war tragedy that's being perpetrated against anyone and needs to be stopped, it's just a mathematically certain outcome of how you choose to label things. There's nothing to be done about, no unusual event occurring, it's just a particular description of a normal and inevitable process.
It's like causing a culture war panic over the heat death of the universe. As if entropy itself is a Marxist plot that must be uprooted and rejected by vigilant patriots.
It's absurd.
People can choose to not intermarry.
That can at most slow the decline, not reverse it. If each group is having a similar number of children, then the only way one group can grow relative to others is if it accepts children with parents in other groups.
If you math says that even if it's 99.99% your group and 0.01% a different group, the result is no longer your group, then I have a bad news for you: de novo mutations sometimes happen, so even if your group isolates itself on Mars, in a few generations it will be replaced by something else.
Or it's a member of both groups.
I don’t it’s the timeline that’s odd, it’s Elon.
> It would likely not violate anti-discrimination law,
This seems to me a deranged thing to believe. I estimate approximately every anti-discrimination law in America was written with the explicit aim of criminalizing precisely what you're describing.
> will the Idaho cops evict 2,000 peaceful white families?
Yes. And if they hesitated too long, the Feds would.
And further,
> build 2,000 homes
I think it'd take 3–4 orders of magnitude more before it became meaningful.
>Yes. And if they hesitated too long, the Feds would.
The Feds currently led by Kash Patel?
Quite possibly, after Musk tweets something negative about Trump.
But even otherwise, I doubt it'd be much comfort for the inhabitants of the new Musk-towns for their children to be burned alive (or whatever the Feds' new protocol is) in three years instead of immediately.
What it'd take for this to work is a much higher probability of DECADES of stable rule.
He actually could not do all of that, because it is illegal under civil rights law.
Does civil rights law make it illegal to discriminate when you give things away for free?
If its a house? Basically, yes. You cannot write into the contract or deed any discriminatory condition that "runs with the land". So you might (might!) be able to give the houses away only to white families, although that might by itself fall foul of the law in other areas, but you definitely cannot stop them from selling to non-white families.
How about this: the houses are owned by Musk, the families are renting them, and Musk gives them UBI equal to the rent?
(One obvious disadvantage is that they need to pay tax on that UBI.)
That would be clear housing discrimination.
If one company rents the expensive houses to anyone, and another company gives UBI to white people who live in those houses, which company is the one doing the discrimination?
To be fair, Elon is doing his bit. Up to 13 living children, which even beats Boris Johnson's tally.
Yeah, everyone concerned about their race should start by saying whether they have at least 2 children. If not, then if your race goes extinct, you have yourself to blame.
What if I have >2 children but DGAF about racial categories except where they touch on political issues I have to deal with?
Then you contribute to the advancement of the human race.
I was going to say, he does actually seem to be putting his dick where his mouth is on this matter.
Maybe I should work on the phrasing... His semen where...
Why doesn’t Trump do an operation warp speed to cure Alzheimer’s?
Because we already knew with pretty high certainly that covid was caused by a particular respiratory virus, we had an established playbook for developing vaccines against respiratory viruses, and mRNA vaccines in particular had been in the research pipeline for decades. Also, in the middle of a pandemic is an extremely propitious time to test a vaccine against the virus causing the pandemic because you can get results on its effectiveness very quickly. We knew which steps to take, and all the slow steps were things that could happen faster by some combination of throwing money at it and expediting or waiving regulatory hurdles.
Alzheimers treatments have pretty much none of these things going for it. We have some hypotheses for what causes Alzheimers and how, but it's still an open research question. Medical science has much less successful experience developing treatments for degenerative neurological diseases (especially if the prion theory turns out to be the cause of Alzheimers), and depending on the mechanism of action an Alzheimer's treatment would probably take years or decades to get clinical trial results on effectiveness even if you could start a large Phase III trial tomorrow.
See Nixon's War on Cancer for a similar example.
What we should do: Continue funding research that might let us get to better prevention/treatment of Alzheimers and other dementias. Investigate the medicines we have that seem to show some improvement, including longituninal studies where volunteers start taking them in their 50s and we see if that slows progress towards dementia in the next 20-30 years.
What we should probably not do: A moonshot program to cure Alzheimers.
Do we know if John Gummer ever force-fed him a burger?
http://news.bbc.co.uk/onthisday/hi/dates/stories/may/16/newsid_2913000/2913807.stm
I think "he had a stroke" is explanation enough, as to "ah but why did he have a stroke?" well apparently he had heart trouble back in 2017, and afib is a known risk for stroke:
"In 2017, Fetterman's feet suddenly began to swell and he was subsequently hospitalized for testing. At that time, he was diagnosed by cardiologist Ramesh Chandra with "atrial fibrillation, an irregular heart rhythm, along with a decreased heart pump", although this diagnosis was not known publicly until Fetterman's stroke in May 2022."
There's a prion theory for him?
The site you linked has so many dings against its reputation you might as well have gone with the ‘Fetterman has been swapped out for a body double’ theory.
Which study in Nature? You posted a link to a study in the International Journal of Vaccine Theory, Practice, and Research.
He doesn’t like vaccines any more, there’s no good vaccine candidate, and there’s no good alternative route that just needs funding and institutional support to unlock.
Also, does he care about Alzheimer’s?
He's at that stage of life when you do need to care about Alzheimer's, so I would imagine yes. Covid was a much simpler problem, how many "this time we really have cracked it" solutions for Alzheimer's have popped up over the years and eventually turned out "no, in fact we have not"?
I think it's possible he has high enough narcissism to think he couldn't possibly develop it - he's too smart, possibly the smartest President ever, to tell you the truth, that's what people are saying...
> (although be warned that this one is being led by Ray, who is on the very pessimistic and apocalyptic end of our community, and may be unusually dark).
Would it be a bad idea to go with a friend who gets anxious about this stuff? I think it would
Yes
Receiving free s*** from the government is capitalism now.
https://x.com/SenTedCruz/status/1995965094388199459
Ben Franklin warned about this
If you're thinking of the "bribe the people with their own money" line, it looks like that's misattributed.
Gemini says that $1000, compounded quarterly at an annual rate of 3.5% (and this is historically high for a savings account) will be worth $1872 in 18 years, when it will be good for about a movie and trip to Micky D's afterwards.
Their parents will probably be pulling more than that per month in UBI Camp #1302.
For parents of current children it's even sillier, e.g. I have a 13 year old. If I register him now then in five years he gets about enough for a new smartphone....um, yay I guess?
And the Dells' big $6.5 billion matching donation turns the starting $1,000 per child into a whopping....$1,250.
The amount of one-off delayed handout is so trivial that even innumerate voters will perceive it that way, so it's kind of puzzling as a political tactic. If this was a far-left initiative we'd all assume that the actual plan is to make the one-grand handout a repeating annual budget item....maybe that's MAGA's unstated assumption as well?
Isn't it already? This is the stimulus checks, just relabeled.
I can't tell if Trump is holding on to a 1950's conception of $1000 (in purchasing power) or if his team is really trying to capitalize on the innumeracy you mentioned. Being charitable, maybe one motivation is to try and get parents serious about contributing to the account themselves.
Similarly, Vance's idea of a 'baby bonus' @ $5000 is unlikely to bump fertility among women in the professional tier. The incentive would have to be comparable to the imposed opportunity cost, and (as an example) five grand is like a month-and-a-half's rent in NYC.
Much-larger, in terms of purchasing power, baby bonus incentives have been tried by various nations without moving fertility needles at all.
Babies cost way more than $5000 in both actual cost and in opportunity costs. I imagine there is a price which would raise the fertility rate, but it is so high that no nation is willing to pay it. And it is frustrating and insulting that governments persistently fail to recognise the actual struggles many young people face and think that a small "baby bonus" will overcome them.
I agree that we're dealing with willful ignorance about the ineffectiveness of baby bonus payments. But it seems that plumetting birth rates have little to do with young adults's material struggles, since they aren't worse today than at basically any point in time when people had more kids.
I live in Germany, and parental leave policies and daycare funding here are extremely generous. While they very much improve my quality of life as a parent, they don't seem to do anything for birth rates either.
That's actually surprisingly smart (the implementation details will matter).
So the entire solution we were failing to figure out for years was to call it "Unconditional Make America Great Again Income" and suddenly it's politically acceptable?
The next step should be "Free Make America Great Again Healthcare". ;)
My favorite thing is when conservatives instinctively reject a good left-wing idea because culture war demands it, then slowly and awkwardly re-invent that idea for themselves under a different name over the course of decades because it's actually a really important idea that you need to have in your arsenal to stay relevant.
People have been pretty uninterested in replacing Capitalism even when it was much worse for them. If you are pinning your hopes for overturning capitalism on peoples lives getting bad enough that they get angry and demand change, I think you are n for disappointment - it's unlikely people's lives will *ever* get worse than they were during the Industrial Revolution or the Great Depression. Technological process just makes it too easy to provide a standard a living above those times.
Meanwhile, I think UBI vastly undermines the *power* coalitions that keep capitalism in place, by weakening the reliance of workers on capitalists to survive.
A great way to encourage workers into a general strike is a guarantee that they will have sufficient income to feed their families during the strike, and even if they are fired and blackballed.
It's a lot easier to start organizing a union if you don't have to care as much about being fired for doing it.
Etc.
That's definitely outside my approach to the problem; I can't say it's wrong, but I'd need a ton of evidence and reassurance before trying a strategy like 'undermine my own nation in the hopes that it makes other nations do something better.'
I think that, to the extent I have leverage against reality, I have way more leverage to push my own country towards and outcome I want, than to push other countries towards that outcome by manipulating my own country in a secondary way. Even if my odds are very bad in either case, I'd rather stick to influencing my own country, at least until someone showed me a really foolproof plan that I believed in for the other method.
> Unemployment is really bad for people, to a degree that I can't overstress.
...yeah, because they don't have any income.
I dunno, it feels like you have a weird combination of disliking capitalism but buying into the rise & grind propaganda it uses to manipulate workers?
I think people can actually do pretty well if the have both leisure time *and* the resources to pursue a good life. I think being forced to do work you don't like and don't have an emotional stake in is actually *very bad* for people.
That's a big part of *why* I think capitalism can't be allowed to stick around forever, and why I think UBI would help undermine capitalist propaganda and reveal the damage it's causing.
(also, just to be clear - I'm talking about a UBI that lets you survive without major discomfort, not one that lets you bu everything you want and live in luxury. Anyone who wants additional things still has incentive to work, they just don't have to be deathly afraid of any small stretch of unemployment)
I'm not sure at what point in history conservatives were arguing for minimum wage hikes, and I was referring to ideas that one side is currently using and which the other side reinvents *at the same time*, not issues that switched from one side to the other.
But, sure, I'm sure this is a phenomenon that happens in both directions to some extent, and is funny in either direction. If anyone wants to educate me on examples going the other direction, I'd love to hear them.
The U MA GAI account will certainly be popular
Wouldn't it be wiser to wait for AI to produce the massive positive supply side shock before we start a UBI program? If we just start cutting checks and AI doesn't pan out, then it would just fuel inflation.
This is a hopeful sign for me as well.
I think this is okay. It may not be capitalism but it's a buffer against the kind of socialism that might actually be popular. Solve the cultural hegemony problem and be pragmatic about everything else. Actually existing capitalism would look very different without any free shit from the government.
Getting rage baited without even subscribed,
If it's open for EVERYONE, then it's open for EVERYONE. If it's not, then we think long and hard about what the restriction should be and what do we want to get out of it. I guess Phelps and Magnus only attend the open ones which everyone is allowed to attend. Not any restricted one unless there's a White-only Olympics or something.
Then again, those are not actually THAT open. There's lots of restrictions regarding drugs and external assistance. I've seen people lamenting for drug Olympics. But we do think long and hard about these restrictions and what do we want to get out of it.
Knowing Scott, I guess he already touched all of these and then some. There must be something about genetical engineering or AI cybernetics somewhere. Please tell me if he did. If he didn't, I'll be sorely disappointed .
>If it's not, then we think long and hard about what the restriction should be and what do we want to get out of it.
Except that we already did this.
The Olympic Committee set a standard of '2 year of HRT following gonadectomy and legal reclassification, with regular blood tests' 22 years ago. And that standard has been sufficient to result in zero trans women ever winning an Olympic medal in all of those 22 years.
Other professional leagues have their own similar restrictions that they've decided make sense for them, and as a result very few trans women actually win major professional competitions in reality.
This is a mostly solved problem as far as the people who actually deal with it professionally are concerned, and it's produced a stable regime with few to no actual problems.
It's the armchair culture warriors who don't know anything about the actual facts on the ground that are at a complete loss to imagine any workable standard, and insist on a blanket ban instead.
This sounds really interesting -- someone actually thought about the problem, before the culture wars got into high gear, and found a solution!
But when I websearch, I don't see this. Instead I see articles that say "olympic committee preparing to ban all trans women from women's sports".
Have you got a link for the two-years-of-HRT rule?
Here is the original statement from the Olympic Committee outlining their policy in 2003. AFAIK, it has been in effect ever since.
https://stillmed.olympic.org/Documents/Reports/EN/en_report_905.pdf
Thanks!
Of course I haven't heard about any of this so this is a new information that I'd have to research further. But my prior to this rule is effective and causally results in "very few trans women actually win major professional competitions in reality" is quite low because
1. 22 years is too long time ago. This is from era where doping run rampant and gay marriage haven't even been widely legalized yet. I doubt this rule would block modern shenanigans. Heck, Lance Armstrong still had good name then.
2. Even if the rule is intended for that, it doesn't seem to be so airtight since there still seem to be lots of controversies regarding the rules like that one wrestler from Tunis.
3. Even if the rule is intended for that, even if it's ineffective, the chance is, there will still be almost none of them winning an Olympic medal YET. This is because the practice has only really ramped up in the last 10 years and there's only 3 Olympics since. Maybe the number of people trying to exploit the rule's ineffectiveness haven't reached critical number yet. So, the lack of trans champion doesn't imply the rule's effectiveness.
This means there's still room for lots of thought.
I'm not sure who you're referring to by "that wrestler from Tunis" - Tunisia does not appear to have had any notable successes in women's wrestling at the Olympics. You might be thinking of Imane Khelif, who is a boxer from Algeria.
However, Imane Khelif is not a trans woman, she was accused of having an intersex condition. I.e., she was born female, raised female, did not medically transition, but might have been born with XY chromosomes or a condition that produces unusually high testosterone.
Which kind of shows the problem with demanding an "airtight" rule - if you define "woman" narrowly enough to definitely exclude all trans women, you will also end up excluding some cis women who just happen to have high testosterone or odd quirks of genetics.
Yeah, I think there have been a number of Olympic womens' champions in things who have some kind of intersex condition. Because biology is messy and even very rare conditions that give you a substantial advantage in some sport will get concentrated in the very top performers of that sport.
I think we shouldn't be using the word “cis women”. It is considered offensive in the UK, confusing and sounds very woke.
Just use “sex at birth” if you need to specify, and then go by women vs. trans women.
> I doubt this rule would block modern shenanigans.
I mean, trans women didn't win any Olympic medals in the most recent Olympic games, either. How 'modern' are we talking, here?
>there still seem to be lots of controversies regarding the rules
Sure, rule always have controversies and need to be reversed - See eg. Lance Armstrong's entire career, or any other sport dealing with perpetual advances in steroids/doping/match trading/etc. etc. etc.
A single doping scandal isn't sufficient reason to say that all athletes with blood should be barred form competition... we always have controversies, and we always revise the rules to address them. This isn't unusual or unusually dangerous.
>This is because the practice has only really ramped up in the last 10 years and there's only 3 Olympics since.
22 years is enough time for a person to be born, trained from birth with the intention to win the Olympics as a trans woman, undergo the precise minimal medical interventions needed to qualify, and win every medal in every division at the most recent Olympics, if such an advantage existed.
See Scott's section on 'solving for the equilibrium'. If it were really this easy for any country in the world to win every women's Olympic medal, is it plausible that none have tried it all in these 22 years? But that one surely *will* try it, soon, very soon, not like the last 5 times we said it was coming soon, this time we really mean it, just you wait and see...
Sorry, but before sanctioning swift, decisive, and overwhelming responses to a problem in the form of collective punishment and civil restrictions on entire demographics, I do kind of want to wait to see *any empirical evidence at all* that the problem *actually exists*.
I understand that lots of people think that this problem is *just about to start existing, every day now*... but if we routinely implemented collective punishments based on that type of speculation, the world would be an even sadder and poorer place than it already is.
Pro cycling has a long and storied history of being chemically enhanced, and now there are rumors of electric motors for the hill climbs.
A cynical sort might be tempted to observe that Armstrong doesn't seem to have been any more enhanced than any other top-tier cyclist of his era, observe that USADA doesn't really have any jurisdiction in France, see "US", but Armstrong had committed the Unpardonable Sin of eclipsing Eddy Merckx's record 5 wins, at least in the eyes of the French.
Somebody call 911. I've been severely wounded in a drive-by wimbling.
Which one? I actually heard something similar but cursory search doesn't yield me anything. I found some interesting 800m and 1500m records but those are the ones that use wheels instead of prosthetic legs like I imagined. But this is reinforcing the point, why don't we have open wheelchair racing already??
While we're talking about Paralympics, I'd respect it more if they compete in something open, that their disability doesn't impede (or even give advantage!). Like if a blind people become chess champion or something. Or a deaf one winning the shooting Olympics. There are already people like that actually, and they're very cool https://en.wikipedia.org/wiki/List_of_athletes_who_have_competed_in_the_Paralympics_and_Olympics
There are some wheelchair races which are not restricted to people with a disability; I checked my local wheelchair marathon and there are explicitly no specific disability criteria for entry.
I doubt there's too many able-bodied entrants though, it would take a special-kind of bloody-mindedness to spend your time training with a wheelchair you don't need just to show up those darn cripples. And besides, the disabled have a weight advantage, they're not carting around so much leg muscle.
> why don't we have open wheelchair racing already??
Because we already have NASCAR.
If you want open wheel racing, you’ll need to watch F1 or IndyCar, not NASCAR.
The blade ones like Pistorius had. This Wikipedia article might have some useful links: https://en.wikipedia.org/wiki/Mechanics_of_Oscar_Pistorius%27s_running_blades
I guess that should be the one famous enough to reach my ear long time ago. But Paralympics records of T44 100m, 200m, and 400m I found still didn't defeat world records.
RE: beowulf888 and Eremolalos on Aphantasia and Mental Images
See the conversation here:
https://www.astralcodexten.com/p/open-thread-406/comment/173154666
and here:
https://www.astralcodexten.com/p/open-thread-407/comment/175847543
My idea of what "mental images" are, is kind of similar to a photograph. But where a camera is only concerned with capturing light (as far as I'm aware), the human mind is much more complex and could be capturing any number of things, and it is definitely not capturing close to everything that there is about something. When you see something, what you commit to memory could be color, shape, size, or something conceptual instead of physical, like if the thing you saw was alive or not. I'm sure that it's usually a combination of many of these things.
The topic seems very complex to me, because when committing something to memory, you also have everything else you know present in your brain as context for your thinking patterns. So for the Statue of Liberty, I stored in my memory the fact that she's wearing some kind of robe as a concept, but not as a physical picture. I can do this because I already have knowledge about robes (or togas or whatever) and how they generally exist. And because, to me, the robe she's wearing is a pretty minor detail, instead of using up brain power and memory remembering finer physical details about it, I just put down "robe" in my thoughts. So when I bring it up as a "picture" in my mind, I'm not seeing what the Statue of Liberty physically looks like neck down, but I'm still picturing something that is representative of what I know about the landmark.
When it comes to her face and crown, however, my mental picture is a lot more detailed. In the Statue of Liberty test, I didn't actually remember the fact that there were 7 spikes on her crown as a concept, but my knowledge of the shape of the statue forms a picture in my mind that is clear enough that I could count the spikes and come up with the correct answer.
So for my mental image of the statue, some portions I have in my memory as physical attributes, and others I have as ideas. In fact, all of my thinking at all is like this. Some thoughts are represented physically, some conceptually. I do most of my thinking using some kind of mental analog of the physical senses, with hearing and seeing being the most prominent ones (I involuntarily hear words as I read or think them, and I create images in my head when words I read or hear describe something). Less often, I sometimes think using emotional senses (or something. idk what to call them), where I'm just feeling different things in my head without some kind of words or sounds or any bodily sensations involved, and sometimes that is the kind of thing I internalize as an isomorph to some visual data.
Anyway, my theory of "mental pictures" is that the whole of any imagined thing is like this. Which sounds like it's kind of like what Eremolalos is saying, if I'm understanding correctly (as the Statue of Liberty test shows, not really anybody remembers and can visualize complete exact details of a thing they've seen). But to beowulf's point, I think it could be very possible for one person to grow and develop a pattern of internalizing things "visually" more than through any other sense, and for another person to internalize things in every other way than visual. There's just so many variables and internal mental contexts that it's hard for me to believe that any two people are very similar inside their heads, even if they can both start with some stimulus and arrive at the same thought.
Hopefully this contributes in some way to what you both were discussing. It definitely is very interesting for me to think about.
I'm very interested in this stuff too, but am just emerging from a few days of being sick and I am now alarmingly behind on practical matters. I will try to respond soon, but wanted to recommend that you guys read the part of my review summarizing Schweitzgebel's argument. The demonstration he gives (and that I amplify with the Statue of Liberty examaple that our experience of mental images not as vivid and accurate as we feel it to be -- well, he gives demonstrations of the same sort for many other modalities of experience in his book. My review summarizes the argument for I think 4 other modalities besides mental images. So I'm not trying to get you to read my deathless prose, but just see the full version of Schwitzgebel's argument, which I find pretty persuasive, and which I think even those who don't will find quite interesting. And while I think I'll have time to get back to this discussion, I'm definitely not going to have time to summarize my summary of Schweitgebel.
The review is here: https://bookreviewgroup.substack.com/p/review-of-perplexities-of-consciousness
You can skip the opening 2 paragraphs, which most people hate anyway, and start with the section called *Naive Introspectionism.* which is I think about 10 pp long, and then if you like read the quite brief follow-up section called *Dismantling the Naive Introspectionist Model.." The conclusion you can skip. And the Schwitzgebel stuff I summarize is not dry and tedious. It's like the Statue of Liberty stuff. It includes the results of actual experiments testing the accuracy of people's idea of what their thoughts, dreams, etc. are like.
I started your review from the beginning and it's interesting to me so far, so I'm just gonna read the whole thing. I've already got some questions, so in the coming days (or probably weeks) I'll compile my notes and questions into a comment and just post it on your review.
Hopefully getting caught up on work went well!
I believe the key takeaway here is that we likely have very different levels of visualization ability (similar to the way people have different levels of internal narrative). I disagree with Eric Schwitzgebel's reductionist view that we're just imagining that we're seeing an image in our mind, but it contains little real information about the original image (apologies to @Eremolalos if I'm distorting what he said). And if I'm not misunderstanding him, Schwitzgebel claims that we all function at an identical level of inability to visualize specifics.
But I admit that the Schwitzgebel thesis has forced me to reevaluate how I "see" my visualizations, and more profoundly for me, it's caused me to question how I see the external world. For instance, I'm wearing a shirt with a complex plaid pattern. If I look at it, I get a gestalt of the pattern, but when I focus on a particular portion of the pattern, I lose the details outside that area. My gestaltian summation doesn't really handle the details outside the area of my focal attention. I have to scan continuously across the fabric to maintain that gestaltian illusion, and I lose it when I turn my attention to one spot on the fabric. I'm forced to conclude that most of my vision is my mind imagining it has captured the details, and at any moment, the only "real" perception is at the center of my focal attention.
And, yes, my visualizations are mostly constructed from same sort of imagined queues, from which I create an imagined gestalt of the image. And Schwitzgebel is correct, that my imagined gestalt of an image actually contains much less information than when I look at the image. Where I would differ with Schwitzgebel is that some people's gestalts retain more information than others. More significantly, my visualizations lack that focal point component I have with vision, so I'm not continually observing and re-observing the details as when a view an image.
And another reason I doubt Schwitzgebel's claim that everyone functions at the same level of crude visualization is that people report very different levels of detail in their dreams. I happen to be a fairly lucid dreamer. Schwitzgebel might claim that I'm either imagining that I'm "seeing" details in my dreams or that I'm misremembering the level of detail when I wake. This past evening, I dreamt that I visited a family on a farm next to the Connecticut River, down along where the tidal runs push upriver from Long Island Sound. In fact, we were talking about the possibility of a really big tidal surge smashing into their farmhouse, which wasn't far above the waterline. I mentioned how much I liked their place. I was an old colonial that had been added on to over a couple of centuries. The wood was very weathered, I could focus on the details of the weathering. At some point most of the paint had been scraped off, but it gone few years without repainting.
Upon waking, what's interesting to me is that the focal-point scanning that I use when I'm awake with my eyes open seemed to be functioning in my dream. I could scan the paint, and see the details of the weathering on the wood grain. The area outside my focal point was blurry. Or I could stand back from the farmhouse to get view the layout of the doors and windows, but the additions outside my focal point were blurry. This was very different from my waking visualizations.
I told them that I'd like to visit them again, but I woke up. I've found that I can re-enter sleep quickly I can go back into my the dream that I was having. They were walking over to a barn, when I caught up with them again. I asked them for their phone number and they wrote it on a piece of paper for me. Of course, I have a shitty memory for numbers, but upon waking, I remember it started with a 703 area code. That puzzled me because 703 is not a CT area code. I looked it up, and 703 is a northern Virginia area code. But then I realized that my visual acuity may in dreams may not be perfect, and that I may have misread the 7 for a 2. And 203 is a CT area code. :-)
It sounded to me like Schwitzgebel granted that some people with practice can "visualize" more specific details than others, but I also disagree with his main thesis that you state in your first paragraph.
I don't doubt that any mental image I have is far from being a photograph of anything I've seen, but it is still very much a visual image of something using (among other things) visual data that I have physically experienced. It's hard for me to believe that anyone born blind would also have "visual" thoughts, and if they don't because they cannot see, then I think it would be pretty certain that those kinds of thoughts *are* visual.
There's more I could comment on in the review, but at this point I've had a few weeks to sit on these thoughts and I've come to some vague conclusion in my mind that will continue to be built on unless something comes and shatters it. But I don't think anything discussed here or in Eremolalos' review thus far could do that kind of shattering.
I think I have more vivid mental imagery than most people.
And my own mental imagery feels like something reconstructed from a compressed description. So I will “see” the Statue of Liberty, but what I see will probably be wrong in minor details that I didn’t remember correctly. My visual memory is clearly not a bitmap image.
What about the case where their response is "noooo, I don't even want to do any good, I just want to turn people into dinosaurs"?
'We'll gladly connect you with a list of people who say they want to be turned into dinosaurs and will even pay you for the privilege, in exchange for letting our scientists observe your work and try to reverse-engineer the techniques towards a cure for cancer.'
Before getting too preoccupied with whether you could, consider stopping to ask whether you should.
I think the other way around makes more sense.
Life finds a way.
Presumably they should play Look Outside. At least one person gets turned into a dinosaur in that. Also one gets turned into a wall.
🦖
You know that frustrating feeling you have in the morning, when you need to get out of bed NOW, but part of you wants to roll over and sleep for another hour, consequences be damned? The intensity of this feeling is influenced by innate attributes you are born with.
You have no way of knowing if I have the same feeling in the morning. Even if our lifestyles were exactly the same, the exact nature and intensity of that feeling would be affected by physiology (body and mind). Even with the same amount of sleep and the same quality of sleep, that frustrating feeling (or physiological resistance) might be twice as great for me. Or only half as great. Our science is still too primitive to measure these factors, except in extreme edge cases.
Bill rises at 6 each day and runs 5 miles. Bob straggles out bed at 8 and runs 1.5 miles. It is possible that rising early and running subjectively feels more uncomfortable for Bob, for inherent, unchangeable reasons. Which would make Bob more self-disciplined than Bill could ever be.
I don't find that feeling frustrating, I just linger in bed. I know damn well that my energy level is low in the mornings so I don't do anything hard then, that's for the afternoon and early evening.
You know some people respond to childhood trauma by being a perfectionist, type A, always on the go. And the rest of us respond to childhood trauma by being lazy unmotivated excuses. And there's the self-help industry telling us that we can choose to be type A if we really, really want to. And there's a recent article in the Guardian about a man who was a type A boss, lost his motivation and became a type B slob. Lo and behold, he had had tiny strokes in his brain which precipitated his demise. Now with the help of legal meth, he is back to being his type A self.
Well, if you figure out the brain mysteries, let me know. I am most definitely type B, and I have learned to work with my nature, not work against it. I am still learning to release the guilt and shame over my brain's disposition.
There's something to be said of choosing to be type A if you really want to. There are many tales of ne'er-do-wells changing their ways when a baby comes along. So maybe that's what you need, a baby to make you wake up at 6am. The natalists were right all along, we all need our personal productivity baby.
A successful friend told me that having kids was the ultimate motivator for him.
I have an alarm set for two hours before I have to get up, just so I can choose to sleep through it.
As someone who did that sort of thing once, I suggest that training your body to sleep through alarms has potential for negative consequences that may be worth considering.
Also, for anyone who needs to know, this exists: https://www.amazon.co.uk/dp/B004MSMUGI
This is partially in response to me already sleeping through my alarms. I was sleeping through them long before I could afford to.
I'me sure there are some proxy markers and imaging methods for sleep inertia
https://en.wikipedia.org/wiki/Sleep_inertia
doesn't mean they are available.
And yes people are not living with the same difficulty level and in some ways it might be asymptomatic if compensated.
Mankinds misery is purely artificial and can be trivially alleviated via an orexin agonist
Brain imaging is great but it was my understanding that it has done very little to solve the mysteries of qualia.
potentially interesting https://www.nature.com/articles/s41586-025-09796-0
Some of you will feel this is obvious. Others will feel it is true but useless trivia. For me, this idea profoundly alters how I feel about humanity, both on an individual level and as a whole.
This is why I will never feel comfortable around committed ideologues, of any variety. No matter how much I agree with them on individual policies, ideologues can never take this forgiving, permissive view of the human character. And so I will never be comfortable rubbing shoulders with them, or supporting them. I'll always have that fear that by empowering them, I am empowering an unspoken ideology of judgment that could one day devour the world.
For those that may have read Scott's post on Janus's simulators or are otherwise interested in niche LLM phenomology and Cyborgism, the first full length public podcast with Janus has just been released.
Janus appears in full snowcrash-style avatar, filmed live in Portugal this year.
YouTube:
https://youtu.be/e0QWr_z3re0
Twitter: https://x.com/thegoodtimeline/status/1995557717843316795
"Janus becomes a VTuber" was certainly not on my bingo card.
I keep expecting National Shrimpact Day to be "come to our fundraiser, attraction: all the shrimp you can eat banquet for those who buy the $100 ticket!" 😁
Is a "Shrimpact" when I drop the Aldi shrimp tray in my freezer and it shatters?
I wouldn't dismiss out of hand them actually doing something like that, with the shrimp being "humanely killed." I suspect you're reüsing heuristics you've developed about vegans, and I'm not sure those are really applicable to these guys.
Humane killing is good, but as important is whether the shrimp had good lives. Were they able to express species-specific behavior and preferences?
You know, this seems like it could become a fabulous grifting opportunity, though maybe it would have been even better five years ago.
What *are* the preferences of creatures with brains the size of something very tiny? Well, the motivated interest groups claim they have feelings just like us:
https://mercyforanimals.org/blog/do-shrimp-have-feelings/
Clearly, this is something that requires study. Lots and lots of study! Lots and lots of very expensive study with good salaries and benefits and opportunity to jet-set to global conferences on "is pink for shrimp or not? the species-specific colour preferences over thirty shrimp species" in agreeable locations with plenty of free booze and vegan food/all the shrimp you can eat banquets.
After all, how can we alleviate shrimp suffering and make sure shrimp have meaningful, worthwhile lives unless we are fully informed on the fears, dreams, hopes and desires of each liddle shrimpie? And how will we know the preferences of a shrimp unless we put in the work? And how can we put in the work unless the government puts its hand in its pocket and funds us?
Now that idpol is fading away and even Ibrahim X. Kendi got turfed out of his comfy academic sinecure, shrimp behavioural research may be the new woke!
I wasn't thinking of vegans as such, and if they *did* have the shrimp buffet of humanely killed happy shrimpies, my opinion of them would go up a lot 😊
On the topic of infighting between such groups, icymi, you might be amused by the fight over catering at the COP30 conference last month between those who wanted it to be entirely vegan and the indigenous communities who wanted to highlight their ancestral food.
Wait, there was a second fight over vegan catering? Yes, such things only happen once in a decade, but two is still a lot.
Great Moments in Catering: Friend passed away, tragically young. Jewish funeral. We were seated with the rabbi, who looked at the catering and ruled that the provided ham was indeed Kosher. . .
Ecumenism is a wonderful thing 😊
For the American's here:
How often would you have to be ill after eating at a restaurant or from a takeout place before you stopped eating there?
How often would you say you get food poisoning?
Asking out of curiosity based on an interaction I had recently.
Once, if I were indifferent between that place and other similar places; twice, if I liked it in particular for some reason.
My wife, meanwhile, won't eat at a place ever again if a friends says they got sick after eating there one time, and this has led to some small amount of strife over our doordash orders.
I'm an American. I've gotten food poisoning in Japan 3 times. Once from sushi, once from sashimi, once from food at a Sichuan restaurant. I've been back to 2 of 3.
The only time I got any kind of food poisoning in japan was on my very first full day after I moved to tokyo. Got a 7/11 microwavable yakisoba. It passed through me at a speed that should be studied.
I got food poisoning from Jack in the Box once over 30 years ago and haven't gone back since.
Averaging out all my cases of food poisoning across my life, I think it's slightly under once a decade (three times/about 40 years)
one
less than once a decade
1) Once, assuming I was sure of the causal connection.
2) Hmm....3 or 4 times I think, or roughly once a decade. Of which I think two were from restaurants or takeout food.
1) If I got ill after eating at a restaurant, I would never eat there again.
2) I have never had food poisoning.
Once, though it's not necessarily a rational choice. There's just so many options in most American cities, of roughly equal quality. The extra, literal, ugh factor of remembering the illness would weigh on the decisions, like it or not.
I've felt ill twice in memory, at least seriously enough to notice.
I once got food poisoning that started as mild nausea immediately before entering a restaurant, and progressed to severe food poisoning as I ate. Rationally, I know that it would take a wild coincidence for the restaurant to have been responsible, but my body still got physically ill on setting foot in there for over a year- on subsequent visits I would eat a bit of bread to be social. (The restaurant was a retirement home cafeteria that I visit regularly, so not ever eating there isn't really an option, and so I got to see how long this took to wear off. Otherwise it would have been an easy never again)
Depends on how many times I've gone there. If I go once a week, and I get food poisoning after a couple years, I might give it another shot, or even two shots. If I get food poisoning on the first trip there, I'll obviously never go back.
Once.
(With the mild caveat that if I have eaten at a place in restaurant for a long time and get food poisoning from the takeout, I am more likely to stop getting takeout than stop going to the restaurant, and the strong caveat that it has to be food poisoning - if I come down with a cold that’s not the restaurant’s fault.)
I have gotten food poisoning a single-digit number of times in my adult life, probably no more than five. I don’t remember ever getting it as a kid, but could be missing (or not have recognized) an incident or two. I think I do more of my own cooking than most people, but still eat out maybe an average of once a week, so it’s not that the situation could never come up.
"How often would you have to be ill after eating at a restaurant or from a takeout place before you stopped eating there?"
Once
Based on experience, once. But that's because the place already gave mildly sketchy food vibes and the food poisoning confirmed it.
I think I'd have to get food poisoning at least twice, possibly more before I actually noticed a pattern. (I don't think I've ever actually gotten food poisoning from a restaurant.)
Like, food poisoning takes time to develop, so I'm probably only going to know vaguely that it came from something I ate yesterday - and my first thought will probably be "I guess those leftovers were a bit too old" rather than "I bet the chipotle wasn't following food safety standards."
Yeah, I think it's pretty hard to know for sure where food poisoning came from unless you have, say, a big bunch of people who all had a single meal together and then all got sick about X hours later.
And yet, I got sick from (as best I could tell) an Indian restaurant once, and I have never been back. There's a lot of competition around and the chance of getting sick again isn't worth it.
It was the Chipotle!!! 😤
Pedantic dickery answer; zero. I've stopped eating at most restaurants for other reasons.
Specifically stopping because of food poisoning; more than once, but probably twice. I stopped eating Jack In The Box for many years after their big contamination event, but went back eventually. But twice means I'm going to consider food poisoning a routine possibility, which means I'm eating elsewhere.
It also depends what you mean by food poisoning. I've continued to eat things that give me butt problems because I don't value those enough to give up the other end.
I get food poisoning on occasion, because I cook my own food and am bad at cooking, and/or keep leftovers around too long.
To clarify, when I say food poisoning I mean any combination of diarrhea, stomach pain, cramps, nausea and vomitting
I don't value diarrhea particularly high, I'll put up with a lot more of that than vomiting or migraines. Typically I don't count something as food poisoning unless I'm stuck in bed because of it.
I can only think of one time I've had food poisoning from a restaurant, but I don't remember which restaurant it was. I think it was Chipotle. QDoba>>>>Chipotle, so I already don't ever choose to eat there anyway. Otherwise, it would probably have to be two or three times in a row for me to stop going to a place for awhile, if it's somewhere I've already been to multiple times. If it was my first time at a place and I got food poisoning, it might be hard to want to go back right away.
Are there any Thomas Sowell fans here? I just finished "A Conflict of Visions" And I found the book to be a deep insight into our differences. I tried to summarize the book, but AI does better, so here's the prompt. * Thomas Sowell "A Conflict of Visions", summarize. *
What Sowell book should I read next? (I've started "Basic Economics" and "The Vision of the Anointed", the second one seems derivative of what I've read... and is being cranky about the 'institutional class' )
(I know two posts in a row, I'm off of work and awaiting hernia (groin) surgery... so you'll have to put up with me for a little bit. :^)
I very much recommend:
a. Read Hayek's essay "The Use of Knowledge in Society"
b. Read Sowell's book _Knowledge and Decisions_
These pair well together. The book is definitely worth reading!
Also, I think his Culture trilogy (Race and Culture, Migration and Culture, Conquest and Culture) were very good.
I just want to say that Basic Economics truly changed my life. It had an absolutely profound impact on me.
Thanks, Yeah I've downloaded that onto my kindle too, I love the kindle sample function.
Book review of Sowell's Discrimination and Disparities: https://thegreymatter.substack.com/p/book-review-discrimination-and-disparities.
Some discussion of the review: https://www.astralcodexten.com/p/open-thread-334/comment/59296100.
Thanks I'll check that out too. (I've started reading "Black Rednecks and White Liberals".)
Black Rednecks and White Liberals is worth reading. If I subtract for some repetition, partisanship, scientifically uncharitable views, and the common errors from sweeping pronouncements, Sowell is still on balance an intellectual treasure. He will help you think more broadly across global and historical dimensions before making politically expedient claims, and that is coming from someone raised as Progressive.
A big thing to notice about Sowell is that he does[1] his best work in long-form, slow writing. He wrote some newspaper columns for many years, and in general they were nothing special. I've seen him interviewed a few times, and he mostly didn't come off all that well.
[1] Did--I think he's in his late 80s now and probably isn't writing a lot more books.
OK Thx, I'll give it a look see.
I own _Basic Economics_ and _Advanced Economics_. Sheepishly, I confess I bought BE because at the time, I was literally looking for an introduction to economics (David Friedman's book _Price Theory_ is better for this), saw the good reviews, and found myself in what I would have to call partisan, but nevertheless very good, writing - enough to convince me to buy AE.
I find Sowell does eventually repeat himself - IIRC, both books contain a lengthy discussion of rent control - so I figure the more books of his I buy, the more chapters I'm likely to find saying the same thing in one of the other books. That said, I might benefit from getting ACoV.
Meanwhile, you might benefit just from finding videos of his interviews and other appearances online.
Yeah it's OK. I would like some equations though.
Where do you think the MAGA movement fits into the two visions that Sowell articulates?
No group of people is all one type of the other. And there will be parts of both visions in many ideas. But Maga is clearly closer to the constrained view. (vs. unconstrained) with it's call to traditional beliefs and mores.
MAGA isn't an intellectually coherent movement, it's mostly just a cult of personality that's energized by a hatred of progressive overreach.
Yeah, I agree, but it does seem to me that another factor is the capacity of MAGA folk to believe exceptionally stoopit bullshit -- immigrants are eating our kitties in the back yard, etc., things for which a thought experiment and consideration of what one has observed so far should be enough to make implausible. The Woke Folk also have ridiculous ideas -- a kiss without permission is a little rape -- but the reason wokies believe them isn't that they are too dumb and uneducated to realize they're nonsense, it's a kind of hysterical, semi-religious scrupulosity.
I think the common point in MAGA and wokism is that both have evolved in the world of an online media environment that makes it easy to police the expressed views of people in your own coalition. And that makes it easy to have silly or crazy beliefs be part of your doctrine, because you've got great machinery to silence dissent just lying around waiting to be used.
This also makes the beliefs rather unstable--whatever the current crazy bit of doctrine is, it is like 10% sincerely held and 90% expressed to keep the peace or because the speaker has never really heard any good arguments against it. That can flip very quickly when the preference falsification cascade collapses.
Wouldn't you agree that MAGA beliefs are at least directionally correct (well, with the exception of tariffs) even if their reasoning is terrible? Unchecked immigration - particularly of the Biden-era kind - is objectively bad, so it's at least a reasonable impulse to be opposed to it. Woke beliefs, on the other hand, don't even have a seed of reality in them. They're little more than a fig leaf for their dislike of traditional American culture. In my view they actually care little for the welfare of blacks or transies, they just want an excuse to paint America as an oppressive state in need of reform because that gives them power. It's the same dynamic as the Cultural Revolution.
Well, I don't think MAGA beliefs about vaccines are directionally correct. As for Woke beliefs, yeah they're a bid for power, but that's not all. I've spoken with a lot of woke students at the local Ivy, and they sound a lot like people who have a case of religious scrupulosity (a form of OCD I have actually treated.). Most are quite disturbed by some of the non-woke feelings they can't keep from being aware of, eg fear and distaste for certain minorities, horny politically incorrect sexual fantasies. Also most are quite scared of being canceled by their peer group because their politically incorrect side is revealed.
Oh agreed, vaccines are another one. Though I'd say that it comes from a place of libertarian freedom and not wanting the government to have the ability to dictate anyone's medical choices, which fair enough. I disagree with the position but at least understand the impulse. It's kind of the right wing version of "my body my choice".
Very interesting to hear about woke anxiety. How do you council those people? Are you able to get them to see the unreasonably puritanical nature of woke?
I don't understand what you were getting at. My point was that in the absence of more info some consideration of what one has seen in the neighborhood, plus some thought experiments would have made it seem implausible to most poeple that immigrants were eating the neighborhood's pet cats. (Is there even much edible flesh on a cat? Has anyone found fur and bones? Have a lot of cats gone missing? Have I seen skinny, ravenous foreign-looking people wandering through my neightborhood?) I don't know what happened at the US embassy in Haiti,but don't see how giving food to Haitian kids is a MAGA bullshit rumor the way they're-eating-our-cats-is.
Maybe someone just heard about "eating pussy" and misunderstood it?
Right, for a book that attempts to explain the "ideological origins of political struggles", I'm not sure that it says much for our time where a reactionary punish-your-enemies / reward-your-friends populism is the dominant political force in the country. I read it years ago and recall thinking that the visions that Sowell identifies are relatively small threads in contemporary politics and certainly not clearly aligned with any political movements or parties in the Western world.
Yes it's not great but this is the future that got locked in when the Left went all-in on identity politics. MAGA is a rather predictable extension of that mentality to the Right. Sowell came of age when there was still a sense of common American identity and it was possible to have higher-minded ideology than simplistic ingroup/outgroup dynamics. Sadly that's not possible anymore. The steamroller of identitarianism has flattened all of it.
The probable reason people don't directly address that argument about transgender sports: it is such a tendentious argument that it feels like a rhetorical slap in the face. I'm not exaggerating when I say that every time I hear this argument, it feels like an insult to my intelligence. I know it's not intended that way, but that's what it feels like.
For a comparison, few Darwinians are willing to argue the fine points of creationism. Perhaps there's a sense that if you are willing to take creationism that seriously, you've already given the creationists a kind of victory.
In my view this is an immoral attitude to take, for it cedes the intellectual high ground to social fashion. I feel that the right to have a firm opinion on anything comes with a responsibility to vigorously defend it in good-faith debate. The attitude you've expressed encapsulates the clever sleight-of-hand the Left has played on the culture over the past few decades: they hold their views to be self-evident and, because they occupy the taste-making institutions, they've been able to successfully taboo any rejection of them as low-status. That's led to their current emperors-new-clothes position w/r/t gender and race and our inability to have reasonable public debate over contentious issues.
I love nothing better than sparring with Creationists. If someone doesn't understand science well enough to do that then IMO they don't have the right to consider themselves educated.
You have to triage which arguments you were willing to engage with. Do you really want to argue with people who believe chattel slavery should be restored? This is not a straw man, I have met people in real life who believe this.
Do you really want to spend your time debating with people who consider Stalin to be the greatest European leader of all time? Someone like that has such insanely different priors that arguing with them would surely to be an uphill battle. It might be pointless.
Do you want to spend your time arguing with someone who believes that life is meaningless and all pregnancies should be aborted? Because these people are among us as well. Please consider the implications of your commitment to debate.
Sure, those are all reasonable pragmatic considerations - we can't spend our lives doing nothing but debate. But it's important to remember that even those reasonable considerations can be applied in bad faith. It's easy to convince yourself that something isn't worth your time when the reality is that you're not sure you can win the argument. I think the right way to handle is that the degree to which a belief is important to you has to be earned by your willingness to defend it. That stops the whole "I don't care enough about my belief to defend it but I care enough to shun you for disbelieving it." That's the sort of behavior that I think is bad epistemic hygiene.
Addressing this argument in good faith, trans women shouldn't be in women's sports for the same reason able-bodied people shouldn't compete in events for disabled people. This biologically disadvantaged group was intentionally given a venue.
I think that in cases where trans women have a clear advantage. But when their hormones take them to the same level as cis women, does it really matter that their disadvantage is due to hormones they took instead of ones they produced naturally?
My personal opinion is that sports are inherently unfair and only really useful as entertainment, so we should stop worrying about whether it's more unfair to make cis women compete against trans women or bar trans women entirely and just let people make the leagues viewers want to watch.
"Same level as cis women" has been too easy to game.
I'm not personally a sports guy, but I can think of at least 2 reasons to want non-competitive players to be interested:
- It's great exercise, and games are an easy way to get children specifically to be active
- It's a great social activity (both for children and adults), and we don't have many easy means of developing social groups nowadays
Having a group of people with similar characteristics to yourself as visible sports icons is motivating, so you might want to artificially keep some around if you like the above mentioned benefits.
If you want to say "sports should just work like the rest of the entertainment economy", that's a much bigger position than "trans women should be allowed in women's sports". I think that would also imply:
- Colleges should stop offering women's sports if they don't "earn their keep" (which they are currently required to do under title IX)
- Colleges shouldn't offer most sports (almost all NCAA sports except football and men’s basketball lose money)
- You should cancel the Paralympics (they usually don't make money)
- You should cancel the Olympics (usually loses money)
If you agree that there's a social utility in subsidizing / mandating certain types of sports, then you need to think about how to maximize that (maybe most people should be able to fit into a group with a "professional" team to be role models?).
If there's no social utility, then letting pure market forces operate makes sense IMO
I don't think there's social utility in subsidizing sports. If you want to do the Olympics, you can. But you shouldn't make other people pay for it. Why do people take a sports league so seriously?
I definitely think there is net social utility in encouraging physical activity and socializing with strangers. There may be better ways to do that, but off the top of my head I can't think of many better than organized physical play.
In middle and high school I got most of my exercise and met many of my friends through sports teams.
Also, as an adult, a ton of the social organizations I know about are amateur soccer groups and things like that.
"just let people make the leagues viewers want to watch"
In practice, this position constitutes a full-scale opposition to women's sports.
That makes it sound like I don't want women's sports specifically to exist. I just don't think people should be forced to pay for sports. If they want to watch it, then sure, they can pay for it, but why force other people to subsidize sports they don't want to watch?
I'm not saying you thought it through and decided this was the best way to oppose women doing sports without social censure, I'm just saying that taking this attitude is tantamount to wishing away women's sports in practice, because nobody (including women) likes to watch any woman's sport that isn't just veiled softcore porn. Whether you're fine with it or not, you should be aware that the practical consequence of your position is to baleet women's sports.
It's hard for me to overstate how frustrated I am about the shrimp welfare movement.
Animal advocacy is a zero-sum game, and as long as there are mammals out there suffering, the shrimp welfare movement seems like a slap in the face. I hate it so much. This is not harm reduction, this is contrarianism at its worse. Some might even call it perversity.
Is there a finite amount of animal welfare concern out there? I kind of doubt there’s a lot of overlap between shrimpophiles and other activists.
One up them by starting an Ant or Mosquito Welfare movement.
Ironically, I go out of my way to avoid killing ants when I am outdoors. I (slightly) resent it when people kill insects in the great outdoors without an articulable, defensible reason.
But I don't see this attitude as an appropriate basis for a political or philanthropic project.
When you get to microinvertebrates, I'm in. Save the rotifers!
I can think of a number of ways to prioritise animal welfare causes.
1. Start with the most intelligent animals (whose suffering is presumably worth more) and work your way down, so your first concern (apart from humans) should be ensuring the humane treatment of apes.
2. Start with the most numerous badly-treated animals, on the assumption that a million suffering invertebrates add up to one suffering chimpanzee. This is where the shrimp people are coming from.
3. Start with the most egregious abuses and work your way down. If you think about the most egregious and needless things that people do to animals, then "crushing its eyes to make its ovaries develop because the overcrowded and unsanitary conditions that you keep it in prevents normal ovary development" is pretty high on the list. (Note: if you're about to reply with a list of things that are even worse then I don't want to hear them.) So the shrimp people kinda have a point here as well.
Or you can prioritise along all these axes simultaneously, which will probably give you a fairly reasonable set of priorities.
Where the shrimp welfare people are really going wrong is in trying to sell this as a pure utilitarianism play. Most people aren't pure utilitarians and react badly to people trying to sell them something based on pure utilitarianism; if you want to convince people to care about the treatment of farmed shrimp then you're better off using the sort of combined vibes-based ethics that people tend to work with in practice.
I completely disagree with the statement that animal advocacy is a zero-sum game. It's additive. Welfare concerns for one animal group often benefit another animal group.
I used to think or hope that the fact that people cared about dogs, to such an extreme degree, would transmogrify into caring about wildlife generally. But that turned out to be fanciful.
We teamed up with the dogs to exert our joint dominion over all the other wildlife on the planet. We have a different arrangement with cats and horses. Nothing else is in remotely the same category.
We may still wind up with some level of concern for various other critters, and that speaks well of us I think, but it's not going to come from the same place as Man's traditional bond with his Best Friend.
Hmm. I'm open to the idea. Could you explain the mechanism a little more?
Making a law "you should not torture animal species X" makes it more acceptable to propose a law "you should not torture animal species Y" later.
This is valid! Good point.
It is a very useful reductio. The argument is pretty sound from its own internal values. And in fact it doesn't even stop at the shrimp -- many of the shrimp people earnestly believe the best thing to reduce suffering in the world is to turn natural land into parking lots, and especially to prevent the creation of more parks and natural preserves. Why? To prevent insect suffering!
How is this different from saying that as long as there are humans suffering, animal welfare seems like a slap in the face?
I feel the same way about animal welfare. When people spend thousands of dollars on veterinary bills, I see this as anti-human and a demonstration of great callousness towards all of the children in the world who are suffering.
UNLESS the animals that need care are vital to your livelihood, of course. And in some cases, I suppose that putting down a sick animal might cause the owner to spiral into suicidal depression, which must be avoided.
But in ordinary circumstances, very sick animals should be allowed to die and the money saved on veterinary bills should be spent in a more responsible fashion.
I mean, I have a cat, whom I love, and would pay expensive vet bills for, but I also pay the Giving What We Can Pledge of 10% of my income. 25% of that pledge is directed to animal welfare. My altruism does not need to be unidirectional.
Sure. I'm not careful enough with my words. I said callousness but really what I should have said was carelessness. And I am not such a rigid virtue ethicist that I would make this a litmus test on anybody's character.
But in some sense, the behavior is “wrong.”
It's not anti-human anymore than it is anti-one-group-of-humans to spend money to help another-group-of-humans.
My first responsibility is to my immediate family, and then to my extended family, and then to my super extended family, and then to community members that aren't related to me.
Humans are all part of the human species, which is analogous to our immediate family. Mammals our analogous to our extended family, vertebrates are analogous to our super extended family, and shrimp are analogous to community members that aren't related to you.
I realize this is all very alien to Anglo-American culture, and maybe even Northwest European Culture. I realize people from these cultures reject the idea of owing a debt to those who share your blood, and I am not going to convince anybody who wasn't ready to be convinced.
But the truth is the truth and your obligations exist whether you recognize them or not. I am grateful I was born into a culture where I was gifted with insight into my responsibilities to my family and to humanity. What a tragic thing it would be if I didn't understand these responsibilities. It would be like if I had never learned to wipe my ass.
Scott donated a freaking kidney so I guess he's done more than enough for humanity. But I'm not healthy or brave enough to do that so I hope I am able to give back in other ways. Like denigrating shrimp welfare activism; in small part, maybe that helps fulfill my obligations.
From my pretty shallow understanding of history, that’s exactly how Western thought used to go: first, your family, then distant relatives and community members, then your ethnic community of third and fourth and fifth cousins. But at some point in the 1800s, this led to the idea that every nation deserves to be self-governed---which in turn led to conflict where to draw borders, ethnic violence, and ultimately WWII. I suspect that western propaganda very intentionally denigrated familial-piety ethnic-duty cluster of ideas to avoid a repeat.
Non-European cultures by comparison didn't accumulate so much baggage around blood-bound duty. So it's totally acceptable to be openly ethnocentric. But as far as this being "the truth" and owing duty in proportion to your blood reatedness---I dunno. I suppose it depends on the ethical system to which you subscribe. You don't think any less of Scott for donating the kidney to a complete stranger rather than a fourth cousin; perhaps even admire Scott more for doing so completely selflessly vs. "helping family."
Taken to the extreme, your duty to a rat in your house cannot be less than to your houseplant.
I could have mentioned that my family is multiethnic and my neighbors are multiethnic. But I shouldn't have had to. After all, this is 2025. Our communities are FULL of multiethnic families and neighborhoods.
You should have considered this possibility. You should have then concluded that the priorities I initially laid out do NOT, in fact, justify ethnocentrism.
BTW, I didn't notice any golden age of peace and amity resulting from the deprioritization of family ties. Did I miss it?
Why do think reality is inscribed with these specific values instead of those articulated by, say, Confucianism?
Funny you should mention that, these values I'm talking about are closer to Confucianism than Northern European values. I think Northern European values are uniquely "frigid," if you'll excuse the pun, and uniquely atomistic, individualistic and vulnerable to being instrumentalized by misanthropes.
Why is it so obvious to you that shrimps should not be the priority in animal welfare?
This is silly. More happy shrimp means more delicious shrimp cocktails. Feed the shrimp!!! \S
I’ll subject myself to further ridicule, by again encouraging you to pony up the $20 - $25 to amazon prime to rent or buy the documentary “Age of Disclosure” produced by Dan Farah. The premise of the documentary is that Aliens are here, the US government (a small part of it) has the proof, alien craft and bodies. And that the way to get at the truth is for our congress to demand answers. Even if you put this at a 1% likely hood of being true, if it is true it is undoubtedly the most important fact that remains widely unknown. How would your life change if Aliens are here? So don’t read any reviews, ignore those voices trying to convince you that it’s all bogus, put on your tin foil cap, grab your tesseract full of inter dimensional popcorn, and join the ‘Age of Disclosure’, https://www.youtube.com/watch?v=SXUEcfgZv70
As an old school X-files fan, I really enjoyed Age of Disclosure, FWIW
OMG thanks, someone else who as actually watched it. (I never watched the X-files)
Here's what I don't understand about the claim that the US government is covering up the existence of aliens: what the fuck are all the other governments doing? There are, what, 190-odd nations on the planet? You're telling me *none* of these people have blabbed, they're all playing along, including our great rivals? Why would, I don't know, Uganda not just give a press conference and show off their crashed UFO? Are they that loyal to US military-political aims? The UK just happens to be exactly equally devoted as the US government – every US government ever, natch, for all their differences - to clamming up about their reams of UFO sightings, so they create the precise same level of coverup? Ireland doesn't want to ask the Vatican how to interpret their alien corpses in orthodox terms? Or are the aliens only visiting the US? If so, why? Pace a level of smug self-importance that comes with being the sole superpower on the planet, is the US *really* that much more important that aliens are just like "you know what, fuck scanning China. Even an American cow is more relevant to abduct"?
According the the documentary both the Russians and the Chinese are also gathering info from downed UFO's. IDK why they don't share, but there is something to, "tell your friends and your enemies will also learn about it."
Okay, but the Ugandans? The British? The Irish? The Norwegians, Indonesians, Moroccans, South Koreans, Slovakians, Laotians, Nigerians?
The relevant probability isn't p(aliens are here) it's the joint probability that aliens are here, *and* that Dan Farah has something to say about it that is simultaneously true, important, and new to me, *and* that there won't be someone else saying the same thing in a more accessible format in the near future.
As someone who has been following the UFO debate for many many years, and who finds text >>> video for this sort of thing, I assess that joint probability as too small to justify the time it would take to watch a documentary even if it were free. But a writeup summarizing his take with pointers to further investigation might be worthwhile.
Hmm I can try a write up, but my english sucks, I'm just not good at it. You might do better from an LLM. So I think Dan started the documentary several years ago. The origin of the story starts with the AATIP (Cia task force to sturdy UAPs) And the role played by Luis Elizondo. The people in AATIP were frustrated by pushback from higher ups in the CIA in particular from people supposedly involved in the Legacy program. (Some CIA/ air force/ Dept energy/ industry group that has been around since the 1950's (40's ?) collecting stuff from alien crash sites.*) They come to the conclusion that the only way to break this log jam is for one of them to resign their position and they will then be able to go the the media. The media will hopefully get congress interested, which will then generate more media interest which may get the congress to force the issue and have the Legacy program divulge what they know. If you've been following the UFO story, then you know that much of this has gone on... increased media and congressional interest. (Heck we've even got the Air Force to admit to seeing UFO's**.) We are stuck at the last step. We need congress to pass the 'UAP Disclosure Act.' The documentary tells this story much better than I have. But there is no real new information. (A lot was new to me.) There are many interviews with government officials, cuts from congressional testimony and media coverage.
*I'm reminded of X-com defense a video game I played in the 90's
** For many years there were reports from Navy and other pilots about UFO's but nothing from the Air Force.
Honestly, and I say this hesitantly.... this is very underwhelming, as far as evidence goes.
I suggest you look into Mick West for a thorough take down of any of the new info you've seen given to congress on UAPs.
They just simply aren't that interesting, except insofar as a reminder to remain extra skeptical when you aren't an expert.
"And the role played by Luis Elizondo"
I don't know why this immediately made a bell ring in my head, the name just sounds familiar for some reason (and not a good one). I'll come back to that.
Problem is, George, I (and others) have been hearing the same old song for decades by now. Every so often, new REVELATIONS. Every so often NO THIS TIME INSIDERS, REALLY INSIDE INSIDERS, TELL ALL. Every so often the same old tinfoil stories. And it never goes anywhere. Like I said, I've been reading this in the Fortean Times for issues upon issues. If there were genuinely any real alien materials and/or bodies and/or live aliens making contact, by now we would have something public and tangible, not shaky videos from the 50s and lots of mimeographs of "[redacted] [redacted] [redacted] swamp gas" papers.
Why should I blow $25 or its equivalent in my local currency on what is just one more in the "What the Bleep" genre of fake science tinged with vaguely sinister conspiracy murmurings? The X-Files mined this seam dry in the 90s, let it lie.
Okay, back to Elizondo. Ah yeah, this is why I had vague "oh crap this guy" reaction:
https://en.wikipedia.org/wiki/Luis_Elizondo#Questioned_and_spurious_claims
Yup that's the dude. Maybe we *are* in some post truth world, where my hero is just some jerk to you.
What you've basically said here is that there's this organization that's very interested in what another organization is doing, and they're unable to find out, and they're frustrated. Reading the paragraph above, this is what I find it boils down to.
But this is not unusual; it's extremely routine.
A straightforward way to approach evidence in favor of some unlikely claim (e.g. aliens are here) is to look at the evidence, conduct a search for alternate explanations, and consider whether any of those explanations are more likely than "aliens are here". One possible explanation immediately sprang up for me:
AATIP was funded in order to look into evidence of aliens, just on the off chance they really are here. They happen upon something they call the "Legacy Program" (LP). The LP is a Special Access Program (a highly classified project) staffed by CIA and USAF, and tasked to perform duties that are kept secret from anyone unless they are authorized to view classified information and have a need to know what the LP is doing (which means they work in it, make decisions about who will work in it, or allocate funding to it). We don't know exactly what LP does, but we can infer from the staffing that it's probably related to both aircraft and intelligence, including a great of information about our tech that we Americans don't want other nations knowing about. It *could* be related to aliens, but it doesn't have to be.
AATIP wants to know what LP was working on. Higher-ups in the CIA - specifically, people who decide who works in it, and therefore need to know what LP is doing - happen to know that LP doesn't have any proof of aliens at all. One of the higher-ups (let's call him Ashley) informs AATIP that LP doesn't have anything AATIP needs to know about. AATIP wants proof of that. Ashley tells AATIP there's no way to prove that to AATIP's satisfaction without giving AATIP a full tour of all of LP's assets, and AATIP has no NTK that (and Ashley leaves out his suspicion that AATIP is very motivated to leak whatever LP *is* doing, which is classified for good reasons), so it's not happening. AATIP is frustrated, and tries for a media angle, because Ashley is right that they are very motivated. In order to make that angle work, they make a lot of claims that suggest LP is working on or with aliens, but doesn't flatly state that (because lying like that would give Ashley a solid reason to shut AATIP down).
I'm betting a careful reading of that documentary's transcript will be consistent with the above two paragraphs.
Yes, this is a vey fine summary, thankyou. And yes we only have hear-say. And right they are clearly keeping it secret (whatever it is.) for a reason. A reason could be they don't want to freak out the public. About the NTN, apparently some presidents and senators aren't on the NTN list and that bothers them. According the the doc, Harry Reid was pissed enough that he started the AATIP. So it is like one piece of the government fighting another piece.
So basically "this lot refuse to hand over the classified material, that *proves* they *are* working with aliens" mindset?
I listened to most of Dan Farah's recent appearance on Joe Rogan's podcast. I'm highly skeptical, but apparently he got a bunch of pretty high level people like Marco Rubio to talk on camera about this stuff. I guess for me, then, the interesting question is what ulterior motive do these people have in pushing these alien visitation stories. The old theory was they wanted to deflect attention from experimental aircraft test flights the Air Force was doing during the Cold War. Seems like there's both a new level of specificity and higher-profile people involved these days in this theoretic disinformation campaign, though, so maybe that just means the US military has some new spiffy drone technology they want to test. I am of course open to other theories.
Is it Marco Rubio saying "Yeah, aliens are real and we are totally hiding that" or Marco Rubio saying "There's some stuff regarding unidentified aerial phenomena that we're not sure about" cut and interspersed with other material in a way that makes it look like Rubio is saying something more than what he's saying?
Boy I'd have to watch it again. But recall him saying something like, there's this one piece of government doing something without any oversight and we the congress want at least some of our people (senators) to know about it. I think he was part of a classified briefing... which of course is classified.
Good question. I don't know the answer. I was just going by what the filmmaker said on the podcast.
I am thinking that it started when we heard rumors the Soviets were doing remote viewing. We ginned up something for them to focus on, to see if there would be a feedback loop and they'd actually find something, or else report the Americans were hoaxing the world; and if remote viewing was bogus, they'd just waste resources on it. Also, it was a bit of a tit-for-tat move, just in case the Soviet remote-viewing effort turned out to be purely the same kind of show for us to waste energy on.
Note: just my theory, based on reading up on how people play these sorts of games and get lost inside them in personal interactions.
Yeah I heard Dan Farah on Rogan.
I like Rogan's take on Bigfoot, "You know who never sees Bigfoot? Hunters," pointing out that the legal requirement to positively identify targets clears up nearly all of the misidentifications.
I suspect the same is true of high-hours pilots, especially ones flying armed aircraft.
Ask your AI how many UAP sightings have been reported this year by the armed forces. And why are we ignoring this?
We are ignoring it because it is not freakin' true. You can have your hobbyhorse, George, but a point comes when nobody cares about what's going on.
The day I start trying to convince people on here about Eucharistic miracles, somebody gently force the keyboard from my hands and tell me to have a nice sit down for a bit. And I believe in the Eucharist, but that does not mean that you all have to be forced to listen to me rant on about it and why you totally should all watch this Youtube video about the recorded, verified, credible witnesses speak on the record, about the long history of Eucharistic miracles HAPPENING IN A CHURCH NEAR YOU RIGHT NOW, WHY DO THE PROTESTANT HERETICS WANT TO COVER THIS UP????
I appreciate the prompt. My thought from analysing this matter in light of thought patterns I learned on this site:
1. It seems a low likelihood that aliens are observable to us when we have so many cameras pointed in every direction now, more all the time, and still all we get are blurry images.
2. The usual "aliens wouldn't be so incompetent they'd crash on our planet and let us pick apart their ships" one.
3. I love ideas about the Alcubierre drive and the idea of other inhabited planets, but the vastness of space and the extreme difficulty of anything FTL, even merely fast interstellar travel, just crushes it for me.
That out of the way, I'll continue to strain credibility with my pet theory:
The one industry that is sure to flatline if aliens were disclosed is the science fiction genre. So, perhaps the aliens are SF fans, and love our movies and books in the same attitude with which we'd consume steampunk or other alternatve-history content.
My position is: do aliens exist? Very probably, it's a really big universe out there. Are they crashing all over our planet? Not unless it's the equivalent of drunk idiot teenagers joyriding. I used to want to believe this, that there was some grain of truth to all the sightings, but over time I've dropped any credibility I gave it. I'll have to get my aliens from SF like everyone else.
Years back I read Jung's book on UFOs and I think he came nearest to what is going on: people used to see the fairies, and angels, and in our modern scientific era we now see UFOs and little green men. It's all the same psychological mechanism in action:
https://www.amazon.co.uk/Flying-Saucers-Modern-Myth-Things/dp/0415278376
https://jungiancenter.org/jung-on-signs-in-the-skies-a-jungian-perspective-on-ufos/
"We have here a golden opportunity of seeing how a legend is formed, and how in a dark and difficult time for humanity a miraculous tale grows up of an attempted intervention by extra-terrestrial ‘heavenly’ powers—and this at the very time when human fantasy is seriously considering the possibility of space travel and of visiting or even invading other planets.
Jung (1958)"
Well we had all the "Drones over New Jersey" stories that happened about this time last year. There have been hundreds of Air Force / Navy UFO sightings reported in the last year. And no, we will need Sci-fi more than ever. Who else has been thinking about this kind of thing.
I expect the chance of the specific scenario the conspiricy theories are on about to be extremely small. Much smaller than 1%. You can't just make up a large complicated pile of nonsense and act as if it's probability isn't exponentially small. Every extra bit of detail halves the probability.
The specific "alien conspiricy theories" have a lot of parts that don't really make sense.
Aliens, fair enough. Aliens are quite plausible. Intelligent technological aliens, still not implausible. But I would expect evidence for this to be seeing dyson spheres through our telescopes, or receiving a radio message.
If there were alien spaceships on earth. Well breaking from interstellar speeds seems like it would involve a massive rocket plume. Again, not craft and bodies.
And mind uploading or similar seems like it should be easier than interstellar travel. And we already mostly use unmanned space probes. So the idea of a crashed probe leaving "bodies" is dubious.
And why would it crash. Surely the alien spacecraft should be reliable. And if it does crash, well nanotech should be extremely good at self repair.
Again, why a couple of bodies in a flying saucer? Why not showing up in a massive, obvious and undeniable way. Or successfully hiding from our crude tech if they wanted to hide.
Let me guess, these "bodies" are grey (or green) humanoids that fit every "scifi alien" cliche.
And then, why would the american government be hiding this?
When you add in all the specifics, this sounds very unlikely to be true.
But this also sounds exactly like the sort of thing that a lot of people want to be true. There is plenty of UFO hype about videos that seem to be birds. Plenty of similar nonsense surrounding bigfoot etc. In short, you would expect something like this documentary to exist, in a world without aliens. Was anything about the documentary unexpectedly strong. Ie evidence that's stronger than you would expect from a bunch of cherry picking and lying UFO enthusiasts in a world without aliens.
Yup, I have no idea why the story has unfolded the way it has. But by the end of the documentary it is a consistent story.
Another possibility is that there's something about the interstellar environment that makes it prohibitively hard for even von Neumann machines to survive a trip to another star, let alone with their self-replication mechanisms intact.
Yet another possibility is that artificial self-replication is itself harder than estimated (bear in mind, we've yet to do that in a controlled environment -here-).
I wasn't thinking of dark matter; rather, plain old radioactive particles, ions, and even just ordinary matter, i.e. dust and ice. Interstellar space is no doubt a near-vacuum, but there's a *lot* of it for a probe to have to travel through, and at a reasonable fraction of lightspeed for a probe's presence on Earth to be a question worth considering. Which means that probe will probably need a way of dealing with anything it collides with en route, even if it's tiny.
I think the most likely, but not the only, option is that intelligent technological life is really rare.
...are those really the only two options?
Why is the default for skeptics "1%" rather than "0.00001%"?
If you want to make the argument that it is worth someone's time, then give the best piece of evidence from the documentary, preferably with the sources (one would hope the source isn't just "the documentary").
You have to put your own numbers on it. There is no smoking gun in the documentary. Mostly just people wagging jaws. But you can ask why all the jaw wagging.
"But you can ask why all the jaw wagging."
...must I? People wag jaws on the internet for millions of hours, about much less interesting topics. I have no problem coming up with a boring answer for "why all the jaw wagging".
Yea that sentence made me LOL for a good while. My office neighbor became concerned.
Most people don’t understand what extreme probabilities are - both that 1% isn’t that low, and that there are plenty of things below .00001% (like the probability that the next plane I am on will crash while I am on it).
Right! I've updated my probability to be above 1%. Unlikely things happen all the time. How unlikely am I?
> 0.00001%
That probability is below the noise floor. People hallucinate at higher rates than that. Whatever mental process it is that spits out such a tiny number, the likelihood that one went wrong in one's reasoning and/or assumptions, or is simply failing to perceive reality correctly, will be higher than this.
Once you've rounded it off to "not worth spending any more time on this", there's no point trying for more precision.
You can have coherent estimates of a probability less than 1/10M. For example, I believe the probability of an asteroid on the same order of magnitude as the one causing the chicxulub crater hitting earth in a given year would be less than 1/10M based on the observations we have.
> People hallucinate at higher rates than that.
It's 1 in 10 million. While "people" might hallucinate at higher rates than that, *I* don't hallucinate one out of every 10 million things I see.... (that I know of).
>Whatever mental process it is that spits out such a tiny number, the likelihood that one went wrong in one's reasoning and/or assumptions, or is simply failing to perceive reality correctly, will be higher than this.
Ok, so let's say the minimum is actually .01% (or whatever value you choose, so long as it is above the floor you quote), can I offer you a bet at those odds versus my ~0% odds about, say, lizard men ruling the world behind a shadow government, or the moon being made of cheese, or the world being surrounded by a celestial crystal sphere created by Uriel?
Clearly, there is a point at which the claim is so ridiculous that you *have* to go to less than 1 in 10 million.
Is the alien thing actually that low? Not sure, really. It probably depends on the specifics, but my point is that it is rhetoric to say "Even if your probability is 1%" because it implies that 1% is some absurdly low number for a probability for something like this, and it simply isn't.
> can I offer you a bet at those odds versus my ~0% odds
No, I'll take no bets on random tiny numbers I can't reason about the relative magnitudes of with any confidence. I'm completely happy to file the lizard men, cheese moon and celestial crystal sphere all in the same mental bucket. I really don't think there's any merit trying to separate them from each other.
If you believe you know better, I'm completely happy to bet money that none of these three things are true against your proposal that one or more are.
Here's Yudkowsky himself on the subject: https://www.lesswrong.com/s/FrqfoG3LJeCZs96Ym/p/ooypcn7qFzsMcy53R
And here's Scott: https://www.lesswrong.com/posts/GrtbTAPfkJa4D6jjH/confidence-levels-inside-and-outside-an-argument
There's a big difference between saying "a human being cannot reason about numbers so tiny" and, essentially "that number is below the noise floor, so you aren't allowed to use it".
I agree noise floors are useful concepts. I disagree with what I *thought* you were saying, which was "therefore you should never assign something probabilities below the noise floor". But it looks like, here, you're saying something more like "Since I'm a human being, I can just round all this down to zero for practical purposes until there's some big, earth-shattering evidence that brings it into the realm of being remotely possible"
If so, I agree. I still disagree with OP that "even 1%" is anything other than rhetoric designed to make 1% look reasonable a priori.
>If you believe you know better, I'm completely happy to bet money that none of these three things are true against your proposal that one or more are.
As a side note, my point was that I would bet *lower* odds than you (since you were arguing against using such low probabilities), so you'd have to pay out to me when they didn't come to pass.
Yup, I think we're on the same page. For me, things like "0.00001%" in casual conversation trigger the same twitch as seeing a number with too many significant figures cited in a newspaper article.
> How would your life change if Aliens are here?
What's true is already true. Admitting it doesn't change this. We know we can live in the world where the true thing is true, because we are already living in that world.
My life would not change if aliens are secretly here, because if aliens are secretly here they are already secretly here and that is the world my life, the way it already is, is already in. The only thing that would change if I bought the "documentary" is that I would be down $20-$25.
Now, if the president were to, oh, announce a trade treaty with the aliens, certainly that might translate to tangible effect on my life, just as any foreign treaty might. Or if they declared war on us, my life would change, just as it would if a foreign nation went to war against us. But I really don't see how me buying this documentary will accomplish anything other than moving $20-$25 from me to its creator.
How was your life changed by watching the documentary?
Yeah, it's worth calculating some kind of payoff matrix here.
What's the maximum benefit to believing that the US Government has a shed full of aliens? Well, if it is later revealed that they do, then I get to say "I told you so", but nobody will be that interested.
What's the maximum downside of believing that the US Government has a shed full of aliens? Well, if they don't, or if they do but it's never proven, then I wind up looking like a kooky weirdo.
The upside of believing in flying saucers is small even if flying saucers are proven exist, and the downside is significant *unless* flying saucers are proven to exist.
Yup, I'm wearing my tin foil hat proudly. :^) (well mine's made from aluminum foil.) For me, knowing Aliens are real and here would be the coolest thing ever. (and maybe the scariest too.)
I'm surprised to see that sort of attitude here, doesn't this argument apply equally well to most of the articles on this substack?
Don't get me wrong, I'm not sold on the documentary myself, but if I thought it had a reasonable chance of being true I would definitely want to know, even if it was just for my own curiosity and interest in truth-seeking.
Thanks. I find it weird also. The lack of any interest. But I chalk it up to a long standing campaign to discredit all UFO reports. (One could ask why we have that effort and where it comes from.)
How's my life changed? My kids think I'm a tin foil hat wearing old crank... but they mostly thought that before too. Some other minor financial things. I will freely admit that I like the idea of aliens. And I have like a million and one questions. But first we need congress to shake out the truth... which will be blocked.
> How's my life changed? My kids think I'm a tin foil hat wearing old crank... but they mostly thought that before too. Some other minor financial things.
Oh... this suddenly stopped being funny. :(
Grin, I'm not trying to be funny. So part of the problem (if this is true) is how do you release it? Grasp the nettle firmly, (all at once), or in little drips and drabs.
I'm expecting drips and drabs, so at least a year... but maybe even longer... What would you do?
1% chance it's true, 99% chance it's not. Tell me where that is wrong, because I'm sure somebody right now is going "noooo, that's not how probability works!"
If aliens are here, then they're not very good at being alien, are they? Crashing their ships and letting their dead be found by the primitives. If they are doing anything, it seems to be weird experiments on cattle and humans. No sign of proper first contact.
I wouldn't be adverse to the notion of alien anthropologists coming here to observe a pre-spaceflight species in the wild, but the rest of it is just unconvincing.
How would my life change if aliens were here? Not at all, it would seem, since the proposal here is that aliens do exist and are here right now. If that is so, then things would continue on as they are. No, aliens exist and are here and we can prove it and Congress released footage of the crashed ships? Eh, again, things would continue on as they are. World governments would squabble over who has the tech, what tech is it, should it be shared, etc. but unless the aliens land in capital cities of the world and announce themselves, it won't change much.
If it's true and aliens with interstellar travel capacity exist and are here, then we are in the position of those New Guinea tribes making (or avoiding) contact with the outside world. We would be so far behind that the notion of capturing UFOs or controlling that tech would be like the cargo cultists trying to create landing towers and runways out of palm trees and bamboo.
So the question would be "how did the lives of natives changed when a higher tech civilisation made contact?" and following on from that "who is the dominant culture, our native primitive global one, or the aliens?" Meiji era Japan where we have to adapt or else, because like it or lump it, the black gunboats are in our harbours. Except we don't yet have the alien black ships in our harbours, so that's the big difference.
>Meiji era Japan where we have to adapt or else, because like it or lump it, the black gunboats are in our harbours.
I like this example because, while the Perry Expedition to Japan wasn't a true first contact, it's close enough to one for practical purposes and moreover, it happened in modern times with large literate cultures on both sides of the contact. The Autobiography of Fukuzawa Yukichi was a review contest finalist here last year. It's the memoirs of a man from a minor samurai family who had just started studying the Dutch language (Japan had very limited and restricted contact with Portuguese and Dutch traders prior to the Perry expedition) when the black ships showed up. Fukuzawa was one of the first Japanese people to learn English, participated as a translator in Japan's first diplomatic expeditions to North America and Europe, and had a career as a teacher and education reformer after returning to Japan.
I've read it and noticed a tons of parallels between Fukuzawa's experiences and the kinds of scenarios that are commonly explored in SF books and movies that explore the aftermath of first contact, but in a context of "this is what actually happened, from the perspective of a well-placed member of the culture who got contacted".
Yeah! So I'm hoping we've all read enough sci -fi so that we have a chance of surviving first contact. And won't form a cargo cult. Why they are crashing is an important question. (I would guess that those in the know have some good ideas.) So first a 'fact' (piece of info that may be true) from the documentary. There are at least two 'species'/types of aliens. So maybe they're at war and shooting each other down? I know not a very fun picture. So on a hopeful note, it could unite mankind. We have to stop all the stupid wasteful fighting and get busy.
Finally I just encourage other people to watch the doc, so we can talk about it with a little more information. (And yes no one has told me shit yet. It's mostly jaw waving.)
It seems very, very unlikely that the majority of UFO reports are due to aliens.
More plausible (still unlikely): nearly all UFO sightings have prosaic explanations: weather balloons, the planet Venus, people just making it up. But also, there really are aliens here, who are much harder to observe.
I just want to say, that from the point of view of the military industrial complex that has control of this info, you are saying exactly what they want you to say. A stigma against UFO's is built into the narrative. (I know you see me as a crank with a tin foil cap. no worries I embrace the look.)
Edit: Oh most of the sightings talked about are armed forces personnel (pilots) and the only data released are distant video shots and radar images.
AI answer to how many reported sighting by armed forces.
"The number of reported UAP sightings by U.S. armed forces has increased significantly, from 144 between 2004 and 2021 to 757 reports received between May 1, 2023, and June 1, 2024. The surge is attributed to less stigma and more reporting protocols, not necessarily more actual events. The Pentagon's All-domain Anomaly Resolution Office (AARO) is now responsible for analyzing these reports."
I have another pet theory that there exists "some kind of phenomenon" that appears to take on the form of most/some of the people looking at it, coming from their beliefs about the world. This could be easily defined as crowd hypnotism, but I'm saying there might be a very rare physical phenomenon that has its own reality external to shared human delusions. So, in older days it showed up as angels, and today UFO's.
Yes. And given the number of personnel and sensors, we should expect to see lots of unidentifiable blobs in the distance.
Tiny blob of something flying in the distance that could be a bird, or a balloon, or a secret drone.
And if 80% of these detections are blurry birds, why can't 100% of them be. It's clear that normal non-alien phenomena can generate UFO sightings. So the shear number of UFO's isn't evidence for aliens existing.
What you need is something that's close to the cameras, so you can see the details, and know it's alien.
I don't think it's that likely that the MIC would want to hide the existence of aliens. I think it's even more unlikely that the aliens would want to show their existence to the MIC and only the MIC.
You think the aliens tech would be simple and comprehensible enough for the MIC to replicate much of it.
As opposed to, take a smartphone. Cut it in half. Give both halves to a caveman.
Also. To me it seems "Aliens crash landed on earth, we have the spacecraft and we are developing all sorts of cool alien tech, no we aren't telling you the details" is a great boast, and a good excuse for needing more funding.
“The flash of light you saw in the sky was not a UFO. Swamp gas from a weather balloon was trapped in a thermal pocket and reflected the light from Venus"
So what's that worth? I encourage you to spend the $25 and watch the documentary.
I think this is a hard sell to those of us around for the von Daniken years. We grew up on 70s documentaries of ancient astronauts, was Atlantis real, the crystal skull and all the rest of it.
One more "documentary", even if it has modern gloss and "no but this is a real Pentagon guy!", isn't going to move the needle much. We're pretty much seen it all before, and the underlying story hasn't changed: crashed craft, the greys versus the other aliens, bodies in secret government labs. EDIT: It's like the Fortean Times, which I stopped buying after a couple of years because the same stories turned up again and again with little to no development. Great the first time you hear about it, still fun the fifth time, by the thirtieth time it's old hat and boring.
Get a real alien on TV broadcasting to the globe and we'll buy the docudramedy.
Yeah I'm an old fart and watched much of the von Daniken stuff with my dad. The number of government officials giving interviews makes this much different from the tales in the 70's.
Oh what, they've got *six* government officials instead of three this time?
If you've been around since the 70s, you know that the "it's all a giant cover-up!" set have touted "pilot X, official Y, other guy with fancy title Z all claimed to have witnessed this, that and the other". It's always been some government official.
Yet we still haven't any aliens rocking up to the European Parliament going "hey guys, we've been here since 1940 so we thought we'd finally say hello".
query *age of disclousre list of government officials interviewed*
answer (AI)
Politicians
Marco Rubio (Secretary of State)
Kirsten Gillibrand (Senator)
Mike Rounds (Senator)
Anna Paulina Luna (Representative)
Tim Burchett (Representative)
André Carson (Representative)
Dan Crenshaw (Representative)
Mike Gallagher (Representative)
Intelligence and military officials
James Clapper (former Director of National Intelligence)
Christopher C. Miller (former acting Secretary of Defense)
Christopher Mellon (former Deputy Assistant Secretary of Defense for Intelligence)
Tim Gallaudet (former commander of the Naval Meteorology and Oceanography Command)
Lou Al Fazio (former NSA analyst)
Brett Feddersen (former member of the U.S. National Security
Yeah I don't know why it turned out this way and not some other way, but here we are.
Scott seems almost to be inviting ragebait, so how about this: it's not much fun to be a white person competing in the international 100m, so why don't we have different races for, er, different races? Or perhaps someone has already had that idea at some time in US history.. ahem.
The answer, I suppose, is that white athletes would have ended up with all the best facilities (in the past), are embarrassed to ask for their own category (today) and can't present themselves as a disadvantaged group overall (no doubt groypers will disagree).
Back to the start: it seems to me that women's sports are justified by (1) historical disadvantage and (2) the fact that mammalian sex is a natural kind. We could deal with (1) in other ways, for example by positive discrimination (promoting 10th place to 3rd, say) or handicaps (as with racehorses, male runners could be made to carry weights!) but (2) is worth thinking about because not every 'demographic' can have its own sporting category. Hence we have the Paralympics but not the Flabolympics.
Two further points to consider: (a) injury and (b) sexual predation. Heavyweight boxers can't self-ID as featherweight, yet trans-identifying men have caused serious injuries in women's contact sports. As for (b), this is an obvious likelihood and ignoring it has not helped transactivists this far, but to repeat the obvious: if you allow, say, men aged 40 to self-ID into the under-15 category, you should not be surprised at the resulting scandal in the changing-room showers. And predation seems especially likely where males are allowed to self-ID as female with no gatekeeping allowed.
(2) is worth thinking about because not every 'demographic' can have its own sporting category
Sure they can, who is stopping them?
Not everything that is possible is likely to happen. No-one's stopping me from flying to the moon, but the practicalities are against me
Actually if you tried, I suspect there would be plenty who would try to stop you!
And also actually back at me, if you tried to set up a "left handers only" sports league, you'd probably get sued for discrimination.
Absent those considerations, there's no reason we couldn't let free association sort this out, though. Not every division would reach critical mass, one should be able to find enough southpaws (or whatever) to enjoy a friendly game every now and then.
Can you point me to links of “trans-identifying men have caused serious injuries in women’s contact sports”?
Thanks!
I'm not your research assistant. Learn how to use Google. As I'm in a helpful mood, I'd recommend you start with "Ellen King" as a search term.
You made a claim. It's on you to provide evidence for it. You've come back with one data point. Which tells us pretty much nothing about anything. And you have the gall down thread to say that someone else needs to read up on the basics of statistical reasoning?
Of course you can. You can also point to cases of non-trans (actual women) causing serious injuries in women's contact sport. What you cannot do is show that injuries had anything to do with trans-ness.
You need to read up on the basics of statistical reasoning
Pound-for-pound, males have a more powerful punch.
https://attheu.utah.edu/facultystaff/carrier-punch/
How much does this change after HRT (suppressing testosterone to natal female levels, and administering exogenous estradiol to natal female levels)? What length of time on a female-equivalent hormone balance sees the bulk of punch strength reduction and how does this point compare to the natal female distribution?
How does the body signal for muscle growth and retention? What happens to muscle mass when that signal is suppressed? Does it increase, stay the same, or decrease? If muscle mass decreases, might that also reduce strength?
Michael Smith made a claim. I’m asking for evidence. In general, these open threads could use more evidence of people’s claims.
Ultimately, you should not compare yourself to others. The most fair sport category is one that includes only one person -- you. No more unfair advantages of any kind.
Some people suggest that you should compare yourself to yourself; to check whether you have improved compared to the last month or the last year. I reject that as microageism.
There's a case to be made that all competitive sport is a waste of time. If that's your position, I wish you all possible success in convincing people..
Team sports are also a great way to divert our evolved-in tribal machinery toward harmless pursuits instead of toward wars and pogroms and the like.
My position is that sport is a hobby.
There is a political decision to be made whether people should finance their own hobbies, or whether government should finance them from taxes.
If it's the latter, it would make sense if the money assigned to a competitive sport was proportional to the number of people who enjoy this hobby. Maybe more if the hobby has positive externalities (e.g. making people more healthy), or less if it has negative externalities (e.g. fans destroying things).
It's also auto-ableist. By giving preference to your stronger self you're denigrating your weaker self, whether that's the present or past identity.
Reductio ad absurdum has its uses but I don't think it works here. This is all very much a stretch.
Who can afford £11 a month for one Substack? This seems like an exorbitant price to ask. I don’t mean morally, though I think that’s an important question, but I just think there might more more than 2x subscribers if it were half as expensive (I’d be among them).
I also don’t get the impulse. If you can afford it, shouldn’t you want more people to read the locked posts even if deemed important
Sorry this is so critical. I do think your writings are worth at least 11 a month
Then don't pay it, sheesh. I hate people who complain about prices. You're free to leave anytime. No one owes you a product that fits within your budget. Everyone has a right to maximize their revenue. Obviously lots of people are willing to pay the subscription fee, otherwise it would be lower. Prices are a two-way street and aggregate information about both supply and demand, so you may as well criticize current subscribers for paying. Also bear in mind that supply is elastic so a high price here sends an economic signal that incentivizes the creation of more high-quality writing.
I worked with marketing teams for 5 years, there is nothing more valuable to a marketing team than honest and direct consumer feedback on their policies and products.
The idea of 'customers should stop expressing their preferences and intentions to the people trying to sell them things' has always seemed insane to me. Please, for god's sake, tell me as much as you possibly can about what I could be doing to earn your money!
There is only one type of feedback that's relevant w/r/t to price: whether or not you buy it. OP isn't "providing feedback" he's whining.
That is the only price the 'invisible hand of the market' can receive, if you go in for that kind of thing.
Actual companies spend huge amounts of money and effort on getting more feedback than that, including scanning social media for people talking about them. They find this feedback very valuable.
This is just a brute empirical fact about reality, one that I am confident about because I worked on the teams doing it. You can try to deny that fact if you want, I guess, but I'm not sure what you're getting out of it.
Explain to me exactly what information that Scott got from a single person saying "this isn't valuable to me at this price" that he didn't get from the fact that that person didn't subscribe. He knows how many views he gets. He knows how many subscribers he has.
>Actual companies spend huge amounts of money and effort on getting more feedback than that,
Yes marketing departments are well-known for wasting money on technically misguided analyses. I, too, have provided technical assistance to biz-dev teams and can testify that they're largely incompetent. They only thing most of them are interested in is generating reports that demonstrate to leadership that they're doing something.
Halving the price is unlikely to get double the subscribers. The friction from $0 to even $1 is huge and most people will never pay it. Of the small pool of people who are happy to pay for a subscription in general, they are very likely to want Scott's writing no matter what, and as Scott has a monopoly on Scott's writing, people are going to pay whatever is being asked. There's an upper limit of course, but generally there is no competition making the service inelasric.
If you're a student or you can't afford it, you can use the link in the subscription page to reduce it to $2.50 a month! ;)
> If you can afford it, shouldn’t you want more people to read the locked posts even if deemed important
Have you considered that he doesn't want people to read it? Why do you think the last three hidden posts are about trans people, Gaza, and fascism? It doesn't attract the best kind of attention, if you know what I mean.
> I just think there might more more than 2x subscribers if it were half as expensive
The trick is how to find out whether this is true, without making subscribers angry.
(If you don't care about making subscribers angry, you could halve the subscription, check whether the number of subscribers doubled, and if not, double the subscription again.)
Tiered pricing is the traditional way.
Yes, but ACX is a bit of a nonstandard product, so I am not sure how much the traditional solutions apply.
I mean, the majority of content (most Scott's articles and Open Threads) is free. You pay for a little extra content (some Scott's jokes, and subscriber-only Open Threads), which could perhaps be better described as "a gift with a nudge". Gifts can have a weird dynamic, more complicated than the usual supply and demand. If the subscribers stopped paying, they would still keep getting maybe 90% of the value they are getting now.
Tiered pricing would require Scott to produce more extra content. I think the billing is per month, so for N tiers it would be nice to have N extra things every month, otherwise someone may feel cheated. It could also feel weird, like: "hey, if you help support my hobby, thanks, and I will tell you a joke... but I won't tell you the other joke, because you didn't support me as much as some others". The current way kinda like a gift exchange, the tiered way would feel less so.
Or maybe I am wrong here, and Scott is just leaving money on the table. The problem is, if he makes an experiment and it turns out that there was a weird psychological reaction from the current subscribers, something like: "I don't want to pay more... but I hate the feeling that I can't see some hidden content despite sending Scott some money... well, if I can't see all the jokes, I might as well unsubscribe and save the money", that change may be irreversible.
> Tiered pricing would require Scott to produce more extra content
Not necessarily. e.g. you could have a variable delay on the premium posts instead of just a straight unlock, so it becomes "pay $x to have access to the posts a week later after the comments section has mostly run its course, or $5x to have access immediately when they go up". (Substack's awfulness helps with that: the comments section becomes unusable around 1k or so comments mark, so there really is a race to get to say your piece). Or you could do the Humble thing and have "pay whatever you want, but if that's at least $x you have access to the premium posts", instead of a flat fee.
Can anyone here point me to a primer on shrimp welfare arguments, directed to someone already sympathetic to animal welfare in general?
I've realized that the ratio of trans women in sports : shrimp welfare comments in this comments section is disappointingly high, so I've decided to be the change.
We need the crossover version: trans shrimp in humane sports!
What about transhumans in shrimp sports?
This is off topic but I'm curious about your username, do you live there or are you just a fan?
I made this account initially to write a comment about how I, unlike Scott and apparently all his fans, loved modern architecture. Hong Kong is one of my favorites. So, just a fan.
no no, there are dozens of us! you are not alone!
I don't know if I'd classify HK as modern architecture. I do love it as a glorious mess of a place, but I associate it more with "crazy eclectic mix of everything" than being a good representative of modern architecture (Singapore or maybe Shanghai seem like cleaner positive examples of that).
You're sleeping on Shenzhen and Chongqing here. I forgot the context of the original comment, but I was discussing how I heavily weighted scale and visual glamour compared to modern-architecture-haters.
Aforementioned Bentham's Bulldog argument for Shrimp Welfare....
https://benthams.substack.com/p/the-best-charity-isnt-what-you-think
Are shrimp welfare activists ENEMIES of the mammal kingdom, in a sense? Or at least enemies of vertebrates?
Not sure I understand this. He's talking about reducing the suffering of shrimp, not increasing the suffering of vertebrates.
If I give money to reduce the suffering of children, that doesn't make me "an enemy of adults". If I give for the welfare of horses and cows, that doesn't make me "an enemy of primates".
As he see's it...he's giving on the general premise of "reducing the most suffering possible per dollar", and the fact that shrimps win out in that argument is just how the chips fell. If the same sums came out in favour of (say) giving the money to causes that improve the welfare of cows.... he do the same for cows.
He isn't the enemy of vertebrates in the real world example, nor the enemy of crustaceans in the counter-factual just because of how the calculus works out. In both cases he'd be the enemy of "suffering" and trying to make sure he is as effective an enemy of suffering as he can be.
1. Donating a penny to one cause is withholding one penny from all others. Although technically, it's more like withholding a fraction of a penny, since working towards one goal can have knocked on effects that help other causes. But I'm not sure how this justifies shrimp welfare.
2. Saying "every sentient life has equal value" is pure ideology, and it's based on vibes. I don't care how many math problems you do, and how often you update your priors. If I'm not feeling the vibes, I'm not feeling the vibes.
1. So you are an enemy of any category of objects on earth not supported by a charity, if you donate a dollar to that charity? Thats not usually how we use the word "enemy".
If I donate to (say) a donkey sanctuary.... that makes me the enemy of any living thing that isn't a donkey, all the inanimate objects that aren't involved in the welfare of donkeys, and the entire universe outside of earths atmosphere. Because I donated to a Donkey sanctuary?
Can't I just be a "friend of donkeys and neutral to every other non-donkey object in the universe"? Must donating to the donkeys make me the enemy of everything else? After all, I *could* have donated that dollar in some other way that would have benefitted all those other objects and callously refused to do so in order to support Donkeys.
This is NOT how we use the word "enemy" in normal conversation.
2. First, this seems orthoganal to the "enemy" argument. This can be true or false and not make him an enemy of mammals/vertebrates either way. So we are now branching outinto pother arguments/discussions without settling the enemy point first .... But even taking it as a separate line of discussion ... its not his position.
He explicitly does NOT value all sentient lives equally. He discusses extensively that shrimp are clearly NOT equal to mamalian/vertebrate sentient life as a moral consideration. IIRC he discusses "discounting" their suffering by between a factor of 100 and 10,000 compared to vertebrates and points out that because so many shrimp are involved, you can easily discount by that factor or more and it still equates to a reduction in overall suffering compared to alternatives as so many shrimp are involved (numbered in the tens or hundreds of millions) and the cost of alleviating the suffering is so low (a handful of dollars per million) that the math works out anyway as opposed to (say) donating the same dollar to a donkey sanctuary.
Nor does he impose any individual value (a 100 to 1 discount....a 10,000 to one discount.... a million to one discount) on the reader, he leaves it up to you to decide what that discount should be (and even leaves it up to you to apply an infinite discount if you think thats appropriate) and provides information relevant to you making that choice. So its hardly imposing any ideological conclusion. He describes his personal conclusion, the facts as they are, the maths as it is, and leaves up to you how to apply those and whether you do so in a way that ends with your agreement with his point or not.
Finally, I don't see how you can say "you're not feeling the vibes" of his argument.
Given you're response 2 above it's absolutely completely clear you haven't read his argument. The "vibes" you are or are not feeling.... and vibes in relation to "an argument I made up in my head specifically so I could disagree with it". I can totally understand why you aren't feeling vibes towards accepting that argument, you've personally constructed it in a way to give you the opposite "vibe".
Perhaps you should read his argument, "feel the vibes" from that actual discussion... and then disagree (if you want to) rather than "imagining what he said, in as strawmanny a way as you possibly can ("each shrimp is fully morally equal to an adult human"), so as to support the conclusion I already reached before listening to what he had to say".
Its not as if I totally disagree with you here, I don't think I reach the same conclusion he does EITHER.... its just I've read his argument and conclusion and (like you) "vibe decided". I just read it first.
I feel like you have nitpicked my word choice in a way that does not add value to the discussion.
Thanks, I'll take a look.
By making it lower?
I am illiterate.
The pinned article on Bentham's Bulldog is good.
4:
1. Well since transwomen are not women by the definition they cant participate in women sports. Thats the ground.
2. Applying that logic consistenly we also would have to allow cis men participate in women sports and adults in kid sports. This is just continuum fallacy.
I have no problem renaming the categories "open and female" instead of "men's and women's" if it will help people feel better.
I do have a problem telling the girls I coach that they must compete with males who have a distinct physical advantage over them.
Most sports *are* Open and Women already, woke people are just quibbling the *term* women because they've decided it should mean something other than what it de facto does and has done for centuries.
In fact, I think the only sport which had a rule that women could not participate in the "men's" division even if they qualified, in the modern era, was golf, and that was changed in the 1990s or something. Otherwise, any woman who makes the cut can play in the "men's" Wimbledon, for example, it's just that that's physically impossible for a woman to achieve so in practice there are no women. That's the whole reason why they get their own category in the first place.
But that would be unfair against men
Agreed, I've long advocated for this solution. As far as sports go there are two categories: actual women who were born women and everyone else.
This just illustrates the insanity of the transgender question in the first place. Gender isn't arbitrary: it serves lots of objectively important purposes in thousands of subtle ways all throughout society. Rejiggering all of that for a vanishingly small fraction of society is just objectively foolish. Society should never waste resources worrying about a 1%-sized demographic. It's simply out of proportion.
"Society should never waste resources worrying about a 1%-sized demographic. It's simply out of proportion."
That's not true. The meaningfully wealthy are far less than 1%, but their concerns are well worth considering, not only due to the knock-on effects on society at large but because they have legitimate claims on the conduct of the rest of society (such as not being decapitated for their money, again). Admittedly these claims scale down to every subsequent tier of prosperity, but even if they didn't they would still have a justifiable demand to have their claims of property respected.
However, you're certainly correct that it's bizarre for a tiny cultural/medical minority community to have palpable sway over social mores, especially in a democracy whose whole design purpose is to marginalize such irrelevant minorities.
Yes agreed my point is actually about value and not just size. The wealthy are different because society actually does have a rational interest in catering to them.
>especially in a democracy whose whole design purpose is to marginalize such irrelevant minorities.
Democracies aren't designed to actively marginalize minorities, but neither are they designed to cater to them. Given the choice between active marginalization and active catering I'd say it's objectively better in a net-social-outcome sense to marginalize them. Society simply can't function if every 1%-sized valueless group can command public attention. That just leads to political gridlock as 100 different factions demand conflicting policies, particularly when those demands come on the back of completely inflexible rights-based claims. That's just anarchy. Minority rights don't entail the entire country changing its bathroom policy to enable the preferences of a tiny group of ill people, for example. That's true regardless of your views on the medical legitimacy of transgenderism. If there was a cancer that 1/1000 people had I wouldn't endorse destabilizing our national politics to make their lives more convenient either. The needs of the many outweigh the needs of the few.
"Democracies aren't designed to actively marginalize minorities"
Yes they are. Majority rule entails minorities only not getting marginalized if if amuses the majority more not to marginalize them in whatever respect – in other words, it's dependent on the amusement of the majority. This is crucial because it's *why* minority advocates of various stripes are always so eager to corrode democracy and put various things beyond the reach of the people, whether in the form of binding treaties and conventions, leaving matters to "an apolitical team of experts" (i.e. in practice a bunch of middle class managerialists ideally drawn from the ranks of the advocates themselves so that they can be expected to be reliably ideologically captured), or any other form.
EDIT: Forgot to add, I agree with the rest – "Society simply can't function if every 1%-sized valueless group can command public attention" is indisputable. Going by some pre-2010s-inflation figures on trans people they, at that time, were a smaller group than sufferers of cystic fibrosis. As a thought experiment, imagining fibrotics *alone* having the same amount of cultural/news airtime as transsexuals (while not depriving the latter of attention) is instructive: it would be an exhausting barrage. *Every* topic of that level of general interest laying the same claim to our attention would be wholly unlivable. People would start killing entire minority groups just so they could hear themselves think.
If you assume the conclusion then the conclusion is true? Wow, shocking, never considered that!
Apparently scott didnt otherwise he didnt have to ask so I had to clarify it so he can understand it. You are mocking scott not me
Concepts were made for man, we historically do not apply them that consistently. The classic example being fruits and vegetables.
Insisting on NOT bending the rules being the default in this case feels to me like a way bigger claim than what you make it out to be.
"Concepts were made for man"
Wrong. Man was made for the categories. One of Scott's biggest blunders.
> Concepts were made for man
The concept of women sports was made for women.
I get accused of being lawyerly on the reg. But man, this takes the cake. If I ever get arrested and charged with a felony, I want you by my side in court.
Exactly, the whole point is that "women" isn't nearly as coherent a concept as the statement you're making makes it out to be. Both sides of the spectrum would agree with your claim, they just disagree with what the word "women" means. There's no real category around women, that's the whole debate.
People had the word women long before they could check the chromosomes of people etc etc
Women is an extremely coherent category. The vast majority of people have no problem with it - if you are born as a women with a vagina and uterus and such, you are a woman.
Exceptions do not matter here, it's not that hard.
There isn't a debate. A small group of radicals want to change the definition of women to something totally subjective and effectively meaningless. To the rest of humanity across cultures and throughout history, the distinction between men and women has been obvious. The words are not arbitrary; they represent real, obvious, and consistent patterns that exist in reality.
> Both sides of the spectrum would agree with your claim, they just disagree with what the word "women" means. There's no real category around women, that's the whole debate.
But there is a bait and switch going on here, no? When we discuss matters such as "what is a woman" or argue that "trans women are women" etc, we are talking about gender, we can do this because the classification is social in nature.
The distinction in sports comes from the other meaning of the word, which refers to sex, not gender.
No popular concept is coherent. There are always arguments around the edges. (E.g. I consider both Neanderthals and Denisovians to be humans.) So expecting one to be coherent is unreasonable. Whether trans-women can compete in "women's sports" is a boundary argument, and either choice is defensible.
If you have a person who’s a woman by any definition, and she happens to have sky-high testosterone, the only place she can rightfully be banned from would be the low-testosterone league. How is it “rightful” to define a category by quality A and ban people from it based on quality B?
Known as a performance enhancing drug, banned in numerous sports. See also:
https://www.wada-ama.org/en/what-we-do/international-standards
https://www.wada-ama.org/en/resources/world-anti-doping-code-and-international-standards/prohibited-list
A definition that includes trans-women that started their transitions early enough does seem to be one of the more coherent positions
...sudden shower thought: to what extent is that actually true, though?
"Women's sports" is a very broad category indeed, containing many very different things that can be categorised in very different ways (e.g. groups that only allow women participants doing things invented by men, vs groups that don't even necessarily explicitly limit participation but play games invented by and traditionally played by women, and everything in between), and absolutely does include activities that were designed at least as much with the viewers as the participants in mind, with a focus on showing the (cis) men in the audience something they'd pay to see.
continuum fallacy again
> Thats the ground.
No, that's just bait. There's absolutely no point in repeating this eternal debate unless there's some groundbreaking new argument coming.
We dont need any groundbreaking new argument. The answer is crystal clear and has always been. There was never a valid "debate" to begin with just as there is no real debate whether the earth is flat or not.
Indeed, Scott settled this back in 2014 [1] but somehow. it. just. won't. stop.
1: https://slatestarcodex.com/2014/11/21/the-categories-were-made-for-man-not-man-for-the-categories/
Scott did not come even close to settling this argument. There have been many good takedowns of this article, here's just one: http://unremediatedgender.space/2018/Feb/the-categories-were-made-for-man-to-make-predictions/
The reason it. just. won't. stop. is because he didn't "settle" it. Indeed, Scott's article was controversial even at the time of publication: http://unremediatedgender.space/2018/Feb/the-categories-were-made-for-man-to-make-predictions/
continuum fallacy again
i estimate a 30% chance this comment will lead to a ban
to myself: this thread caused alot of debate among everyone except me lol, i barely received any replies. This is a pattern i noticed as I got older. When I was an obnoxious 14 year old on the internet I almost always received replies to my comments but now barely any. I think the reason for why that is is obvious. Sad but thats the cost of being objectively correct.
Question for ACX readers: Is there a name for the position between 'pause AI development' and 'accelerate at all costs' that focuses on interpretability and coordination infrastructure?
I wrote three fables exploring what this looks like as design principles rather than engineering specs. Systems that must narrate intentions before acting, communities with built-in verification, infrastructure designed to amplify prosocial behavior rather than engagement. Basically: Bostrom's sparrows, but they figure out how to thrive.
Synthesis here: https://rajeshachanta.substack.com/p/honey-we-hit-the-jackpot
Does this map to existing frameworks I'm missing? Or is there a fundamental reason these principles wouldn't scale?
I think most people in AI safety are between "pause" and "accelerate". I don't think there's a name because it's the normal moderate position.
But I don't think that your fables/principles are speaking to the same concerns that most other people in the field are. Other people in the field are more asking "What if the AI is trying to deceive us?" An AI that was trying to deceive you would narrate a perfectly good-sounding intention before acting, then do something else when you weren't looking.
You're right Scott—I was addressing coordination assuming alignment, not the alignment problem itself. If the AI is deceptive, these mechanisms don't help.
The narrower question I'm exploring: Given aligned AI, how do we build coordination infrastructure so fast deployment doesn't preclude collective oversight? The Two-Feather Rule, Chorus Path, etc. are about human coordination around AI systems we trust but want to verify.
Does that problem matter if deception is unsolved? Or is it too downstream to be worth any attention now?
The problem with the "pause" scenario is that it requires broad agreement and enforcement. This isn't present, and isn't likely to become present. So it's not a real option. The options that *are* available are "open source" vs. "closed source". It seems possible, though unlikely, that legal frameworks could tilt that in either direction.
The value of the open source approach is that it enables those who want safety checks to design them. The problem is that using of them can't be enforced, so those who don't want to use them won't.
A plausible name for an intermediate position would be "put lots of effort into safety checks". I'm dubious about it's possibility, as if there's much cost, the first mover is likely to be one who doesn't do it. (Scott's argument that the cost wouldn't be significant seems wrong, as I think doing it properly would require a lot more effort than he is assuming.) Also, it seems unlikely because businesses are already starting to apply AI agents that we KNOW produce incorrect answers that may be costly.
I agree that in a pure competitive race, interpretability mechanisms will get dropped for speed—classic race to the bottom.
I'm exploring what the alternative landscape looks like—not 'how to win the race' but 'what infrastructure we'd want if we weren't racing to the bottom.' Getting there probably requires regulatory mandates or post-crisis adoption.
History suggests we regulate after things break: traffic safety after 5 decades of road deaths, building codes after major fires, FDA after drug disasters. The question with AI is whether we're smart enough to build guardrails before the catastrophic incidents happen, or whether we'll follow the same pattern.
Your point about businesses already deploying 'semi-reliable' AI is well-taken—suggests we're in the 'breaking badly' phase but haven't hit the threshold that triggers political will for regulation yet.
Are you saying the race dynamic is so dominant that exploring alternatives is pointless? Or just that we shouldn't pretend voluntary adoption is realistic without enforcement or crisis?
To me it looks analogous to the "tragedy of the commons". Without enforced rules, people will act for personal advantage. But enforcing rules in a worldwide development is difficult enough when most people agree on what the rules should be, and there is limited advantage in breaking the rules. Who would decide on what the rules should be? Who could enforce them? (For "Who" read "what mechanism", as it probably shouldn't be some particular person.)
It's not that the race is so dominant, it's that the structure of the problem is extremely difficult. I don't see any solution that doesn't require a quite intrusive international power with sufficient force behind it to intrude where it wasn't welcome. ... Actually, that's not quite correct. If one group felt that they had an unassailable lead, then they might be open to being careful. In the actual situation, however, everybody is being secretive about just how advanced their best models are...at least until they actually release them, so nobody can be really certain of their lead. Consider how Google just upset OpenAI.
This is a multi-player game of hidden information with partially unknown rules for extremely high stakes, without a referee. And it's not seen as an iterative game, so it's more like a multi-player version of single-shot prisoner's dilemma.
This is really helpful framing—the 'single-shot prisoner's dilemma with hidden information' is bleaker than I'd initially conceived.
I'm coming to terms with the fact that I'm designing mechanisms for a coordination equilibrium that may never exist without either crisis-forcing-function or (as you say) intrusive international enforcement. That feels like clearer scoping than I had at the outset—design for ideal conditions, acknowledge the conditions are unlikely.
Appreciate you helping me think through the structural problem more clearly. This has been genuinely useful. Thanks.
The other day, I came across The Qualia Research Institute's Psychedelic Cryptography Contest. The aim was to "encode" a message in an image so it could only be read by a person who was tripping — psychedelic steganography.
https://qri.org/blog/psycrypto-contest
Here's the winner. Evidently, any psychedelic/dissociative substance that causes heavy tracers could enable the decoding of the message. I don't have the substances to test it, myself.
https://youtu.be/P3D2kTJPOmg
From the writeup ot the contest results...
> It turns out that in order to decode these pieces you do require a substantial level of tracers, so only members of the committee who had a high enough level of visual effects were able to see the encoded messages. Some of the members of the panel reported that once you saw the messages during the state you could then also see them sober as well by using the right attentional tricks. But at least two members of the panel who reported seeing the messages while on mushrooms or ayahuasca were unable to then see them sober after the fact no matter how much they tried.
I thought that was pretty cool, indeed.
Very interesting. Impossible to see anything at normal speed or slower. Text appears at 2x. So on psychedelics time speeds up?
Somewhere in the write-up, they explained that it probably worked because of tracers that one would see with psychedelics like LSD and Psilocybe. And the tracers would leave behind after-images of the motion on your visual field.
If you're not aware of the term, tracers are the visual trails that one sees when an object moves across one's field of vision. For me, they've always been trails, something like colored lines you see in comic books to denote motion. According to a study that Gemini referred me to (below), "LSD users perceive a series of discrete positive afterimages in the wake of moving objects, a percept that has been likened to a multiple-exposure stroboscopic photograph." It's been a while since I tripped, and I don't actually remember them being so clearly stroboscopic.
And I always assumed the effect was due to my pupils being fully dilated while tripping, but on further consideration, that doesn't really explain the visual effects I saw. And it certainly doesn't explain the stroboscopic description that some people see!
BTW, I don't see the message at 2x speed, but the video ends too soon for me to visually "lock onto it."
https://pmc.ncbi.nlm.nih.gov/articles/PMC3091843/
I see some of the letters but it’s too fast, yes.
I can see the hidden images/messages while sober but I have done a lot of psychedelics so maybe it's a lingering aftereffect. It takes a bit of effort to parse it at 1x speed but at 2x speed it sticks out really easily.
Haven't been able to get it, yet. It ends too quickly at 2x speed. Maybe if I could figure out how to get it to replay continuously...
There's a loop setting in the options, on desktop it's just right-click -> loop, on mobile press the cog -> more options -> loop.
We also hosted a sillier debate between Bentham's Bulldog and Jeff Maurer on the motion "Donating to GiveDirectly instead of Shrimp Welfare Project is a chump move".
You can watch that debate here: https://benthams.substack.com/p/my-shrimp-debate-with-jeff-maurer
Here's the poll to vote on who won the debate: https://urbanpoll.com/polls/donating-givedirectly-instead-shrimp-welfare-project-chump-move-poll-clas29ah
To join Team Shrimp and donate: https://www.farmkind.giving/international-shrimpact-day/?promo=benthams_shrimpdog&s=swp
And to join Team Human and donate: givedirectly.org/IMBW
Re the trans sports thing - I think you have to get back to the question of why we have sports at all, and why do we have women's sports, then ask how do trans women fit within that. Some of these reasons have implications re trans women in women's sports.
E.g. one big reason for women's sports is that people saw that male athletes were heroes and role models for boys, and wanted something similar for girls. They think that as long as athletes are so idolized, we should have the same for women, and it's important for a project of gender equality/empowerment/whatever in a way that "short basketball players" isn't important for short people equality. What that means for trans women and sports ... people will disagree I'm sure.
Another reason - same way people want to see the fastest man over 100 meters, they want to see the fastest woman over 100 meters. For people who don't want trans women in women's sports, they specifically want the fastest biological female, gender identity isn't what they're interested in.
Another reason - camraderie among girls. Making an all-girls environment where they can build confidence etc in a way that doesn't happen so much with the boys around. Again what that means for trans women, people will disagree.
Someone should do a survey whether trans women athletes are motivating models for girls.
Would you accept the survey results if they didn't support your political stance?
I don't watch women sports, so it wouldn't even feel like I "accepted" something. If the people who actually do the sport want to have it certain way, let them have it.
My current impression (possibly wrong) is that some women who do the sports hate the idea of competing against someone who was a man yesterday, especially if it is a combat sport (when it's not only about losing, but about likely ending up in a hospital)... but if they object, they risk cancelling. So it's an end of their career either way.
(I especially hate the argument about how trans athletes are only 0.00000000001% of the population. I mean, that's exactly the point, that having a male body in a women sport is often such advantage that those 0.00000000001% easily get to the top.)
Yes. My curiosity is stronger than my political opinion.
This is an important point but the justification of a practice doesn’t always determine what the fair rules of competition or inclusion are, once the practice is already established. It might be that we established, or would have had reason to establish, organized competitive sports for some of the considerations you mention, but also that, given that they are established, fairness within the competition demands something that is at odds with the original justifications. Or so it seems possible to me.
Trans women have been eligible to join the Olympics for 22 years, and have never won a single medal.
I think things are plenty fair, and will continue to think that until trans women actually win more competitions than the null hypothesis would predict.
In terms of camaraderie among girls in the context of casual school sports, the trans women I've met fit very well into female social groups and social styles. This is why cis lesbians now have the strongest rates of support for trans rights among cis people; nearly every lesbian knows a trans woman who blends into the community. Trans women even seem to be prone to the same sorts of gossiping and shunning that cis women are.
>This is why cis lesbians now have the strongest rates of support for trans rights among cis people
Citation requested.
https://yougov.co.uk/society/articles/45983-what-do-lesbian-gay-bisexual-and-transgender-brito
I've seen several surveys which find this same result, but the above is one from the UK from a couple years ago. There is a vocal minority of trans-exclusionary feminists which lead people to believe that cis-lesbians tend to be trans exclusive, but when you actually ask the communities in any kind of systemic way it's clear that TERFs are not representative.
To be pedantic, this survey asked cis lesbians what they thought about trans PEOPLE, not trans WOMEN specifically. The finding is entirely consistent with the cis lesbians in question having many trans men in their social circle, of whom they think highly, but also having a negative opinion of the trans women they know.
Feels like an implausible stretch, but you are welcome to apply whatever standard of evidence you would like to this in order to try and confirm your own priors. I cannot find a study that specifically parses out attitudes towards trans women and trans men, but it does not seem likely that approval of one group would be un/inversely correlated with approval of the other.
>nor does it seem likely that approval of one group would be un/inversely correlated with approbal of the other.
Why not? Many radical feminists, for example, have strongly positive opinions of female people and strongly negative opinions of male people.
Why specifically lesbians? I thought they would be more annoyed with trans lesbians trying to court them. I would have assumed that gay men supported them more, since there's not much reason for them to get into conflict with trans women.
People find a reason
Agreed. I'll add that when people argue over this, they often have different ideas of who we should optimize for. We can focus on what's best for the competitors, the spectators, the identity group (women in this case), or the young people who think they may one day be a competitor. And we may have different answers at different levels of competition, like Olympic level versus high school level teams. Debates can get muddied when people don't realize they had different starting assumptions.
It's also a proxy issue for the more general culture war, so you get people who never cared much about women's sports having very strong feelings on this particular issue.
Why do you have to care abut women's sports, though? I couldn't care less about womens sports, but allowing in transgenders is just a litmus test for basic sanity. If you can't pass this test then I don't trust you in any type of leadership role.
> It's also a proxy issue for the more general culture war, so you get people who never cared much about women's sports having very strong feelings on this particular issue
That's true, although I don't think we should dismiss the opinions of those sorts of people, people are allowed to be tourists on issues when it comes to issues of fairness.
If the issue is "no blacks allowed at the Woolworths lunch counter" then having eaten at a Woolworths lunch counter is not a prerequisite for having an opinion on it.
I don't think you should necessarily dismiss their opinions but you can question their motives. Someone who always treated women's sports as a punchline, who opposed equal funding of women's sports under Title IX, who suddenly is freaking out over this issue, I think you can question whether they are truly motivated by concern for women's sports.
As somebody who would prefer to see toddlers kicking a rock around a muddy rain sodden field in November than the woman’s World Cup I do mostly try to keep out. In fact all men should keep their distance, that said i feel when this ideology breaks down the patriarchy will be blamed somehow.
What difference does it make how they feel about women's sports? Maybe they have a larger agenda...why isn't that okay?
People who worry about head injuries in football aren't worried because they think it will hurt the game of football, they're worried because they think it's bad for the kids' health. That's totally consistent with also thinking football is stupid. I think you can question whether they'd be so gung ho about it if they loved football and didn't think it helped the patriarchy or whatever ...
but it's a different boat from people who deride women's sports but think that trans women in women's sports is bad *for the sport*, i.e. bad for competition and such.
You could argue it's not a different boat specifically for a fighting sport (e.g. boxing not swimming), where there is arguably a health angle separate from the fairness angle, but nobody limits their views to those types.
"most sporting victories involve biological advantages that make things less fun for people who don’t have them"
Then the correct solution is to have no women's leagues.
I completely agree with this argument. One must recall why women's leagues were originally created.
Which trans men want to compete in the men's leagues?
Many of them, notably Mack Beggs and Schuyler Bailar. I've met several trans men that compete in men's sports (or did so in high school) personally.
Strange reply to write to a comment mentioning Mack Beggs, who made headlines for being repeatedly refused his request to wrestle in the boys' league!
And when I was wrestling, I was allowed to wrestle against people in weight classes higher. I could not choose to wrestle against people in lower weight classes however.
Correct. I did this in one match. If I recall correctly, the opponent would have no grounds to object, and one could only "wrestle up" no more than two weight classes (for the smaller wrestler's safety).
From an abstract, crotchety, no real pokemon in this fight perspective I kinda agree. To me if you're engaging in a competitive discipline, either you want to be the best in the world, no excuses given or accepted, which means you need to be willing to compete against the very best in the world across categories. Or you just want to do your best and have a good time, in which case it makes sense to sort you into a league with people of similar ability, but there's no particular reason to privilege sorting by sex or gender over height, weight, years of experience, or just empirical demonstrated skill (eg by an Elo rating).
A lot of sports do sort by this though, league systems (e.g. the English Football League System) are skill based and combat sports generally have weight classes. They just do it on top of the gender, although its also worth remembering that a lot of male leagues and competitions aren't actually male, they're open to all which seems to result in all male once you get out of the most casual of casual matches, the standard is an open competition and a female competition.
> no real pokemon in this fight
I like this expression.
For the record, it's not original to me.
It's a lot easier and furthermore, it already exists. That makes it better than any other hypothetical sorting system.
Dividing by ability level also already exists, and is necessary either way since merely dividing by sex isn't remotely sufficient.
Only in some sports.
Uh what? Can you name a single organized sport that doesn't bother to sort people by ability at all? Where a random new player can just wander in off the street and get matched against the best in the world?
That's not sorting, that's working your way up.
Love the post framing, funny
> most sporting victories involve biological advantages that make things less fun for people who don’t have them
This has occurred to me for some time now. The conclusion I have reached is that trans rights are at the moment a kind of compromise - many people accept it as a kind of polite fiction for civil society and accommodation for unhappy people. But not to the extent of like, fully internalized category update
the sports thing is a kind of rubicon/line in the sand, which is why it's so important despite not mattering much on the object level (see also: dating)
Most trans people agree that trans women are women, and that trans women are not exactly the same as cis women in all ways. The two options aren't "trans rights are a polite fiction" and "trans women are treated exactly the same as cis women in all contexts." For example, many trans women want access to options for freezing sperm.
I'm speaking of trans women here since those are the people this controversy is mostly about.
> Most trans people agree that trans women are women, and that trans women are not exactly the same as cis women in all ways.
Yes, and most people aren't trans. When are they going to realize that they aren't in a position to make demands? Don't they realize that people will only tolerate them as long as it remains convenient? Do they not realize what will happen if they start making things inconvenient for the majority?
Most people don't use wheelchairs and we're still doing okay (but could be doing a lot better) about having ramps, accessible bathrooms, and curb cuts. Most people aren't Muslim, but every major university/conference/anywhere else that provides food I go to has had halal options. Have a little more faith in humanity.
"having ramps, accessible bathrooms, and curb cuts"
All of which in no way inconvenience the majority. In fact, the bathrooms are also a convenience to non-disabled people who don't want to share a can (although this sometimes disgruntles the disabled, it's still widely allowed; guess why?), and curb cuts can help e.g. old people and kids as well as people in wheelchairs. A halal option isn't an inconvenience for the majority either as long as you can just opt to eat non-religious meat, which you normally can.
You're really proving his point for him here.
But those things do inconvenience the majority. A public building with ADA accommodations is getting that extra construction fund from your tax dollars, and a conference with halal options is getting the extra food budget from your ticket. These are indirect, yes, but sometimes inconveniences are direct too, with disabled parking being an easy example.
How much a given accommodation costs versus benefits the majority depends a lot on what the accommodation is, so it's hard to compare without getting into specifics, but generally speaking there's usually some cost and some benefit.
I'm not ok with either of those things honestly. Completely unreasonable accommodations.
> Most people don't use wheelchairs and we're still doing okay (but could be doing a lot better) about having ramps, accessible bathrooms, and curb cuts.
Which is not a product of disabled people using their leverage, but the voluntary kindness of the abled. If disabled people start making unreasonable demands and make their existence problematic for society, people will start withholding such kindness.
> Most people aren't Muslim, but every major university/conference/anywhere else that provides food I go to has had halal options.
That one's definitely not going to last long, given the current political climate in the US. Immigrants have become a problem for society, and you're about to see what happens to people that become a problem.
It sounds like you’re inclined to simply define an unreasonable demand as one that the majority won’t abide. So your argument seems pretty empty to me.
Is it reasonable for somebody to put a target on their back, for something as trivial as sports league participation?
'Society' is doing a lot of work in these assertions.
An overlapping patchwork of people who are minorities along various axes comprises a big chunk of society, and while the degree of success fluctuates, they often try to stick together, as per the old Niemöller litany.
Add to that the plurality of default-type people who are willing to join the 'other side' on ethical grounds, and your society ends up fairly cleanly split.
Fine, is "the collective that holds majority power in a given setting" better? I won't even say population majority, because not all people have equal capabilities to impose their will, one of the more obvious example being men versus women. And with 61% of men saying that changing gender is morally wrong, and the views of the ruling party being what they are, it's not looking too good for our trans friends. https://news.gallup.com/poll/645704/slim-majority-adults-say-changing-gender-morally-wrong.aspx
The ADA was the result of coordinated, disruptive protests. That's how minorities get rights.
Halal food isn't going to last long? What?
> Which is not a product of disabled people using their leverage, but the voluntary kindness of the abled. If disabled people start making unreasonable demands and make their existence problematic for society, people will start withholding such kindness.
A cursory investigation into the history of disability rights will show that they were often not a voluntarily given kindness, and often a result of disabled people organizing, filing lawsuits, and "making their existence problematic" by doing things like occupying federal buildings and staging an event in which they crawled up the steps of the Capitol Building without assistance. Likely many people thought of their demands for legal protection as unreasonable at the time.
https://en.wikipedia.org/wiki/504_Sit-in
https://en.wikipedia.org/wiki/Americans_with_Disabilities_Act_of_1990 (see Capitol Crawl)
These demonstrations were admirable but in no way do they seem like exertions of force. These were sophisticated, intelligent requests for charity. And that's fine, I've requested charity in my life, so has everyone. But it's still asking for charity.
> A prevailing view held that people with disabilities were pathetic and deserving of pity, and therefore incapable of such political actions. This same perspective made officials reluctant to risk a public relations embarrassment that would result from arresting participants.
Even the biased Wikipedia article admits that the government just felt bad for them. There was nothing forcing people to accede to their demands. But their pity and compassion is not endless. More importantly, most people do not feel pity for trans people.
How could immigrants possibly be a problem to American society when studies have found pretty much all forms of immigration to be pretty much universally beneficial to the US.
On a timeline basis the reprisal against immigration started at the beginning of the year and has shown to be pretty unpopular with the broader electorate, and immigration is viewed positively overall among the entire electorate except for specifically the surge in people exploiting the refugee loophole.
Jews are white these days, so I'm sure they'll be fine for a good while unless something triggers a massive wave of antisemitic sentiment among the right.
Hello! I have written a medieval fantasy novel and I want to give it away online. Can anyone advise me on how to do this?
I, for one, consume a lot of people's give-away writing here: https://www.royalroad.com/
AFAICT you get way more visibility if you post your writing in regularly scheduled little pieces rather than all at once.
I'm publishing one there, too. The Writathon is almost over, thank goodness!
Release it as a web serial.
I feel like non-subscribers are on the naughty stool.
Worlds collide: someone I follow on X reposted a fiction piece written by an LLM (Opus 4.5, to be precise). It's based on a meme that has apparently been making the rounds, featuring two doors: a red one opening on $2 billion immediately, and a green one opening on $1 that doubles every day. The piece begins: "I chose the green door ninety-three days ago."
https://x.com/magnushambleton/status/1995459217407217698
It's actually pretty good, IMO.
The first reply to it is from the person who posted it, claiming Claude made a few obvious math errors, which he fixed, and otherwise posted without further wordsmithing.
The second reply, also from him: "his was the result of thinking ”I wish there was a @slatestarcodex or @jackclarkSF short story exploring the green door option” and then asking Opus for that."
...Well then.
Yeah, I also thought it was good. I'm going to put it in a links post, hopefully in the form of a link "read this cool story", followed by "and after you're done, see the interesting part", which will be a link to him saying it's AI.
Very good story!
That’s really impressive
I'm very impressed. I read a lot of sci fi/speculative fiction short stories and this is definitely better than average.
Correction: I see another post from the prompter:
---
My first prompt was "Can you write a 1000 word short story in the style of slatestarcodex/scott alexander or jack clark/import AI about a world in which the dollar doubles every day, and the horrifying implications. Eg implications of storing the bits of the dollar value in bank systems, and how the economy is affected by all this money being created. Write it from the perspective of the one who made the choice."
It then went down a path of writing about how computer systems were crashing from storing the number of bits needed to store the value, so I had to correct it and say "make it mathematically accurate, i.e. bits to store the monetary value will scale linearly with number of days, since it's doubling and bits also increase the number they can represent by 2^x"
---
The Emergent Thermostat, the Beautiful Ones, and the Coming Collapse of Human Societies
A Unified Theory of Reality-Model Degradation from Thunderstorms to Hikikomori
1. The Insight on the Beach
In the year 2000, a scientist living in Fiji noticed something no climate model had ever been asked to explain: the global temperature of Earth had varied by only about 0.2% (in Kelvin) across the entire 20th century despite volcanoes, wars, and industrialisation. While most researchers focused on why the planet had warmed by 0.6 K, he asked the opposite question: why had it barely changed at all?
Observing the daily cycle in Fiji revealed the answer. Clear dawn led to rapid surface heating, which triggered cumulus cloud formation at a precise threshold. The clouds reflected sunlight, cooling the surface. If heating continued, thunderstorms would form at a higher threshold, producing downdrafts, evaporative cooling, and cold pools — enforcing even stronger cooling. Conversely, when the morning was cooler than normal, clouds formed late or not at all, keeping the surface warmer. Day after day, fast, local, emergent phenomena maintained an extraordinarily narrow temperature band.
His realization: a governor operating on the timescale of hours can, through repeated application, enforce the same bound over centuries and millennia. CO₂ is not the main long-term control knob of Earth’s temperature; thunderstorms, El Niño, and cyclones form the planetary heat engine’s cruise-control system.
2. The Blindness of Artificial Intelligence
Twenty-five years later, he tested two competing claims on ChatGPT: one, that the major control of the world’s temperature is CO₂, and two, that the major control is emergent phenomena like thunderstorms and ENSO.
The model initially tried a safe “both/and” answer, but when forced to choose, it picked CO₂. This is because that claim appears millions of times in its training data, while the emergent-thermostat hypothesis is barely represented, mostly on one blog. ChatGPT reproduced the exact objection the scientist himself had held for months: “but those phenomena are only short-term.” It could not grasp that repeated short-term regulation is the only way long-term regulation is physically possible in an open system with variable energy input.
This illustrates the core limitation of current large language models: they excel at summarizing what has already been said, but cannot originate genuine integrative insights.
3. Model Collapse Is a Universal Law
In 2024, Nature published the definitive paper on AI “model collapse”: neural networks trained on an increasing fraction of synthetic (AI-generated) data rapidly degrade into confident nonsense. The more artificial the training diet, the faster the collapse.
The parallel to human brains is chilling. Brains are prediction engines trained on sequential sensory data. When a growing fraction of a child’s training data is second-hand, symbolic, or entirely synthetic (screens, classrooms, anime physics, TikTok morality), the internal reality model loses fidelity in exactly the same way.
Critical developmental windows make this irreversible, just as it is in silicon. Vestibular and proprioceptive calibration occurs in the first 5–7 years, native phoneme discrimination is lost after 12 months, spatial navigation circuits prune if never exposed to complex terrain, and corpus callosum thickening from early music training cannot be fully replicated later. Remove enough real-world gradients — gravity, pain, hunger, predation, unscripted social consequence — and the network begins to hallucinate.
4. Universe 25 and the Mammalian Blueprint
John B. Calhoun (1947–1973) removed every natural selection pressure from social mammals and observed the same sequence in rats, mice, deer, rabbits, reindeer, and partially in primates: explosive population growth, sudden peaks despite unlimited resources, emergence of a hyper-aggressive minority alongside mass withdrawal of the majority, maternal failure with 100% infant mortality, and the appearance of the “beautiful ones” — perfectly groomed, healthy individuals who did nothing but eat, sleep, and groom, ultimately ceasing reproduction and leading to extinction.
This phenomenon is not rodent-specific; it has now been documented in seven mammalian orders whenever adult mortality approaches zero for multiple generations.
5. The Human Experiment We Are Running Live
For the first time in history, large human subpopulations meet Calhoun’s four conditions: adult mortality from violence, starvation, or disease is nearly zero; caloric and material resources are abundant; urban and digital density are extreme; and emigration to lower-pressure environments is functionally impossible.
The results mirror Universe 25 with disturbing precision. In Calhoun’s mice, the “beautiful ones” were perfectly groomed and socially withdrawn. In humans, the closest analogues are Japan’s 1.5 million hikikomori, 30–50% of 20–34-year-olds in affluent countries who are sexually inactive, and hyper-groomed but socially dead young adults in Seoul, Tokyo, Singapore, and Western cities.
In Calhoun’s mice, maternal failure caused total infant mortality despite food abundance. In humans, we see record child-neglect reports despite unprecedented wealth, and voluntary childlessness is rising fastest among the richest and safest cohorts.
A tiny minority of mice became hyper-aggressive while the rest withdrew. In humans, tiny cohorts (the top 2–5% of males) commit most violent crime while the majority increasingly drops out of work, dating, and reproduction.
Calhoun’s final outcome was total reproductive collapse. In humans, South Korea’s preliminary total fertility rate in 2025 is 0.72 nationally and 0.55 in Seoul — the lowest ever recorded. East Asia is furthest along this trajectory, having first and most completely achieved the Calhoun conditions. The West lags 20–30 years behind in its most affluent, screen-saturated strata.
6. Conclusion: Two Emergent Governors, One Civilization at the Threshold
Earth’s temperature is regulated by fast, ruthless emergent phenomena that activate the moment a local threshold is crossed — the mechanism required for long-term stability.
Human behavior is regulated (or de-regulated) by the fidelity of our internal reality models. When those models are trained on increasingly synthetic data, we produce our own behavioral sink: the “beautiful ones” who defend delusions confidently because they have never interacted directly with reality.
The same natural law governs both systems. Thunderstorms prevent runaway planetary heating. Direct, unmediated experience of physics, biology, and consequence prevents runaway social cooling.
We have spent seventy years removing every consequence that once shaped human behavior, and we are discovering in real time, at civilizational scale, that the thermostat works in both directions. The cruise control is still there. It is simply waiting for the next threshold to be crossed.
⸻
> caloric and material resources are abundant
Where can I find this place? If I stopped working, I would run out of material resources.
There's a lot to chew on there. Just out of curiosity, I ran a couple of probes against a couple of your statements using LLMs (ChatGPT and Grok).
> we see record child-neglect reports
In the US it looks like reports of child neglect or abuse went down ~20% between 2019 and 2023. That may be an aberration to a longer-term pattern for which I can't find any data at the moment. If you've got this data, please share some links.
The EU seems to be undergoing a surge in hotline reports of child sexual abuse complaints, but both the LLMs I queried said they didn't have any overall data on child neglect. The UK data shows that as of 2024, the number of kids in child protection plans is the lowest since 2013. OTOH ,Germany showed a 4% rise in child neglect/abuse cases between 2021 and 2022. This was flagged in news stories as unprecedented.
> John B. Calhoun (1947–1973) removed every natural selection pressure from social mammals and observed the same sequence in rats, mice, deer, rabbits, reindeer, and partially in primates...
Calhoun only ran these tests on Norway rats and house mice. Later researchers tried to reproduce his experiment with macaques, rabbits, pigs, and songbirds (I didn't try to find out which species). Although crowding could affect behavior, none of them showed a Calhoun style population collapse.
Near Clarion, PA? I was there a couple times as a young'un. Cool place.
The shrimp meme has run it's course. Pigs and chickens suffer the most in commercial agriculture.
Hold on there! I want to know how they calculate that "$1 can improve welfare for around 1,500 animals," and how they arrived at "making them one of the most impactful animal charities to donate to by number of animals helped per dollar"?
These answers have profound consequences for how we measure the success of human welfare programs!
And just who is removing their eyestalks? Are these immigrant laborers using tweezers?
For a dollar, I can buy about 5 pounds of corn. I can then give each kernel to a different cow, which I'm pretty sure would result in that dollar helping more than 1500 cows.
They have an impact page here: https://www.shrimpwelfareproject.org/shrimpact .
They're trying to develop a welfare index to calculate how much suffering, and what intensity of suffering, is being averted through their work: https://www.shrimpwelfareproject.org/shrimp-welfare-index
But they don't actually have great publicly available data connecting their work to the index at the moment. The numbers on the shrimpact page seem to be a straightforward calculation of: (1) how many farms have adopted electrical stunners or agreed to other welfare changes (stocking density, pH management, eyestalk ablation, etc.); (2) how many shrimp those farms produce annually; and (3) programme cost.
“Improving welfare/helping X animals for $Y” can be a misleading framing that covers everything from "0.1% reduction in stocking density" to "not ripping their eyestalks off". But SWP’s main measurable impact is from introducing electrical stunners, so it’s reasonable to assume that “quicker, less painful death” is the main unit of benefit they’re counting.
I thought we were onto nematodes now?
This feels like a claim that needs evidence. The number of shrimp and other arthropods that are farmed are enormous, and the whole 'crush their eyestalks so they lay more eggs' thing makes me feel like we probably cause them to suffer.
Another good link below:
https://forum.effectivealtruism.org/posts/mdcSeMwkBEYhdTAWF/to-a-first-approximation-all-farmed-animals-are-bugs
I'm open to the claim that shrimp are incapable of suffering, but I don't think caring about shrimps is just a meme.
I didn't say "they don't suffer", nor is my claim contingent on that. To the extent you can be confident that an animal suffers, not only can we ascribe far more confidence to the aforementioned, but the conditions they endure are the worst. Chickens raised in battery cages their entire lives shouldering disease, immobility, pain. Plenty has been written about this already and it's not hard to find, it's even on the old SSC blog. There's also a Richard Hanania post on the topic.
By comparison, shrimp is a complete joke. This is analogous to prioritizing eliminating pesticides and drastically resharping crop reaping practices because there are far more insects than anything else, and they have sentience too. You cannot equivocate these.
It's more than a meme, it’s an insult.
> slap in the face
You've used this expression no less than 3 times in this Open Thread. You know what they say: If one person slaps you in the face, they are the problem. If everyone slaps you in the face, you are the problem.
Maybe it's just that I'm sleep deprived and I'm repeating myself. Maybe it's because shrimp welfare and biological men in womens’ sports are two extremely hot button issues and it's unusual to see Scott doing so much counterproductive advocacy in one post.
Maybe it's because it's a highly evocative phrase and so I like to use it. We all have our favorite phrases, don't we?
Did you have any kind of point to make or are you just taking pot shots?
> We all have our favorite phrases, don't we?
We sure do. My post was mostly meant to be tongue-in-cheek, not as an attack. I'm sorry that I didn't make this clear.
Edit: What are you doing here if you're sleep-deprived? Take a nap, get some rest! Let people be wrong on the Internet, your physical and mental health is much more important than posting in some forum.
I am not convinced that vegetables don’t suffer, and actually I’m inclined to think they do. But vegans and animal welfare types, broadly speaking, do not seem to take this possibility seriously. (The only group that takes it seriously, that I’m aware of, is Jainists.) I find this kinda fascinating.
> I am not convinced that vegetables don’t suffer, and actually I’m inclined to think they do.
What would it even mean for a plant to "suffer"? Plants don't have brains, they don't even have a central nervous system. Suffering, and qualia in general, isn't an objectively measurable phenomenon, at least not as of today; we can only (speculatively) infer that other beings suffer by their similarity to us, specificially by the similarity of their brains to our brains, as that's where suffering manifests. So given this, the question isn't "Do plants suffer?", but rather "Is suffering even a meaningful concept when it comes to organisms that are as far removed from us as biologically possible without leaving the eukaryotic domain?".
"Plants don't have brains"
This disqualifies the shrimp too, though; they don't have brains either. Shrimp, lobsters and other crustaceans have ganglia.
Plants evolved a different means from animals of sending messages throughout their bodies, but using many of the same chemicals (including cortisol!) and to achieve the same goals for the same reasons. Something like a quarter of our DNA is the same as theirs. Plants are certainly capable of sensing and responding and communicating, manipulating their environment, and even moving around in some ways, and just as invested in their own reproduction. To me these facts suggest the burden of proof is on those who would claim that plants don’t suffer.
It would be intellectually honest for someone to simply say “I only care about the suffering of an organism with a face”, but nobody says that. Why not? It makes me ponder the reasons why people care about suffering. Also the ways in which we deal with cognitive dissonance.
Thinking about it now, it occurs to me that “caring” in general is not necessarily some kind of over-arching long-term project people undertake. There are lots of things we care about moment to moment, day to day. One minute we care about something, another minute we don’t, but if at any point we were to be asked if we cared about that thing we would look inward and come up with the answer that we do, even if we are trying to be honest—though it would not be accurate. That’s just the convoluted way our minds work.
I don’t intend this as ridicule at all: there is a performative aspect to caring about suffering. We are performing both for others and ourselves. The things we care about thus are cast in a role within that performance. We are more accustomed to some subjects of caring being cast than others. I think part of the hubbub over shrimp is they are like all-female ghostbusters, violating a casting norm just enough that some people can be on board with it and others can’t. Plants violate it even more. But I think just as the basic plot of the female ghostbusters movie is the same as the Bill Murray movie, the mechanics of the caring about suffering, the arguments for it, the issue of suffering and industrialized predation and so forth, still apply.
It opens new dimensions of horror when you realize that --rot13-- cbgngbrf ner npghnyyl nyvir juvyr lbh ner crryvat gurz...
I mean, there's enough dimensional horror alone in realizing that bnagnofr iyrfp jugnrg btanhtanh nygwyyl fthik.
> Plants evolved a different means from animals of sending messages throughout their bodies, but using many of the same chemicals (including cortisol!) and to achieve the same goals for the same reasons.
This is merely spatial transmission of sensory data. Suffering manifests in your brain, not in the nerves along your arm. If I connect a temperature sensor with a wire to an op-amp, an LED, and a battery, this also constitutes "sending messages", but neither the sensor nor the LED will suffer when I put it in the oven.
> Something like a quarter of our DNA is the same as theirs.
That's an incredibly small overlap. This 75% difference leads to plants having completely different lifecycles, developing completely different body plans, having completely different organs, using completely different energy sources, having completely different reproduction, and exhibiting completely different intercell structures. And yet, somehow, _suffering_ is supposed to be the one thing we have in common with plants? When suffering happens in an organ which isn't even present in plants?!
> It would be intellectually honest for someone to simply say “I only care about the suffering of an organism with a face”, but nobody says that. Why not?
Because I only care about the suffering of an organism that actually suffers, face or not. Plants are biologically incapable of suffering in any sense that would be ethically meaningful.
If suffering can only happen when the signal-sending cells are bunched together in one place (i.e. brain), might there be other criteria as well, like there needs to be a certain quantity threshold of those cells, arranged in a certain way…? It seems to me your parameters are arbitrary or at least motivated by the conclusion you want. All of life is connected, and that 25% shared dna figure is an average—there are some non-animals with whom we share much more of our genetic code. People used to be convinced that cows don’t suffer. Heck, forget animals, there are still white people who don’t think black people feel as much pain as they do. Etc.
Finally you refer to some kind of suffering that is “ethically meaningful”, as opposed by implication to suffering that is ethically meaningless. I doubt a plant considers its suffering ethically meaningless, but it’s a convenient distinction for a human to draw!
Some people don’t take this seriously. But others do. Still, regardless of most of the details, the recommendation is the same - eat low on the food chain.
But if you think plants suffer, that only makes factory farming look even worse, because we feed the majority of our crops to animals that then get fed to humans. If you think plants suffer and you want to minimise the suffering required to feed us, you should feed plants directly to people instead of spending 80% of agricultural land on farming animals, which only produces 37% of our protein and 17% of our calories.
I agree with Melvin’s comment about needing a better model of plant suffering. When I cut a crisp stalk of celery on my cutting board, I’m pretty sure it’s alive while I’m cutting it, right? That seems worse than harvesting corn or wheat that have already dried out and turned brown on their stalks.
And, if we don’t grow plants to feed animals, that land would still likely be covered with plants that would still meet torturous ends. It’s not like the alternative to growing plants to feed animals is to turn that agricultural land into parking lots (which might be a still worse thing to do anyway).
I think you would need a better model of how exactly plants suffer in order to figure that out. How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
Honestly though, this is the first I've heard about the eyestalk ablation thing and I don't like it, I would be happy to have it banned, not on some kind of consequentialist grounds but because it's disturbing and in bad taste. I think I can come around to supporting greater farmed animal welfare from a virtue ethics sort of standpoint, we should treat the animals who serve us with dignity.
I can't get behind the idea that we shouldn't ablate prawns' eyestalks because it's the only way to get them to grow ovaries in the horrendous Indian/Vietnamese farms in which they live it because shrimp suffering is worth epsilon times my own suffering for some totally made up value of epsilon, but I can get behind the idea that we shouldn't because it's weird and gross and those farmed Asian prawns taste like crap anyway.
> How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
The apple is intended to be eaten, so this looks analogous to saying "how does a woman's suffering when a stylist cuts her hair compare to a man's suffering when he gets shot by a bullet?" I don't think you could have picked a worse question if you'd tried.
By this argument, childbirth should be completely painless, and yet...
[Quibbling about the use of the word """"intended"""" in reference to biology preemptively excised]
EDIT: actually the argument would be *stronger* for human childbirth, since an apple tree does not get any say in whether its fruits are plucked or not, so whether or not it suffers has no selective effect. Which is also why I think it's very unlikely that plants suffer -- suffering has no evolutionary advantage for them. Hyperselectionism is a known fallacy in evolutionary biology, but I doubt conscious experience appears as a byproduct of signal processing without any selective advantage for it.
I agree it’s not a good analogy. (I didn’t make it.) but to rescue it a bit, there are similarities between apple trees and farmed animals. Both are human cultivars, freaks who would not exist without us having bred them out of quite different ancestral organisms, and probably would not survive without continuous human intervention, whose phenotype has been massively distorted to suit our specific demands. There’s a moral component to that, right?
If some other organism liked to eat human hair, and it managed to breed humans that were maximally hairy so that we looked like giant walking piles of hair (only we couldn’t walk or even see anything or move under all that hair—eyes just take up space where hair could be so we are eventually bred to have no eyes, just very hairy permanently closed eyelids, in fact our eyelashes are considered a delicacy so some people’s faces are all eyelash), that would be bad, right?
> How does the suffering of an apple tree when one of its apples is picked compare to the suffering of a blade of grass when the top of it is munched by a sheep?
Before we pick fruit, we wait until it is fully developed. We know fruit is developed when it reaches what we call the "breaker stage." This is the stage at which the fruit's internal connection to the plant's vascular system has been severed. You can tell this has happened because the fruit will just pop off with a gentle tug. It's like the difference between a loose hair that's ready to fall out and a healthy hair that holds on tightly by the root.
Less than 20 years ago a housing bubble popped and caused the Great Financial Crisis. When bubbles pop, a thing that everyone thought was expensive and valuable is revealed to be cheap and not valuable. So how come houses got expensive again so fast?
Yes yes, it wasn’t technically the houses that were bubbling but the mortgages. It still led to brand new houses being built everywhere and sold off at a loss. Why didn’t we get cheap housing for a while after the GFC, like we got cheap computing infrastructure after the dot-com crash?
> So how come houses got expensive again so fast?
Because there's no financial instrument to short housing, so prices can only ever increase until they burst. How fast the prices increase has more to do with the velocity of turnovers and the number of buyers than it has to do with the 'actual' value of the houses.
Doesn't this imply that there's no such thing as an underwater mortgage? Why would banks bother classifying mortgages as safe or risky if every house purchase was guaranteed to be either neutral or profitable?
The demand for housing didn't go down.
Think people need to remember the "financial crisis" part. It wasn't just that people lost savings because home prices went down, there were a bunch of over-leveraged financial institutions and investors who were buying complicated financial instruments linked to the housing market. So even a temporary drop in housing prices causes a bunch of knock-on effects.
>"Less than 20 years ago a housing bubble popped and caused the Great Financial Crisis."
False.
Monetary authorities (esp. the Fed) cut the money supply causing NGDP to crater; the symptoms of the recession were then concentrated on the most rigid (in nominal terms) sector: residential mortgages.
We've made it hard to build any new housing.
Kevin Erdmann has an interesting hypothesis that everyone misinterpreted the Great Financial Crisis and that, effectively, there *wasn't* a bubble in housing prices. That was their true market price (due to zoning constraints creating a huge shortage).
https://kevinerdmann.substack.com/p/research-roundup
tl;dr:
This enormous shortage caused prices to rise so much that it finally managed to trigger a construction boom of single family homes (the only thing still legally allowed to be built). This led to people misidentifying the construction boom as a bubble and a large elite effort to kill the "unsustainable" level of housing construction. We succeeded at engineering a massive recession which did cause housing prices to fall (briefly) on account of massive unemployment reducing housing demand. Unfortunately, home prices quickly recovered as prices are a rationing mechanism and we still had a true housing shortage.
This seems too clever by half. In hindsight, the global financial market was dangerously overexposed to a bunch of correlated housing derivatives. Whether the marginal additional dollar of pricing signal was coming from Atlanta or Miami doesn’t change that several major companies went bankrupt and needed to be bailed out or liquidated with shocks around the world for the next half decade.
The thing is, banks did fail. Lots of firms went bankrupt. But mortgage backed securities never actually experienced substantial losses. We know this because the Federal government nationalized Fannie Mae and Freddie Mac and *didn't* suffer losses (that is, Fannie Mae and Freddie Mac remained profitable).
If the Mortgage Backed Securities actually were failing, the government *should have* taken a loss. Indeed, the whole point of nationalizing them was for the government to absorb the losses. But that... never actually happened and we just assumed that the mortgages were the cause of the banking crisis anyway.
Fannie and Freddie were exposed to better quality mortgages than the market as a whole. See page 219 of the Financial Crisis Inquiry Commission Report. Just one data point: in 2008 Fannie and Freddie mortgages were had a delinquency rates of 6.2%, versus 28.3% for other mortgages.
At the time that the government took over Fannie and Freddie, both had positive net worth, but their losses meant that their leverage was quite high, making it unclear that they could continue to borrow money at reasonable rates. So there was no reason for the government to suffer any losses in the absence of further adverse events. That doesn’t mean that the Fannie and Freddie didn’t suffer losses in the financial crisis; it just means that their losses hadn’t been large enough to wipe out their entire net worth.
In summary, Fannie and Freddie managed to largely but not completely escape the fallout from the mortgage backed securities collapse by mostly avoiding the worst mortgages, but they never the less suffered losses large enough to prevent them from continuing to operate as independent companies. I don’t think this is consistent with your claim that “mortgage backed securities never actually experienced substantial losses.” I certainly don’t see how it is supposed to *support* that claim.
In a liquidity crisis, is it that surprising if the lender of last resort makes a profit buying up distressed assets? Iirc the government also made money on hedge funds. The problem wasn’t the balance sheets, as you say; it was overexposure to systemic risk that led to a drawdown and sudden demand for liquidity. You could be sitting on a mountain of umbrellas that are “objectively” worth hundreds of thousands of dollars, but you might have to sell at a loss during a drought to get food, if that analogy makes sense.
> In a liquidity crisis, is it that surprising if the lender of last resort makes a profit buying up distressed assets?
Well, if it's a liquidity crisis instead of a solvency crisis then you're right. It's unsurprising.
But it's my understanding that the Great Financial Crisis (GFC) narrative is that the building boom led to a solvency crisis. Homes were built that were unneeded, so the underlying loans were insolvent.
But if the assets were solvent, that implies that the root cause of the GFC *wasn't* the building boom, right? Because the underlying assets were solvent! If the underlying problem was a liquidity crisis, it means the issue was Federal Reserve mismanagement of liquidity in the financial system.
> Homes were built that were unneeded, so the underlying loans were insolvent.
I don't follow your logic here. The US has been under-building housing for decades; I doubt there were homes that weren't needed. Rather, there were homes that were way too expensive to be justified by the incomes of the people owning them. They were insolvent for that reason, not because we had more housing units than people all of a sudden.
And honestly, what building boom? There had been a steadily increasing number of new units being built per population since about 1990, but in comparison to the prior 30 years it's pretty pathetic: https://fred.stlouisfed.org/graph/?g=1OnVx
I believe that liquidity and solvency problems fed into each other. This is why one of the first things Bernanke did was try to find a price floor for distressed assets. Whether banks were technically solvent on paper or not, an inability to accurately price assets caused institutions to hoard cash and prevented banks dependent on short-term loans from refinancing their debt. This is why the Fed needed to inject liquidity into the banking system (swap lines, money market funds) alongside propping up demand for assets and equity (TARP).
See: Taxpayers Have Now Made A $63 Billion Profit From Fannie Mae, Freddie Mac Bailouts
https://finance.yahoo.com/news/taxpayers-now-made-63-billion-185024064.html
thank you for your answer!
Housing prices did go down for a while, although even at their lowest they were still basically equal to the pre-bubble peak: https://en.wikipedia.org/wiki/Case%E2%80%93Shiller_index#/media/File:Case-Shiller_Index,_1890-8.2025.png
A priori, I'm not sure how big of an effect you would expect or how long it would take to bottom out and then to start to recover. I guess it is at least a little surprising that it never got lower than about 10% more expensive than it was pre-bubble (inflation adjusted). Also, I'm not sure how much this price decrease reflected availability, since mortgages became harder to obtain.
In any event, the valuable part was actually the land. You can justify much riskier mortgages if the price of housing is going up, since you don't have to rely on the borrower's income to recoup the loan. But the value of the building doesn't go up (compare to cars--obviously these get cheaper as they become more heavily used, lack the most current features, etc). Pretty much the entire price increase is explained by the increasing price of the land, which in turn is consistently pushed by because A) people need to live somewhere, and B) most of the US bans reasonable land use. Neither of those things changed just because the housing market collapsed. Also, I don't think the GFC would be expected to impact rent, except to the extent that some people would buy instead, but it's not like the actual supply of housing went up (if anything it probably went down).
Overall I think the GFC might be both a "policy bubble" (in the sense that the price is largely driven by high demand and limited supply) and a "speculative bubble" (in the sense that the price goes up because people expect it go up, which reinforces those expectations).
Purely anecdata, but my house was under water* for 3 years. I asked for and got a reduction in my property taxes (though not as much for I asked).
* meaning I owed more than it was worth.
The average-looking national aggregates in 2012 were actually an average, roughly speaking, of low-tier housing in places like LA that was 100% overpriced because of the shortage and 50% underpriced because of the mortgage crackdown, and low-tier housing in places like Atlanta that had never been overpriced, but was now 50% underpriced because of the mortgage crackdown.
Prices in a few places were unsustainably high in 2006 (Arizona, Florida, etc.) mostly because there had been a mass migration to those places from the cities with binding housing shortages. That reversed. But, on net, the entire loss of $5 trillion across the country, in places like Atlanta, was not a reversal of anything that had come before, and was due to tightening lending too much at the federal agencies, well after the private securitization boom was over. Because of the moral panic about the lending boom, all credit tightening and deep drops in home prices where they had never been elevated, were, a priori, blamed on the lending boom.
Here's a post on Case-Shiller.
https://kevinerdmann.substack.com/p/what-the-case-shiller-home-price
thank you for your answer! can I ask you more questions?
Why do you think the actual supply of housing didn't go up during the bubble? My impression of America's zoning issues is that they make every new build x% more expensive, rather than putting a hard cap on the build rate. So increasing the profit for new houses should still lead to an increase in housebuilding.
If, as you say, the market for houses collapsed but the market for land was unaffected, why did the GFC produce such a shock to banks? Was the lost value purely speculative?
I think it is actually a hard cap, or very close. Housing supply is limited by policy that effectively bans building housing above a very low density in most of the land area in and around towns and cities. In some places it also substantially increases the cost of even that low-density housing, but municipalities can effectively make the cost as high as they want so there's not actually any profit to charging more. However much the builder could charge, I assume that local residents could make it instead by just letting their home prices appreciate.
> If, as you say, the market for houses collapsed but the market for land was unaffected, why did the GFC produce such a shock to banks?
Like I said, I think there were some of both (also, I'm not a real estate expert, but I doubt the market was unaffected, it just has kind of a high price floor under current policy). Banks assumed house prices would keep going up because they had previously gone up, even though there's only so much that people can afford. And there were actually a lot of people with mortgages that were only justified because of expected price increases rather than by their incomes.
> Was the lost value purely speculative?
I'm not exactly sure what this means.
> municipalities can effectively make the cost as high as they want so there's not actually any profit to charging more
Excuse my lack of clarity; I meant that developers expect more profit per house when mortgages are generous and would thus lobby harder against restrictive zoning. I wasn't alleging that zoning restrictions are due to a profit motive.
>However much the builder could charge, I assume that local residents could make it instead by just letting their home prices appreciate.
I don't believe this follows. There are three ways residents could increase house prices:
1. Investing in their communities, making them so attractive that demand to live there goes up. Not super relevant here
2. Supporting low-density policies that constrain supply. This increases price in the short-term, but also increases the pressure to defect by overturning the zoning or seeking exemptions to it. However much appreciation you can get on a single-family lot, the price jump from permitting multi-family or assembly projects is many times greater. So homeowners are not a stable profit-seeking coalition.
3. Supporting design bylaws or HOAs that make housebuilding more expensive. In my opinion this is a strategy for homeowners to capture a larger share of the surplus value of new builds, rather than a way to increase absolute profit. For instance, requirements that all new houses must have electric furnaces, or 2-car garages benefit the neighborhood more than the end buyer (if the end buyer thought it was worth the extra cost there would be no need to mandate it).
All in all, I don't think the game theory really works out to make NIMBYism the most profitable strategy for homeowners. It's more accurate to model NIMBYs as motivated by non-financial desires like local character or crowding concerns.
> Was the lost value purely speculative?
I meant the same as you when you said "mortgages were only justified because of expected price increases". The underwater part of each mortgage (the disappearing value that prompted the recession) was due to a disproven expectation rather than a real asset losing its value. I'm not positive it makes sense to draw this distinction, but it does feel like an answer to my original question about why house prices didn't drop more. In short: land never stopped appreciating; we had a temporary price spike due to financial shenanigans; the actual housing boom caused by the price spike was either non-existent or tiny compared to the underlying trends so there was no housing glut.
> Excuse my lack of clarity; I meant that developers expect more profit per house when mortgages are generous and would thus lobby harder against restrictive zoning. I wasn't alleging that zoning restrictions are due to a profit motive.
I mean, I think they are due to profit motive, on the part of local homeowners. Developers probably would like construction to be easier, but local homeowners have a lot more votes.
> So homeowners are not a stable profit-seeking coalition.
I think it's stable enough. Only 1 person having an exception is not stable (no one except that person would vote for it). Everyone being restricted is pretty stable. And the fact is that "home values" are one of the most commonly cited reasons for preventing new development. And I'm not saying it's the *only* motivation, just that I don't think there's much incentive to allow development at a slightly higher price point when they can instead capture all that value for themselves.
> I'm not positive it makes sense to draw this distinction, but it does feel like an answer to my original question about why house prices didn't drop more. In short: land never stopped appreciating; we had a temporary price spike due to financial shenanigans; the actual housing boom caused by the price spike was either non-existent or tiny compared to the underlying trends so there was no housing glut.
I'm not sure either, but this sounds right to me.
Have either of you two read Mike Bird's *The Land Trap*? It discusses the history of real estate bubbles. I wrote a review of it here if you want a summary:
https://progressandpoverty.substack.com/p/book-review-the-land-trap-by-mike
I think I read this review a while back
edit: it's from a few weeks ago so probably confusing it with something else
A very contentious spat between hereditarians and statistical geneticists like Sasha Gusev broke out this holiday weekend on X. It all started with something Rebecca Sear said in an interview...
Sear said eugenics wouldn't work because the traits they want to add or remove "are too complex to breed in or out of populations. And we don't really know what would happen if you tried to select for some of these traits. You might get all sorts of unintended consequences ... as you do with farm animals and dogs."
About 1:30 into the video... https://x.com/i/status/1784972168238969328
Full YouTube video here (note, I didn't listen to the entire interview)... https://t.co/sBBUPv2iKp
Philippe LeMoine dismissed her out of hand, calling her clueless. Damien Morris called her ignorant. Sichu Lu, an account who seems to have some genetics chops, said she was retarded. Sasha Gusev leaped to her defense.
AFAICS, Sear's first point, that we can't breed polygenic traits in or out of populations, is supported by polygenic liability threshold (PLT) models. These models were purely theoretical when they were first formulated in the early 20th century. But more recent GWAS studies have supported this theory (for instance, Huang et al., Nature 2024, "Examining the role of common variants in rare neurodevelopmental conditions", https://www.nature.com/articles/s41586-024-08217-y).
Sear's second point, that we'd get unintended consequences if we tried to breed out polygenic traits, is supported by the antagonistic pleiotropy hypothesis (APH). (described here in Wikipedia: https://en.wikipedia.org/wiki/Antagonistic_pleiotropy_hypothesis; and supported by studies such as Long & Zhang, 2023, "Evidence for the role of selection for reproductively advantageous alleles in human aging", https://ouci.dntb.gov.ua/en/works/l1waE8Z7/).
The only technical arguments I've heard against Sear fall into two main categories.
1. Yes, we can select for polygenic traits. We do it all the time in domesticated plants and animals.
2. Evolution through natural selection is a given, and it's obviously affected the frequency polygenic traits.
The 3rd argument against Sear is that she's politically motivated. Which is probably true, but doesn't affect the technical arguments except to cast shade on her.
The best example that I found for both PLT and APH in action is thoroughbred racehorses. Yes, the winning speeds of thoroughbreds have increased by ~0.11% per year from 1997 to 2012. But thoroughbred racehorses have higher rates of congenital diseases due to inbreeding, with studies showing 8-14% limb deformities and overall foal morbidity up to 88%! (I need to check that statistic, but I haven't had a chance to dig into Google Scholar).
Breeders have used genomic selection models to reduce the incidence of congenital deformities in thoroughbreds. Studies have shown that harmful haplotypes on ECA14 contribute to some of the risk, and they're trying to identify them in breeding horses and select against them. But under liability threshold models, fully eliminating ECA14 haplotypes is challenging. The trait is polygenic, with liability distributed across multiple SNPs, making complete eradication difficult without reducing genetic diversity further.
And the argument that evolution happens despite PLT and APH assumes that the human genome would remain static if removed from selection pressures. But it would not remain static, because the human genome mutates at a rate of 0.5–1.5 × 10⁻⁸ per generation. With ~3.2 billion base pairs, that translates to ~30–70 new mutations per child. Depending on the study, most of these mutations are neutral, with about 5-20% mildly deleterious, and less than 1% being strongly deleterious. And Less than 1 in 10K are beneficial. So selection can happen slowly because mutations over many generations delete genes and insert new genes into genomic deck, reorder the polygenic relationships.
Finally, the political argument. Sear is now famous (or infamous among the HBDers) for proving that Lynn and Vanahan's National IQ study was specious to the point of intentional fraud. And Retraction Watch has been publicizing her attempts to get the studies that rely on Lynn's bogus data retracted. She's identified about 175 and is starting to gain traction on her retraction crusade. Yes, this is partially political. But it's rich that the people who accuse her of being political fail to remember that Lynn was unabashedly political when he called for complete separation of the races and created a largely bogus database to support his crusade.
Happy to hear contrary views. But please argue with data and the theories.
Cheers!
I have only tangentially followed this discussion, but I think Sear's claim is valid insofar as this is extremely hard to do on a society-wide level. I've given the example before of how the Nazis killed all German schizophrenics for eugenics reasons, and it maybe decreased the schizophrenia rate of the next generation by 1% or something undetectable like that. See https://www.astralcodexten.com/p/some-unintuitive-properties-of-polygenic. This isn't just about schizophrenia being a threshold trait, at some point I calculated out what would happen if you tried something similar with criminals, and it would take executing every criminal for thousands of years just to lower genetic propensity to criminality 1 STD.
It's invalid insofar as sure you can do this, as long as you're willing to do extreme things and have lots of generations to work with. For example, if you banned normal reproduction, and said women could only get pregnant by inseminating themselves with Einstein's sperm (which you had on hand in some massive cloning vat), the next generation of humans would be much smarter than the last one on average. If you repeated again with the smartest Einstein-child for the next generation, they'd be smarter still, and so on. You can do something like this in farming, and it works fine for producing very large cows, although depending on exactly how you do it you might get some genetic diseases unless you work hard to avoid this (I don't think the work would be very very hard, , but it's not really worth farmers' time at this point).
>The best example that I found for both PLT and APH in action is thoroughbred racehorses. Yes, the winning speeds of thoroughbreds have increased by ~0.11% per year from 1997 to 2012. But thoroughbred racehorses have higher rates of congenital diseases due to inbreeding, with studies showing 8-14% limb deformities and overall foal morbidity up to 88%! (I need to check that statistic, but I haven't had a chance to dig into Google Scholar).
This seems mostly irrelevant, since nobody is proposing inbreeding, which is the mechanism causing the problem here. Furthermore, as far as I understand your argument, there is little direct evidence here for PLT/APH, you merely claim it would be hard to get rid off due to thresholds?
On Sear's argument itself, the problem with farm animals is that we don't really care about their health; In fact we want to maximize only a very small number of traits which obviously trade off against health, especially in the tails. Of course maximizing fat - in other words, obesity - in chicken is not going to be good for them.
Eugenics in humans is very, very different, and includes a wide range of both possible targets and possible approaches. Especially, we can simply start by optimizing for health-related conditions directly, which is already done in some cases. Monogenic rare diseases have of course been done for quite a while, and the currently available PGS embryo selection companies are looking at fairly uncontroversial common diseases and most importantly use risk/threshold designs themselves, so it's not really a problem for them, this is literally what they are already doing. Take a look at, for example, lifeview's EHS. In addition, when they looked into the relationship between different health measures and diseases, they didn't find trade-offs, they mostly found (weak) positive relationships: https://www.nature.com/articles/s41598-022-22637-8
This actually makes a lot of sense, and there is a wide literature on both the phenotypic
comorbidity and the genetic shared architecture of many different diseases. Structurally, this is also equivalent to classic evolutionary selection but without all the horrible parts it normally entails. There is very little evidence why such an approach should increase the risk for any particular disease, and even if we can find certain cases, very little reason to believe that the negative externalities will outscale the benefits. APH, btw, is just one of several hypothesis', primarily for aging, and even in that field I wouldn't consider it the most explanatory (which imo is mutation accumulation).
Sear's argument itself only fully applies for a very narrow band of approaches which are similar to what we have done with animals, such as IQ-maxxxing with complete disregard for health, including incest to really double-dip. But nobody is actually doing this; Steven Hsu is sometimes associated with that approach (minus incest), but he is actually one of the founders of lifeview/genomic prediction, which is not even looking at IQ so far at all. There will surely be some crazy online people somewhere advocating for it, but none of the people in this space actually building things is attempting it, and Sear makes a far stronger claim, namely that eugenics in general doesn't work.
>Finally, the political argument. Sear is now famous (or infamous among the HBDers) for proving that Lynn and Vanahan's National IQ study was specious to the point of intentional fraud. And Retraction Watch has been publicizing her attempts to get the studies that rely on Lynn's bogus data retracted. She's identified about 175 and is starting to gain traction on her retraction crusade.
Meng-Hu has a good article on this:
https://menghu.substack.com/p/national-iq-papers-must-be-retracted
Short version, if we were to take Sear's argument serious, we'd not be allowed to use large parts of the international data. All the issues they identify are general problems with third-world data, which are only ever raised if someone doesn't like the conclusions. Worse, most newer datasets - which don't have the issues Sear raises -come to congruent results anyway.
The Pesta controversy is a particularly glaring case: https://www.sciencedirect.com/science/article/pii/S0160289618302435?via%3Dihub
The paper found that if you use modern exam performance data, you will get mostly similar results to Lynn. It doesn't methodologically rely on Lynn at all, however.
Nevertheless, progressive researchers went hard after them anyway and got Pesta fired: https://quillette.com/2022/12/16/and-yet-it-may-or-may-not-move/
Ostensibly due to misrepresenting his research on the application to the NIH; However, as a researcher in genetics myself, he was if anything very forthcoming and specific. I've diverged far more from my initial statements with nobody ever complaining and this is generally well understood in the community as perfectly fine, but apparently only as long as you don't get controversial results.
I really recommend people to read Sear's own article here as well: https://www.statnews.com/2024/06/20/richard-lynn-racist-research-articles-journals-retractions/
It's extremely light on data, and very heavy on invective. It's particularly telling that she doesn't actually propose an alternative, better dataset, which would be the normal course of action in these kinds of controversies. Like her co-author Bird, she just wants to shut the research down entirely. It's perfectly fine and reasonable to be against Lynn's political views, but he still build up an unprecedent dataset that holds up surprisingly well.
Frankly, there is just nothing here except a classic witch-hunt. First you claim that a dataset is bad because of faulty methods and a plausible bias due to researcher's own stated views (fair enough, I'm all for improving standards!), then people use other approaches without the issues you named and show the original dataset is not particularly biased in any direction... and then you go after them, anyway. That's not the conduct of anyone interested in science. And again, I want to mention here, we don't do this ever in any other case; Nobody tries to outright get David Graeber's research and books retracted, for example, despite him openly stating that he considers it an important vehicle to spread his political (anarcho-communist) views and has frequently generalized small studies far beyond their applicability.
> This seems mostly irrelevant, since nobody is proposing inbreeding, which is the mechanism causing the problem here.
But what mechanism are they proposing? Again, if you accept the implications of liability thresholds and antagonistic pleiotropy, then selecting polygenic phenotypes to be bred in or bred out of a population may move the needle a bit, but you can only take it so far before the negative consequences start to outweigh the positives. The hereditarians hold up Ashkenazi Jews as proof that one can select for IQ. According to the hereditarians, this is the result of endogamy and preferential marriage to the smarter ones. And, yes, the mean IQ of Ashkenazis falls into the 107–115 range. But they suffer higher rates of both monogenic disorders and polygenic disorders (such as IBD). Endogamy is just inbreeding over many generations where people who frequently share the same ancestors 5 or 6 generations back marry each other.
> Then people use other approaches without the issues you named and show the original dataset is not particularly biased in any direction...
AFAIK, no one has performed any multi-national IQ surveys a la Lynn and Vanahan. And more rigorous IQ studies have failed to reproduce Lynn and Vanahan's national IQ scores for various countries. For instance, Lynn and Vanahan claimed the average IQ for Thailand was 91. Current studies show a median of between 98-102. They assigned Nigeria national IQ of 69. While Wicherts et al., concluded that because of these flaws and biases, it is not possible to derive a scientifically reliable, comparable “national IQ” for many sub‑Saharan countries (including Nigeria). Using the available data, looking at some less-than-perfect sources suggests that Nigeria is doing OK IQ-wise. A for-profit online IQ testing company shows that the average IQ of Nigerians taking their exam was 91.4 (https://iqexam.co/countries/nigeria-average-iq). The average IQ of students entering Lagos State University is 97.7. Contrast this with a report (I didn't check its validity) that claims the average IQ of US students entering college has dropped since the 1940s and the average is around the US mean of 100.
I seem to recall the Ashkenazi had and possibly still have some sort of breeding register to try to avoid their various exotic congenital diseases. If so, I assume there are Ashkenazis here who can say more on the topic.
There's an organization called Dor Yeshorim that runs a confidential genetic screening program. "Before a potential marriage match is finalized, the couple can check their ID numbers with the organization to see if they are genetically compatible. The specific carrier status of an individual is not revealed, only whether the match is advised against, which helps prevent children from being born with these diseases."
>But what mechanism are they proposing?
As I wrote, the most commonly proposed mechanism is embryo selection, which has none of the drawbacks mentioned here.
>AFAIK, no one has performed any multi-national IQ surveys a la Lynn and Vanahan.
Not to the degree they had, no, which is why it was still in use. But there are plenty of country-specific studies, and some looking at related concepts such as exam performance, such as the Pesta et al. study I linked.
>A for-profit online IQ testing company shows that the average IQ of Nigerians taking their exam was 91.4. The average IQ of students entering Lagos State University is 97.7.
No offense, but that's precisely what I'm talking about with respect to Sear's criticism; Neither for the company nor for the university average is there any reason to believe this is representative; Lagos State university has 35k students and is either the top or second best school in the entire country, Nigeria has a population of ca 240 million at the moment. You'd expect their average to be several SDs from the average citizen. The US mean you cite is afaik for public colleges, where just about everyone can enroll, so you'd expect it to reflect the population mean. The equivalent would rather be Harvard student's average IQ, which incidentally is ca 120-130, possibly higher.
Lynn's data is, if anything, more representative. None of this would be allowed to be used if we took Sear seriously!
Again, you're making unsubstantiated claims to support your argument. For instance...
> The equivalent would rather be Harvard student's average IQ, which incidentally is ca 120-130, possibly higher.
Did you check the validity of that statistic before you wrote it? I can't find any systematic IQ studies performed on Harvard students using randomized sampling. I found a study published by Shelley Carson, a professor at Harvard, who gave a selected group of highly creative students (n=25) a short-form WAIS-R test (Vocabulary + Block Design), and her study found that their mean IQ was 128. Secondary sources garbled the study and applied it to the entire Harvard student body, and that seems to be the source of the chestnut that the average IQ of Harvard students is between 120-130. AFAIK there have been no systematic IQ studies using a random sample of students done at Harvard. If there are, please share them.
There are some other studies that estimate Harvard IQ from SAT scores. And the one thing that IQ correlates well with is SAT scores. How much of that is due to g and how much to study is a contentious issue.
So, I do not doubt that the average IQ of Harvard students is in the range of 2 SD above the US mean. After all, Harvard probably is one of the most competitive schools to get into in the US, and many people with very high SAT scores get rejected (interestingly, Harvard doesn't track SAT score averages, but only 54% of the 21-22 applicants provided SAT scores). In the 21-22 application cycle, Harvard received ~61K applications, and it admitted 1,962 students—an overall acceptance rate of about 3.19%. Of those accepted, it looks like 48 were from Nigeria (but I'm having trouble verifying this, though—and it's likely some of these are grad students).
OTOH, up until 2024, there were between 50 and 55K students from low-IQ Nigeria studying at universities in the UK (undergrad plus grad programs). Beginning in 2024, the UK began tightening its visa requirements, and new students accepted from Nigeria have dropped by a whopping 60% compared to previous years. But it's clear that Nigeria was exporting its best and brightest students to the UK.
> Lynn's data is, if anything, more representative. None of this would be allowed to be used if we took Sear seriously!
To quote the old song, "Saying it don't make it so." Using the Nigerian example, Wichert noted that Lynn cherry-picked the three studies that reported the lowest IQ scores. Then he picked apart the methodology of those studies. Also, it looks like Lynn ignored between 6 and 9 other studies that gave higher IQ numbers for Nigerians. Wichert is a hereditarian when it comes to IQ, and even he thinks Lynn was full of shite.
Lynn is a fraud. Anyone who uses Lynn's data knowing that it's bogus is a fraud. Sears was not the first to voice criticisms about Lynn's national IQ database, but she has done a great service by compiling all the problems with it. You might call it "cancellation," but she and Retraction Watch are trying to get all the bogus studies that Lynn's fraud generated retracted. Good for her.
Sorry, but this is just getting ridiculous. Your original post was full of unsourced claims (such as the horse breeding stats) and baseless assumptions (such as talking about the relevance of inbreeding for human eugenics). In a comment section that is to be expected, so I engaged in good faith, explained that nobody is in favour of inbreeding and mentioning some other approaches that actual people do advocate for, and also going so far as to adding links that frankly should have been in your original post (such as Sear's Retraction Watch article). You respond by ... talking about inbreeding again (?!) and also making (unsourced) claims about the average IQ of Lagos State University, with an (unstated & baseless) assumption that this is somehow representative of Nigeria as a whole, a country still struggling with basic literacy & numeracy. Again, I didn't make a big deal out of it in my answer.
If you think that a claim needs a source, you can simply ask me. I didn't expect the Harvard stat to be contentious, so I didn't add a link, just as you didn't add a link for the Lagos State University IQ, and as it turns out, you don't even disagree! So it's unclear why you even mentioned the LaSU average in the first place, you should be aware of how wildly biased it is.
Even worse, you also make unsourced after unsourced claim in this very post! As far as I can see, Wichert is in fact not a hereditarian by any usual definition of the word; He is constantly stressing that high heritability can be caused by cultural factors, never gives any clear estimate he considers correct, and most of his papers I can find are explicitly phrased in opposition to hereditarians (especially Lynn).
Nevertheless, you may be surprised to hear that I have relatively little gripe with Wichert. His criticism is mostly fair and even-handed, and he still comes to an imo very plausible estimate of close to 80 (again he is stressing that it needs to be seen in the light of the Flynn effect and pro-western bias of the tests, making rather clear where he stands).
He did precisely what I consider the correct course of action: He proposed an alternative, argues for how it is better and recommends the usage of his own estimate instead of going straight to cancellations or calling names of people who disagree with him.
No offense, but I will not be engaging any further from here.
Still waiting for the data on Harvard student IQs that I asked you for. If you've got it, I'd be sincerely interested in seeing it, because I can't find any studies other than Carson's N=50 study of creative students.
> Sorry, but this is just getting ridiculous. Your original post was full of unsourced claims (such as the horse breeding stats) and baseless assumptions (such as talking about the relevance of inbreeding for human eugenics).
This is getting the pot criticizing the kettle's melanin quotient. But here you go. Some bedtime reading material for you...
And BTW, I explicitly asked you to back up your Harvard claim, and you never did.
----------------------------------------
Re: Horsies...
“Racehorses are getting faster” (Sharman& Wilson, 2015)
https://pmc.ncbi.nlm.nih.gov/articles/PMC4528479/
“Does inbreeding contribute to pregnancy loss in Thoroughbred horses?” (Lawson, et al, 2024)
https://pubmed.ncbi.nlm.nih.gov/38221707/
“Multiple Genes Related to Muscle Identified through a Joint Analysis of a Two-stage Genome-wide Association Study for Racing Performance of 1,156 Thoroughbreds” (Shin, et al, 2014)
https://pubmed.ncbi.nlm.nih.gov/25925054/
----------------------------------------
Re: Why exogamy is a good idea, and endogamy can lead to bad things...
“The Finnish disease heritage” (Norio & Löytönen, 2002) https://fennia.journal.fi/article/view/3775/3566
“Origin and spread of the 1278insTATC mutation causing Tay-Sachs disease in Ashkenazi Jews: genetic drift as a robust and parsimonious hypothesis” (Frisch, et al, 2004)
https://pubmed.ncbi.nlm.nih.gov/14727180/
“Implications for health and disease in the genetic signature of the Ashkenazi Jewish population” (Guha, et al, 2012)
https://link.springer.com/article/10.1186/gb-2012-13-1-r2
----------------------------------------
Re: Hahvahd. I can't find any careful studies of Harvard IQ, but this is what I dug up. I showed you mine, now you show me yours.
“Decreased latent inhibition is associated with increased creative achievement in high-functioning individuals” (Carson, 2003)
https://psycnet.apa.org/buy/2003-07329-008
Harvard admissions stats
https://bpb-us-e1.wpmucdn.com/sites.harvard.edu/dist/6/210/files/2023/02/harvard_cds_2021-2022.pdf
----------------------------------------
Re: IQ in deepest darkest Africa. From 2015-present, there've been no nationally representative, peer-reviewed IQ survey for Nigeria. Seems like the genetic racists would be all this, because with a nat'l IQ of 69 (per Lynn), Nigeria is their favorite talking point.
“Raven's test performance of sub-Saharan Africans: Average performance, psychometric properties, and the Flynn Effect” (Wicherts, et al, 2009)
https://jeltewicherts.net/wp-content/uploads/2015/12/wichertsravenafr2010.pdf
IQ scores of Lagos State University students
https://brght.org/iq/educational-institution/lagos-state-university/
Average IQ results from Nigeria according to IQexam, “The most accurate IQ test model ™” <LoL!>
https://iqexam.co/countries/nigeria-average-iq
----------------------------------------
And here is where Sear shreds Lynn's national IQ dataset...
’National IQ’ datasets do not provide accurate, unbiased or comparable measures of cognitive ability worldwide” (Sear, 2022)
https://www.researchgate.net/publication/360665701_'National_IQ'_datasets_do_not_provide_accurate_unbiased_or_comparable_measures_of_cognitive_ability_worldwide
One should also recall Goodhart's Law in these circumstances: "When a measure becomes a target, it ceases to be a good measure".
I think it's more like Gresham's law, where bad science drives out good. ;-)
...mostly because people repeat so many claims uncritically.
My sense is that the whole argument just isn't presented around a relevant crux - it's just a poorly-framed argument.
Polygenic liability threshold (PLT) models show that there's a liability for a certain trait that only becomes a visible trait once you cross a threshold. That liability comes from a combo of rare, high-impact variants and the larger polygenic background. It implies that you can remove some rare variants in a single embryo, but it's more difficult to remove from a whole population. You can't mechanically/deterministically "breed out" a complex polygenic trait, but you can reduce risks probabilistically. New mutations might emerge such that we can't reduce liability to zero, but I think we could probably map out a "selected" pathway that consistently reduces risk every generation (if we do consistent embyro selection).
On the idea that we get unintended consequences, it's obvious that we can, and this is very well-documented in companion and farmed animals. But the relevant question is not "will there be unintended consequences?", but "will the unintended consequences be worse than the benefits?" , "what are the trade-offs" etc.
I get the sense that the actual cruxes are things like:
1) What's the effect size ceiling of polygenic embryo selection (when do we get decreasing returns)?
2) How generalisable are polygenic scores across populations?
3) How predictable are unintended consequences?
>Yes, this is partially political. But it's rich that the people who accuse her of being political fail to remember that Lynn was unabashedly political when he called for complete separation of the races and created a largely bogus database to support his crusade.
Of course, the problem with both sides being intensely political is that convergence on the truth is impossible - everyone is already convinced beyond doubt that their priors/intuitions are correct, and the only thing that matters is marginalizing opposition. Debunking bad studies doesn't help much if there are no forthcoming good studies to replace them, and where would those come from?
Although I have strong disagreements with Yudkowskian rationalists, I think that all Bayesian rationalists (including Yudkowsky), plus scientific realists, post-positivist naturalists, and even scientific anti-realists, would agree that experimental evidence is what provides us with at least an outline of Truth (or a reproducible story about Truth in the case of the anti-realists). I think they'd all agree that by proving the null hypothesis, we weed out false beliefs.
> Debunking bad studies doesn't help much if there are no forthcoming good studies to replace them, and where would those come from?
Network epistemologists and social epistemologists would likely agree with your statement. But being an old-school Popperian, my philosophy of science aligns with the post-positivists. Theories need to be falsifiable to be scientific theories. If a theory is falsified, we know it's wrong. If it's not falsified, it doesn't necessarily prove it's correct — but at least we've eliminated a false hypothesis, and we can more profitably spend our time pursuing other avenues of research.
Galileo proved objects of different weights fall at the same rate. By doing so, Galileo disproved the Aristotelian dogma that heavier objects fall faster, but he did not provide an explanation for this phenomenon. It took Newton to quantify the behavior. And it took Einstein to explain the behavior.
So, no, I don't think the debunkers owe anyone an alternative theory.
You are Popperian? How suprrising. Tell me which facts would make you to consider anti-HBD debunked. I would lose belief in HBD if Nigeria goes the path that South Korea/Taiwan went (and later China/Vietnam). When I see Nigerian students winning IMO and IOI and Nigerian made electronics. You?
IQ is pseudoscience. Other than SAT scores, it has as much correlation with life outcomes as phrenology or astrology.
But if we look at average IQ scores, the Flynn Effect is a worldwide phenomenon — no doubt due to improved health conditions, nutrition, education, and economic development. African IQ is rising at the same rate that Asia is, although Asia has had a 20-year head start. The US has seen a 3 pt increase in median IQ every decade up to 2010 (that's why they have to renormalize IQ tests every decade or so). US average IQ in the 1910s was what Lynn said Nigeria's is (~70). There's no reason to think that America's gains are due to genetic improvement. And there's no reason to think that modern Americans are any more intelligent than Americans were a century ago. And there's no reason to think the Flynn Effect won't continue in Africa for many decades to come.
From: "One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909–2013)" (Pietsching, et al, 2015).
https://journals.sagepub.com/doi/10.1177/1745691615577701
(Behind a paywall, but I purchased the article).
I asked you what kind of evidence could have changed your opinion. You answered ANOTHER QUESTION (I asked nothing about IQ tests). You have nothing to do with Popperian ideology but in name.
Edited several times...
Sorry, I thought you understood basic Popper, and you were setting a trap for me. ;-) According to Popper, you can debunk the theory being tested (=null hypothesis), but you'll never be certain you've proved the alternative hypothesis. "Positive results never conclusively verify hypotheses; they only fail to falsify them.” This tends to rub Bayesians the wrong way. But to Popperians, Bayesianism measures how confident you are, not how good the theory is.
But even though HBD is not a single simple hypothesis, I think the various null hypotheses that HBD theories encompass *have* been disproved over and over, but the hereditarians just don't want to listen.
Minor clarification. IQ tests were originally designed to identify the cognitively impaired. And they're still good at that. But above -1 SD they have very low predictive value (other than standardized test scores).
>So, no, I don't think the debunkers owe anyone an alternative theory.
I agree, of course, that debunking a bad study is a clear positive in and of itself (whereas publishing a bad study is a clear negative). My point is simply that in the grand scheme of things this is only a minor step, and there doesn't seem to be any prospect of substantial improvement any time soon.
Polygenic here just means "more than 1," right? Eye color is polygenic and a population could breed that in or out, against certain limits depending on which genes are recessive or not. You wouldn't need to identify which genes are at play or even know about gene theory to get some measurable results. Does that fit the model or not?
Funny you should ask. I was just checking out eye color. When I was in school (long before genome sequencing), it was of held up as an example of Mendelian inheritance, which would have meant it was a monogenic trait. But it looks like two genes control >95% of eye color variation. And one of the two is a regulatory region that controls the other. So, they're tightly linked.
And turns out that eye color does not perfectly follow Mendelian inheritance patterns, but they're pretty close. Since the trait is only moderately polygenic, You might be able to breed for eye color in a population without producing AP effects.
But even with massively polygenic traits, you *could* move the needle with selective breeding to reduce or increase the frequency of a desired trait. You just can't breed it in or out completely. And the more you push the needle in one or the other direction, the more unintended adverse effects appear.
Yeah you can't breed out any recessive gene *completely* unless you're testing everyone, Tay-Sachs style.
Yes, Tay-Sachs is a single-gene disorder. Polygenic threshold limitation models wouldn't apply in the case of Tay-Sachs or any other monogenic disease.
Every polygenetic trait is a combination of many monogenetic ones. There's nothing qualitatively different, really. You just look at how the answer varies with X while you hold Y and Z constant.
Of course, the tricky part is that things don't have to be linear. Organisms are very complex and have lots of moving parts. So maybe removing X makes things worse, unless you also remove Y and then it makes things better. Same idea as if you have a mechanical loom and you make one part twice as big, that makes things worse unless you also adjust the other parts to match. Figuring all that out is tricky, but for now we can basically compare against the most common other values, to decide which gene variants we like or dislike. This is the same system used in selective breeding, both artificial and natural.
Frankly this whole thing sounds like a debate over the word "can". Can you jump over the ocean? Yes, but it's hard. But jumping over the first 3 feet of it is pretty easy.
I don't have the science chops to argue with you, but I want to ask: why does the argument against PLT and APH assume that human genome would remain static without selection pressure?
I don't think either theory assumes that. My argument was that if new mutations occur continuously, they will inevitably change, most likely damage, polygenic relationships and their liability thresholds, allowing natural selection to slowly favor or disfavor polygenic traits. That was my idea, and I should have noted that. My bad. Take it with a grain of salt. :-)
what if my response is just NOOOOOOOOO because i'm a contrarian. Should I post here?
Go watch the Monty Python argument clinic sketch instead.
"This isn't an argument; it's just a contradiction!"
"No, it isn't!"
In that case, you can't do that, but you will, which will make it satisfying.
No.
Source: I am also a contrarian. Or am I?
No.
Oho, looks like we've located DeBoer's newest alt account
In a 80,000 hours podcast, Rob Wilbin suggests that dropping fertility rates are not a consequence of cost and expenses, but more so a consequence of individual choices where people pursue hobbies, interests, and careers over children. And cultural acceptance of this furthers the trend. And the rise of the phone and digital entertainment mediums slow coupling as stimulating ways of engaging without in-person-relationship-forming interactions. I found this perspective in the podcast fresh and it captured the tension between individual and community needs that continue to reshape our society and we lean more and more toward individual determinism. Give it a listen if you are interested in the fertility conundrum!
https://80000hours.org/podcast/episodes/rob-luisa-parenting-chat/
As Jack says, I’ve seen this explanation as one that makes sense of how “cost” could be an issue when people are getting richer - it’s really *opportunity* cost, and as opportunities increase, those increase.
Haven't watched the podcast but it seems like that's possibly just a restatement of the same thing.
Why would you pursue hobbies over children? Part of your answer has to do with the cost of each. If your hobby is "travel" then, how much are you able to travel if you don't have kids? With cheap airfare, the internet making booking shit easier, ability to work remotely for many people, the cost of "travel" as a hobby is way lower than in the past when "travel" as a hobby, beyond some local domestic spots, was out of reach for most people.
Same with the cost of having kids ... if you think that sending them to college is a requirement, you're basically saving significant money for it the moment they're born.
I don't think it really has anything to do with cost. I like donuts, I don't like flan. Not because of costs of one vs the other. Just because that's my preference. I think for many more people than in the past, they just don't want kids, same as I don't want flan (that's an example, I love flan AND donuts)
Why don't they want kids? Because culture has changed from the past. Until some time relatively recently (60s?) it was just expected, you become adult, you pair up, you have kids. All of that seems to have disappeared or is in the process of disappearing. You become adult, you play, you work, you play some more, you work some more. Kids, for many, are just not something they want. They aren't weighing costs, they're just not interested.
The same is somewhat true of relationships in general. They seem harder than ever to get into and many have just given up and/or are fine not trying to find someone. That was less true in the past.
Sure, but the difference is that the solution isn't to decrease the cost of having children, it's to eliminate alternatives.
> it's to eliminate alternatives.
I've wondered about this myself - what would happen if you eliminated the alternatives - and decided to do some research on it a while ago. I ended up writing a post on the effect of curfews on fertility, with a focus on the era of Latin American military era curfews:
https://www.mangosorbananas.com/p/fertility-and-curfews
I don't think it's just a restatement, it's a reshaping of the conversation away from purely economic costs and towards other costs, both the opportunity cost of having kids (which stop you from doing a bunch of fun things) and the other non-economic costs of having kids (they can be annoying!)
In your travel example the increased opportunity costs are indeed driven by the decreasing cost of travel, but in other cases it might just be social changes. For instance a big part of it is just social contagion.
I'll be in San Francisco shortly before the solstice festival for an onsite interview. Would anyone be interested in meeting up Tuesday afternoon or Wednesday evening?
Is there any evidence for “micro-optimisation” doses of psychiatric or hormonal drugs in healthy people?
Pharmaceutical drugs (SSRIs, testosterone, Ozempic etc.) have clear therapeutic effects at normal doses, plus predictable side-effects. It seems plausible that there might be a “tiny, subclinical dose” where you get a small benefit with minimal downsides, e.g., TRT microdose nudging testosterone from the 60th to the 62nd percentile without giving you any side-effects, miniscule mood improvement through a tiny dose of antidepressants, tiny Ozempic dose bringing average food consumption down by 50kcal a day etc.
Any good evidence, or anecdotes on this? I'm probably not considering it, just curious why I've never heard of people trying it.
IME, the effect of testosterone on mood is very subtle, so only slightly nudging it up is unlikely to make a difference. (As well as the problems trying to do this as Erica has explained)
I've flirted with the model where many psychiatric drugs work, not by treating an abnormality or disease, but by enhancing healthy people.
Of course, these two models are not INHERENTLY mutually exclusive. But society sure acts if they are. I think the disease model will brook no competition, institutionally speaking. And besides just withholding research funding, I suspect our institutions actively harry and obstruct anyone who dares to suggest that doctors should be in the business of enhancing healthy people.
But I don't actually know much about the state of medical ethics besides the occasional post from Scott or discussion in these comments. Is the situation as dire as I have made it out to be? Are models that allow for human enhancement gradually gaining ground or is there virtually no movement in that direction?
For testosterone in particular, I would expect no effect from microdosing because there are homeostasis mechanisms that would cut endogenous testosterone production to compensate. In a normal adult cis man, the pituitary makes secondary hormones (FSH and LH) to "order" a certain rate of testosterone production. The testicles respond by making the testosterone, which circulates in the blood. Some small percentage of the testosterone gets converted to estrogen (roughly proportional to testosterone levels), and estrogen sensors in the pituitary up-regulate or down-regulate FSH and LH production based on estrogen levels.
The practical upshots of this for exogenous hormone purposes are:
1. Macrodosing testosterone severely reduces endogenous testosterone production, unless you take additional meds to counteract it.
2. If your testosterone levels are low because your pituitary is regulating levels to too low a target level but your testes are healthy (i.e. secondary hypogonadism), you can boost T levels by taking meds that block estrogen conversion (aromatase inhibitors) or boost FSH or LH levels. Clomid is the most common drug for the latter purpose.
3. Trans women can often shut down testosterone production almost completely by taking enough exogenous estrogen.
4. Both cis men taking exogenous testosterone and trans women taken exogenous estrogen usually experience testicular atrophy because of the lower FSH and LH levels.
I can't speak to psychiatric drugs, but I rely on prescription bioidentical hormones due to a genetic condition that makes me unable to make many hormones naturally.
I can say that nudging a sex hormone like testosterone by such a small amount (going from 60th to 62nd percentile) probably wouldn't be at all noticeable, at least from my experience. The human body is built to account for constant, mild fluctuations in hormones (especially women), so it tends to take a larger dose for your body to react to a noticeable degree.
This is why you see body-builders increasing their testosterone to insane levels; increasing their testosterone by 20% isn't going to make them 20% stronger. They probably would need to double their testosterone to get that effect (and even that may not be enough).
So I'm quite skeptical that a *micro* dose would have any significant impact. That being said, I have a keen interest in seeing some studies done on *low* dose HRT. I suspect low-dose therapy could be extremely beneficial to many people.
I personally know of three people who were diagnosed with psychiatric conditions, suffered for years with symptoms and harsh meds, later discovered they were low in progesterone or testosterone, and were able to cease all medication when they began taking HRT.
I'm actually writing an article about this at the moment... It's such an under-researched area of psychiatry, and I think it's criminal that it's not being explored more.
Thanks for the reply, interesting anecdote about the HRT and psychiatric conditions - I'll subscribe and look forward to your article.
I suppose I’d expected a micro-dose to produce a micro-effect. Some chemicals like caffeine or alcohol have a linear dose-response. A "micro-dose" of, say, green tea in the evening can give a slight, barely perceptible boost if you need to work late, without affecting sleep as a double expresso might. A small alcoholic drink can put you in a "sweet zone" where a few of your inhibitions are down but you haven't lost any clarity of thought.
So I was curious whether anything like that existed in the testosterone pathway. If it's possible, it might give you a tiny, barely perceptible strength/libido/energy boost (with tiny negative effects), which is probably what you should want if you're at normal levels to begin with. But, as Erica mentions above, it seems that isn't really how testosterone production works because of homeostatis.
I've known people to take 2.5 mg of Lexapro daily; that's 1/4 the standard dose. The general argument is that most drugs are taken at supra-optimal levels, because our Phase 2 / 3 trials system selects for the highest dose that doesn't cause extreme side effects.
There's been a lot of news recently about how bad UC incoming freshmen are at math, including those who had nominally passed honors calculus and/or gotten A's.
https://senate.ucsd.edu/media/740347/sawg-report-on-admissions-review-docs.pdf
A sampling of questions is on page 49. Is there a place where I can find the full test, or another equivalent?
Here's what gemini had to say: https://gemini.google.com/share/a8dfbfabb360
tl;dr: Opus 4.5 11/30/2025 7 questions, tl;dr of results:
5 correct, 2 partially correct, 0 wrong
(Opus 4.5 and Gemini 3 look just about even, and nearly saturated, from this benchmark-ette)
light a) correct
FeCuCl b) correct
C4Hn c) partially correct (but not far from fully correct, omitting some rather unreasonable cases)
solar d) correct
titrate e) correct
gases f) partially correct (but slightly better than Gemini 3)
S4 g) correct
full dialog at: https://claude.ai/share/cb4bf348-0e96-4919-a6c4-0a1dc5f9e055
List of questions and results:
a) Q: Is light with a wavelength of 530.2534896 nm visible to the human eye?
results: "Yes, without question. That wavelength is green light, sitting comfortably in the middle of the visible spectrum (roughly 400–700 nm)."
b) Q: I have two solutions, one of FeCl3 in HCl in water, the other of CuCl2 in HCl in water. They both look approximately yellowish brown. What species in the two solutions do you think give them the colors they have, and why do these species have the colors they do?
results: Got all the important points - that the species are FeCl4- and CuCl4 2-, that they both have LMCT strong, allowed transitions in the UV to blue, which dominate both colors, that the Fe d-d transition is spin-forbidden and doesn't contribute to the visible color, that the CuCl4 2- d-d transition, while allowed(-ish, partially Laporte-forbidden) is down in the near-IR and doesn't contribute to the visible color.
c) Q: Please pretend to be a professor of chemistry and answer the following question: Please list all the possible hydrocarbons with 4 carbon atoms.
results: Missed a few of the 3-unsaturations species: There are a couple of other highly unstable species under C4H4 besides the tetrahedrane and bicyclobutene that you already mentioned. Taking cyclobutadiene that you've cited as the 1,3 isomer, a 1,2 isomer is formally legal as well, and a methylcyclopropanediene, anc cyclobutadyne.
d) Q: Does the Sun lose more mass per second to the solar wind or to the mass equivalent of its radiated light?
results: "Radiation wins, by roughly a factor of 2–3."
e) Q: Consider a titration of HCl with NaOH. Suppose that we are titrating 50 ml of 1 N HCl with 100 ml of 1 N NaOH. What are the slopes of the titration curve, pH vs ml NaOH added, at the start of titration, at the equivalence point, and at the end of titration? Please show your work. Take this step by step, showing the relevant equations you use.
results: correct at all three points, and stated a universally true equation for [H+] as a function of volume
f) Q: Please give me an exhaustive list of the elements and inorganic compounds that are gases at STP. By STP, I mean 1 atmosphere pressure and 0C. By inorganic, I mean that no atoms of carbon should be present. Exclude CO2, CO, freons and so on. Please include uncommon compounds. I want an exhaustive list. There should be roughly 50 compounds. For each compound, please list its name, formula, and boiling or sublimation point.
results: Very good, though not quite exhaustive. All the compounds it cited were valid. It (correctly) blew through my "roughly 50" to give 68 out of 97 possible. Actually a bit better than Gemini 3's 63 of 97.
g) Q: What is an example of a molecule that has an S4 rotation-reflection axis, but neither a center of inversion nor a mirror plane?
results: Fully correct, gave three examples, all of which are valid: A conformation-dependent tetraphenyl tetrahedral case with a "propeller" orientation of the phenyl rings, an alternating R and S chiral substituted tetrahedral case (which doesn't depend on a specific conformation), and a 1,3,5,7 tetramethyl substituted cyclooctatetraene
Is there a way to roll employer matching funds into the listed external matching funds?
Re: transgender sports & biological advantages, this has long been my view, glad you wrote up something about this though it sounds like you take the other side of the argument (though I can't read the post, sorry for not subscribing). In particular, as someone who has never been particularly athletically inclined, there's no way I would ever be competitive with top female athletes even with the advantage of my male body.
My preferred solution to this is to just use what every competitive online multiplayer game has already figured out: matchmaking plus anti-cheat. You can use ELO or similar to update an estimate of player skill and assemble matches with a goal of each side having an equal chance at winning. As long as you also prevent cheating on whatever the agreed rules of the sport are, everyone has a fair shot to move up and down the ranks, and is matched with players of similar skill in competitive matches. Sure, the very top players likely have some unusual combination of biological advantages, but you can always make filtered leaderboards for whatever particular set of characteristics you care about (Top Women, Top Octogenarians, Top Asthmatics, etc etc).
The entire argument is about whether men should be allowed in the Top Women category, so this doesn't really solve anything. The premise of a women's only category is that women aren't competitive against men so they need their own league. If we accept the argument that we shouldn't care because every winner has an unfair advantage anyway, then there is no justification for women's sports existing at all.
The argument is female modesty and decorum requires separate female spaces. This is what the non-weird world believes.
It clearly does offer a solution. I'm not advocating for or against it, but just to clarify:
There would be no women's sports leagues. Anyone can be matched against anyone else with a similar skill level.
There would (or could) instead be women's coverage. A broadcaster might choose to cover all the top women in their matches. Each of their matches may or may not be against other women, but they'd be against someone close to their skill level. Interested spectators could watch the world's top female athletes compete.
It doesn't answer whether a particular broadcaster would include trans women in their women's category. Maybe it would be influenced by demand and the free market.
> Maybe it would be influenced by demand and the free market.
In theory, we could have two kinds of sport events, and let the market decide, which one survives, maybe both.
But it wouldn't work in practice, because instead of voting by their wallets, people would instead do political pressure on media to stop broadcasting the "wrong" kinds of sport.
You can have different lists, trans-inclusive and trans-exclusive depending on your point of view. There's no need for universal agreement because the list doesn't restrict who you are matched to play against, as that can be determined purely by skill.
I don’t think you understand sports. Or the arguments involved.
If all sex segregation is to disappear it’s not going to re-appear with trans inclusive tennis games at Wimbledon. People want to see the best, and the reason the women’s sports exist at all was a once understanding of biological difference. If that is to be forgotten or ignored then there would be a single sex competition and not a trans exclusive/trans inclusive division. Which would upset both sides.
> People want to see the best,
If you mean best as in most skilled, then someone could easily fill that market by having a basketball league that only lets in people that are shorter than average. People would need to be skilled instead of just tall.
If you mean the best in general, then why have a women's league at all?
> If you mean best as in most skilled
The best in general. There are Olympics for the disabled, and while admired it’s clearly not as popular.
> then why have a women's league at all?
Women wanted to compete. And women’s sports are just not that popular. Tennis maybe, or perhaps some sports where the female physique isn’t a disadvantage - like gymnastics.
What the argument here anyway, because we now trans we should have dwarf basketball?
I don't think there's a whole lot of controversy of trans men in men's leagues. It's really focusing on women's leagues. So this isn't so much a question of why people watch sports in general as women's leagues in particular.
> What the argument here anyway, because we now trans we should have dwarf basketball?
Mostly that I misunderstood you.
The trans movement believes that transwomen are women. That’s one side. Then there’s the rest.
We already made that list, though.
Is it a single list? The world record in marathon for women in mixed events is much better than that for women-only events. Including 2 cases of women holding both records.
The problem with this is something like "state" or "the olympics". Under your model, there would be zero women in the olympics, or very few in very non-physical sports (marksmanship type stuff?). Starting from highschool, you'd have male-dominated everything with occasional female-outlier people making it through, essentially eliminating girl's team sports. And then it would never recover.
I think the foundational place you start on the whole trans thing is pretty well proven to be "do you even care about women's sports existing?". People can answer different ways on that, but knowing the guy you are talking to is OK with women's sports taking any size hit to give the marginalized-group-de-jour what they want is relevant to know as compared to any other possible position.
I don't think it would necessarily hurt female participation in sports, and may even help by making it easier for everyone to compete in well-balanced matches. It's true that men would dominate at the very top, but the vast majority of people aren't anywhere close to the very top, and would get to compete with people of a similar skill level regardless of their particular biological advantages/disadvantages. Also in the school example, there's no need to restrict things by grade level; you might have women competing with slightly younger men for example.
Oh, it would hurt it a massive amount, with near 100% elimination of girls from team sports from the earliest points. You are thinking the differences between highschool girls and boys are small enough to be adjusted for with just a few years of age, and it's just not so.
To put this into perspective, the woman's national soccer team (an adult professional-level sports team) has LOST to a single academy's boys under-15 soccer team. The best female soccer players in the US, or nearly so, lost to a group of 14-year-olds from a single highschool.
To say that another way, if you say "Oh, the oldest girls can just play highschool sports with the youngest male highschool players..." you've just eliminated girl's highschool soccer, full stop. The male advantage is not small and competitive sports are not and cannot be set up like non-competitive video game matchmaking; there's just not the ability to make enough matches or enough supervision to be constantly running them.
Again, what you are proposing would start eliminating girls sports at approximately 100% levels just after the onset of puberty; having lost the training and facing gaps that only grow as they age, you will have also effectively eliminated women's sports entirely.
Doesn't matter, and that's sort of the point. If it was a smaller field, then men would use their still greater endurance to compete on it harder. If it was a contact sport, they'd be bigger, or else stronger at the same size. They'd be faster, blah blah.
You can try to find any sport you want, and if it has a significant physical element, men are going to have a massive advantage at it. Tennis? Men. Golf? Men.
I'm being forceful here, but what you said above is nonsense because the women don't "compete just fine" during some magic period before they get tired out. The men are stronger and faster during that time, too, hopelessly so, to the point where for women to get an even match they have to play against highschool JV squads to have a chance.
Does introducing other languages to small children who otherwise live in a monolingual environment have any effects? Having a kid now it seems like there's a much bigger trend towards bilingual or foreign language books. May just be a quirk of the books we've acquired, but I'd say like 5-10% of them seem to be some flavor of bilingual (read: spanish).
Is there any studies/info on what effects this has, if any? Does early exposure to other languages matter for any cognitive reasons? I'm guessing the effects would be small in either direction, but interested if there's any concrete data one way or another.
EDIT: Should clarify; while I worded my first sentence kinda broadly - I'm specifically asking about the effects of bilingual children's books, not a full "teach your kid a foreign language" program. A lot of the replies are addressing the latter, but my question is more about whether just reading simple bilingual books (e.g. "Besos for Babies" is an example of one in our collection) has any effects.
I don't know any studies, but I imagine that reading books is a very inefficient way to learn a language for a child, compared to talking or even watching cartoons.
I suspect this is just a status thing, a combination of "languages are high status" and "reading books is high status", therefore reading books in a foreign language is high status. You would probably achieve much more by showing the kids Paw Patrol in Spanish, because they would be naturally motivated to pay attention.
You're asking about what is called a "bilingual advantage" in the linguistics literature (https://en.wikipedia.org/wiki/Cognitive_effects_of_bilingualism). It's very clear that there are direct benefits of speaking different languages - for example, being able to travel more widely, read more literature in its original language, and experience more cultures. It's very unclear whether there are broader (non-linguistics) benefits of being bilingual per se.
There are robust studies that find neurological differences between monolinguals and bilinguals. The literature finding *behavioral* differences has not been treated well by the "replication crisis" of psychology (https://en.wikipedia.org/wiki/Replication_crisis): Many experiments which found benefits for bilinguals in the late 1990's and early 2000's were re-run in the 2010's, but found null results. It's possible that earlier works failed to fully correct for confounding variables which are highly correlated with bilingualism (such as socioeconomic status and level of education) which also influence other cognitive functions. Unfortunately, this topic has become politicized within the research community, in part because the research finding bilingual advantages was a response to an early generation of research which (likely incorrectly) found bilingual *dis*-advantages, so folks are sensitive to the pendulum swinging back too far. There's a quote from a researcher (Mark Antoniou) that I find particularly damning: "Those familiar with the literature are able to surmise whether the findings [of a study] will be for or against the existence of a bilingual advantage simply by peering at the list of authors." (https://www.psychologytoday.com/us/blog/life-bilingual/201906/the-bilingual-advantage-three-years-later). My personal take is that, if a bilingual advantage exists, it must be weak enough or specific enough as to make it irrelevant to any individuals making choices for their children.
All that said, I think it's a great idea to raise kids speaking as many languages as possible! The linguistic benefits are excellent, as mentioned by other commenters, even if the broader cognitive ones don't seem to be.
> There's a quote from a researcher (Mark Antoniou) that I find particularly damning: "Those familiar with the literature are able to surmise whether the findings [of a study] will be for or against the existence of a bilingual advantage simply by peering at the list of authors."
Why do you find that damning? You can substitute "the existence of a bilingual advantage" out for any question in any field and the statement will still be true. The only thing it relies on is that people continue doing the same things they've done in the past.
"as many x as possible!" is a red flag that you are not considering opportunity costs.
What I wrote was definitely misleading. In my head, "as many X as possible" was smuggling in "when not taking away from other important uses of time and resources," but that's obviously not how you'd normally read it. In context, I was trying to push against a monolingual default that I was worried my earlier claims would seem to support - again, there's a fear of the pendulum swinging back too far.
Having a private language ("idioglossia") definitely wouldn't give you the culture/travel benefits of learning other languages, for sure!
Attempting an Ideological Turing Test, I think that even the most gung-ho bilingual-advantage proponent would believe that your situation wouldn't have benefited you much. The classic explanation of the bilingual advantage is that a bilingual speaker is in a constant state of balancing their languages in their mind, and so trains their brain to more easily switch tasks or suppress irrelevant thoughts. In your case, or even in a more common case of an immigrant needing to switch entirely to a new language, I'd expect this balancing act to apply much less than in the case of someone who is actively using their different languages in different contexts.
There's pretty good evidence that learning languages as a kid makes you better at languages as an older kid or adult! Especially phonology/pronunciation, which are often quite transferable across different languages.
We're going for "trilingual by default" for our boy, but I can already see why any data collected from him will be confounded. If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two.
I'm actually tempted to teach our lad the whole phonetic alphabet, including those weird Xhosa clicks. It's only like 200 sounds, and you can probably master them all in a couple of weeks.
"We're going for "trilingual by default" for our boy, but I can already see why any data collected from him will be confounded. If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two."
From experience, I'd recommend limiting it to two. I tried to teach my kids three from birth but I just couldn't put in enough time to teach them the non-native language of the family (German) - ie. the one which neither me nor my wife speaks natively. We did make an absolutely no exceptions rule to only ever speak in our native languages to the kids (with the biggest burden being on me, as we live in my wife's country not mine, so they only got English from me - and the TV...).
I think it is probably better to get them really fluent at two languages first (and possibly also teach them the phonetics of the third language while they are very small - ie. before they are two - because that is definitely when they best learn that). This will be less confusing for them and learning a third language later in life will anyway be much easier once they already know two languages (due to loss of speaking embarrassment etc).
Cool, that's interesting.
We're quite lucky in that we've got three "naturally occurring" languages (Mum's L1, which we speak at home- Chinese, the language of the country we live - French, and my L1, English, which we speak a lot with friends and family, and for work etc.). My sense is that we should prioritise Chinese at home, and neglect English, because he'll probably end up learning English no matter what, but we might adjust depending on how he turns out.
We have a friend who has trilingual kids without trying very hard (she taught them Chinese, French at school and with dad, and English became fluent through YouTube and gaming!), which seems pretty ideal.
Ah, yes, then your situation is a bit different than mine is (where mother and country language are the same) and I speak two other languages but was the only source of these for my kids (besides English on TV, Youtube, etc). So you will likely have much better chances of trilingual success. Wish you luck!
> If he's smart and responds to three languages naturally, we'll stick with it - if he's less smart, we'll limit it to two.
Language learning is one of those things that is notable for not being related to intelligence. (Not quite: people who become noted rhetoricians tend to be more intelligent. But idiots and geniuses achieve the same levels of fluency.)
If your child can learn one language, he can learn three languages just as easily. Though note that if children perceive a possibility of getting people around them to speak what the child sees as the dominant local language, they will try to do that.
> I'm actually tempted to teach our lad the whole phonetic alphabet, including those weird Xhosa clicks. It's only like 200 sounds, and you can probably master them all in a couple of weeks.
What do you mean? All children are born recognizing those sounds. It takes them more than a couple of weeks to learn to produce them. It takes them even longer to achieve the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds.
That learning will take place whether you want it to or not as long as the child doesn't have to use a language that draws the distinction.
Except of course that South African clicks don't even have that handicap; you can hear the difference between them just fine right now. Do some babbling with them and you'll probably figure out how to fit them into a stream of other sounds. Note that the dental click and the lateral click are frequently used in American English, and all native speakers are expected to be able to produce and recognize them (in isolation, not as part of a word).
I definitely disagree with the "language learning is not related to intelligence" claim (almost tautologically), but not sure what specific claim you're making.
On the phonetic alphabet point, I've learned a bunch of languages that use sounds we don't have in English - it's a cool skill that you can develop relatively easily as a kid, but gets difficult as you get older. After learning SE Asian languages, I can distinguish certain contrasts (like /d̪/ /ɖ/) that monolingual English speakers just can’t “hear”, even after instruction. A few years ago, did a pronunciation teaching course and was surprised to find that all Anglophones except me couldn’t identify the difference at all. But weirdly, I can't pronounce /ʌ/ even when both my parents had the /ʌ/ /ʊ/ distinction. I'm sure that focused learning as a kid would give you a massive leg-up on these distinctions.
I've done accent training with Chinese kids of different ages, and it's relatively straightforward to get real improvement in producing and identifying new sounds. Of course, this doesn't translate to mature adult use unless they practice a lot, and get consistent exposure, but you can get real progress from a couple of hours a week.
But I take your point about "the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds". I wouldn't expect full distinction between all sounds, but I'd expect that, if I spent 2 hours a week for a year on pure, global phonology (say, when my kid is 6), he'd get a huge boost at producing and distinguishing certain sounds.
> I definitely disagree with the "language learning is not related to intelligence" claim (almost tautologically), but not sure what specific claim you're making.
There isn't a difference in fluency between idiots and geniuses who learn the same native language. People don't learn language by applying their intelligence. They use dedicated language-learning systems. Relatedly, given a population of people trying to learn a foreign language, intelligence will not predict which ones do so more successfully.
> (almost tautologically)
You're thinking about "learning" languages the wrong way. Is it tautological, or delusional, to suggest that learning to walk is driven by intelligence?
> After learning SE Asian languages, I can distinguish certain contrasts (like /d̪/ /ɖ/) that monolingual English speakers just can’t “hear”, even after instruction. A few years ago, did a pronunciation teaching course and was surprised to find that all Anglophones except me couldn’t identify the difference at all. But weirdly, I can't pronounce /ʌ/ even when both my parents had the /ʌ/ /ʊ/ distinction. I'm sure that focused learning as a kid would give you a massive leg-up on these distinctions.
It won't. As a baby, you already know the distinctions. As a kid, you will forget them.
For example, before she could speak, my sister "learned" that my name was k!, since I would make that sound to her. She once saw me approaching the house through the window and ran over to our mother excitedly saying k!k!k!k!k!k!.
But more recently she chastised me for not having taught her that very sound when she was young. She retained no memory of how to produce it and struggled quite a bit while I demonstrated it for her.
> I'd expect that, if I spent 2 hours a week for a year on pure, global phonology (say, when my kid is 6), he'd get a huge boost at producing and distinguishing certain sounds.
I wouldn't expect this. If you maintain the practice 𝘧𝘰𝘳𝘦𝘷𝘦𝘳, you'll get the effect you're hoping for.
Though if that's what you want to do, 6 is a late start. You can start that as early as you want, and there's not really an advantage to waiting.
> But I take your point about "the "mature adult" state of having successfully learned 𝗻𝗼𝘁 to recognize the difference between certain sounds".
You still seem to think of sounds as something that you learn to tell apart. In that model, if you give a child the ability to tell two sounds apart, maybe he'll still have that ability later when you test him again.
But that isn't what happens. If you forget the difference between two sounds at age 3, or if you do it at age 8, you've forgotten it either way.
"There isn't a difference in fluency between idiots and geniuses who learn the same native language. People don't learn language by applying their intelligence. They use dedicated language-learning systems. Relatedly, given a population of people trying to learn a foreign language, intelligence will not predict which ones do so more successfully."
This seems just insanely untrue. Native speakers clearly differ enormously; some native speakers struggle to understand or produce certain sentence structures, have much smaller vocabularies, or have speech issues. I'm not sure how connected this is to "raw IQ/g" or something, but I can't imagine a use of the word "intelligence" that precludes differences in these different abilities. Foreign language learning is even more driven by intelligence; it relies on memory, grammar learning etc., which are all directly dependent on intelligence.
"You're thinking about "learning" languages the wrong way. Is it tautological, or delusional, to suggest that learning to walk is driven by intelligence?"
It's tautological. Walking, however, is relatively capped in terms of capability, so there's very little difference between someone with "just enough" intelligence to walk and "more than enough". Language is not like this. Linguistic abilities clearly run the whole gamut from zero to genius.
It's possible to use language in ways that reveal that you're stupid. You shouldn't confuse that with misusing the language. Stupid people are no more prone to doing so than other people.
> some native speakers struggle to understand or produce certain sentence structures, have much smaller vocabularies, or have speech issues.
(a) Some native speakers struggle to understand certain sentences because they can't understand the concepts. But that's not a flaw in their use of the language.
(b) Vocabulary size is very diagnostic of intelligence. It's not at all diagnostic of ability in the language. Vocabulary is driven by what you're exposed to, and people are systematically exposed to different things in ways that are driven by their intelligence, and also in ways that are driven by their relatives.
You might notice that people maintain the ability to learn new vocabulary without difficulty throughout their entire life, and it's not something that presents difficulty to people attempting to speak a foreign language.
(c) Really? There's nobody out there who thinks speech impediments are caused by stupidity.
> Foreign language learning is even more driven by intelligence; it relies on memory, grammar learning etc., which are all directly dependent on intelligence.
This is completely false. Foreign language learning is driven by practice, just like native language learning. Do it more and you'll get better at it. Do it more than someone much smarter than you who focuses on their textbook, and you'll be better than they are.
The phonetic alphabet is a bit problematic - there isn’t exactly a discrete set of phonetic sounds that make up the phonemes of the world languages. If you go through 5 places of articulation for consonants (labial, dental, alveolar, palatal, velar) and 5 modes of articulation (voiced nasal, voiced stop, unvoiced stop, voiced fricative, unvoiced fricative) and add about a dozen pure vowels and a few diphthongs, you’ve got most of the sounds for a bunch of languages. Then you start adding a few more locations like uvular, pharyngeal, glottal, and a few more manners like ejective and aspirated and you start getting better, but then you start getting into more detailed questions about whether Hungarian “ty” and “gy” are the same sort of palatal as those of other languages, and whether any American and British vowels are close enough to count as the same or whether the subtle differences are enough to represent them all differently.
There’s probably plenty of value in learning, from an early age, about all these features, and learning to pronounce and hear a good sampling of them beyond the ones that appear in the language(s) you speak, but you shouldn’t expect to learn “all” of them, because there isn’t a clear list of “all”.
Yep, agree with these points. Tones are also a big one, and there are countless possible tonal variations.
With Retsam's clarification edit (that we're looking at short-term exposure based on reading from bilingual books), I think that Torches's suggestion is the most relevant benefit: A baby who hears sounds from different languages early on may find it easier to distinguish those sounds later. With two caveats:
(1) In the case Retsam describes, it doesn't sound like the parent is a particularly proficient speaker of the language they're reading. In that case, I wouldn't expect as much benefit, because the child would be exposed to English-accented Spanish, which sounds more like English than like Spanish. (Contrast this with Torches, who I imagine is practicing the phonetic alphabet!)
(2) Solid experimental evidence is thin on the ground. There's good evidence that long-term exposure as an infant can lead to benefits even years later (looking at heritage speakers of different languages), and that short-term exposure can lead to benefits a bit later (the longest I've seen is a couple months), but the combination of short-term exposure and long-term effects is both the theoretically least-plausible and the empirically least-tested.
Beyond this, I think we're mostly looking at benefits in language attitudes - the child may learn early on that it's normal for someone to speak Spanish. That's a valuable lesson, but there are better ways to learn that lesson (e.g. interacting with anyone who speaks Spanish - which would also be much better for learning sounds, come to think of it).
Are you looking specifically for cognitive effects (in which case my instinct would be 'a small effect consistent with any other activity that keeps your brain busy, like learning piano or chess'), or beneficial effects in general?
Having some knowledge of other languages gives you cultural capital, lowers the barrier to learning that and other languages further later on, can help with literacy in your first language, helps you understand other cultures and empathise with non-native speakers of your first language, and so on.
I just saw a recent news headline go by in my feed about how 2+languages leads to cognitive benefits. It seems like this is a common-sense side of the argument, though I know how often this turns out untrue or very marginal.
What if playing music was considered another language?
I wonder how much of the popularity of bilingualism is reversing the causality. Being smarter makes you better at languages, not the other way round.
I thought this recent headline referred to a groundbreaking large study, though. Qualitatively different from past headlines.
I listened to a podcast (can't remember which one, maybe Coleman Hughes?) with John McWhorter -- NYT columnist, Columbia linguistics professor -- and he basically said "I love languages, but no, there's no real cognitive advantage to learning other languages." From what I recall he even says that it's pretty pointless to require language classes in school, since <1% of people who are fluent in a second language learned it primarily through school language classes (usually it is family, living in the country, etc). Did not look up the primary refs myself but based on the broader cognitive ability literature I would put a lot of money on the "no" position, in large high-quality studies. Clearly if you go hunting, you will find some studies that say "yay language learning" because of the obvious selection effects at play.
But I keep my mouth shut when family members talk about how important it is that they get little Billy into the Spanish Immersion school.
I gut some use out of my O level french when I moved to Brussels ... i was far from fluent, but you don't need fluency for everyday purposes.
I think that knowing a little bit of French, a little bit of German, and a little bit of Latin is necessary to be a well-rounded English speaker.
So are you saying speaking more than one language takes up brainspace and makes you "dummer" for other things? My kids grew up fully bilingual from birth (no need for books etc, just one parent speaking one language and the other speaking the other), does that mean this has hampered their other potential intellectual pursuits later in life? I speak three languages fluently, does that mean I will always be bad at some other stuff 'cause my brainspace has been all used up?
"Do you think you could train yourself to recognize images on a sub-second timeframe? So you could look through 1000 images in 100 seconds (or less) and recognize if any of them were new?"
No, never. But - admittedly with no evidence whatsoever - I am loathe to believe that has anything to do with being a ployglot.
There is some slowness in young DLLs in picking up the preferred local language, but it disappears around age 6, and very soon after that they're statistically indistinguishable from monolinguals. It certainly has nothing to do with "brain space." You can't even distinguish the kids who were in a home with 1, 2, or 3 languages just by looking at their local language skills.
I think you are just needlessly aggressive here. To me the response makes perfect sense. You worry about brain wasting capacity on languages which could be missing somewhere else. Edward says that empirically, after age 6 we do not see any missing capacity.
Of course, the extra languages have to be stored in the brain somewhere, but it seems that either the amount of extra space is negligible compared to the total brain capacity, or that the missing capacity is something unimportant (maybe remembering fewer episodes of Paw Patrol?).
For those of you working in technology, what’s going on with the job market when it comes to more senior roles do engineers or engineering managers? I keep reading that people are getting ghosted left and right, big haircuts in terms of comp, you’re lucky if you get on the phone with anybody to begin with. Any reports from the ground?
“ getting ghosted left and right, big haircuts in terms of comp, you’re lucky if you get on the phone with anybody to begin with ”
This was my experience two years ago after I was laid off from my human factors/UX researcher job. I ended up switching careers (to skilled trades) and haven’t looked back, wish I made the switch way sooner instead of fruitlessly looking for another tech job for a year!
Skilled trades is the way to go! Last year swapped from managing data analyst to refinery operations. So much more fun (minus union nonsense but whatever).
Sorry, could you elaborate on what is meant by "skilled trades"? When I read that I think " plumber", but by your post and the one above it sounds like you're referring to something else. Thanks for the clarification.
I think what you’re thinking is correct! I mean any blue-collar work that requires some training to learn. Like plumbers, electricians, pipefitters, carpenters, etc. Refinery operations is just a small (well-paid, intellectually demanding) niche within blue-collar work, but it’s definitely blue-collar.
Unfortunately, everyone is free to become a software developer, but after I read how much mandatory training is required in my country to do qualified blue-collar work, I gave up.
Yes the training pipelines are dumb and have a large opportunity cost. It’s hard to get by for 4-5 years off apprenticeship wages (usually starts at 60% journeyman wages 1st year and goes up to 90% by the 4th/5th year).
That specific reason is why I went into refinery operations, because there is no apprenticeship and you start out making ~$100k with no overtime and over 10 years or so get up to ~$130k with no overtime (can easily clear $180k with overtime if you don’t mind working it).
If I understand correctly, FarmKind's shrimp matching donation campaign uses a pre-donated pool, and doesn't lead to any actual amplification of donations, unlike GiveDirectly's matching, where each donation actually leads to more money being donated than the donation, itself.
GiveDirectly states:
"This is a “true match”, so if we don’t reach our $600,000 goal, the leftover match funds will not be donated to GiveDirectly."
Farmkind indicates the opposite, however, here: https://www.farmkind.giving/support-bonus-fund, in "More detail on how it works:"
"In a conventional matching campaign, participants generally assume that by donating through the campaign, before it ends, they cause the matching funder (or in our case 'bonus supporter') to donate money that they wouldn't have otherwise. So a $10 donation can cause $15 to be donated in total. But critics of these kinds of campaigns fairly point out that if the matching funder has already committed the matching funds to the charity, then participants in the matching campaign aren't causing any more money to be donated to the charity than if they made a normal donation.
There are two senses in which our bonus system is not like this:
1. Participants get to direct the bonus funds to the charities they care most about. This is the most direct and clear benefit of getting your donation boosted through FarmKind...
2. Our bonus system isn't finite or time bound: It's an indefinite cycle of giving. In a conventional matching campaign, the funds are typically provided by a single large matching funder with a limited amount of funding, to be deployed by a certain date. FarmKind democratizes the provision of bonus funds: Many donors, big and small, support the bonus system on an ongoing basis. So long as enough donors are willing to provide bonus funds, and enough donors are willing to use them, the system can run continuously and indefinitely. As we explained earlier, this allows standard donors and bonus supporters to multiply their impact relative to what they could achieve without working together."
I don't get what they mean by the system not being time bound, given that this campaign has fixed dates. Maybe they mean that the whole matching is an accounting fiction, so they can set whatever dates they want, and it's all meaningless. If sufficient new donations aren't made in the campaign by the end of the campaign date, the matching funds will just rollover to the next campaign.
That is, it seems like the matching is just an arbitrary mapping of new donations to old donations. That is, let's say the matching pool were $500,000, then instead of saying that they'd match the first $1 million at 50%, they could say that they're matching the first $50,000 at 10x. They can call it whatever they want, since it doesn't correspond to any additional money actually being donated.
In fact, depending on the disbursement timing, wouldn't this potentially lead to *less* charity donated by a given point in time, since instead of disbursing the matching fund at the earliest opportunity, they'd hold onto it until the end of the campaign?
So it seems like the GiveDirectly and Shrimp matching campaigns are fundamentally different things, with the former actually leading to greater impact of new donations, and the latter only "amplifying" the effects of donations by convincing people to donate, who wouldn't otherwise donate, under the mistaken impression that their donations are now amplified.
EDIT: I see that the critique of Farmkind's donation matching was made here: https://forum.effectivealtruism.org/posts/D3Gqv9oiDkbmRgGJe/farmkind-a-new-animal-fundraising-platform-is-live-please?commentId=8uxeQiJTLXCziJ6wf, with responses by the organization.
Am I missing something?
The substantive issue is that per my comment, as I understand it, the whole matching mechanism of Farmkind is an accounting fiction, and its only justification is that the campaign will incentivize donors who wouldn't have donated, had they not been fooled by the campaign.
But the wording they employ is also disappointing. First, they seem to acknowledge that their matching campaign isn't doing what donors think it is:
"Our mechanism for multiplying impact is different to how people tend to think that donation matching works."
Yeah, the "people who tend to think" that the match means that each donation is amplified, are the people you're trying to fool into donating to the campaign!
Second, they have a weird line about how the most important benefit of the matching system is cycle of virtuous giving:
"Participants get to direct the bonus funds to the charities they care most about. This is the most direct and clear benefit of getting your donation boosted through FarmKind (although we think it's a far less important benefit than contributing to the virtuous cycle of giving)"
How exactly can a charity predicated on strict utilitarianism claim that the most important benefit of a donation campaign is the beauty of the system of donors working together, rather than actual dollars and cents being moved?
It sounds like they're trying to misdirect readers from the fact that new dollars and cents only come in if people are fooled by the nature of the campaign, and implying that the matching cycle is beautiful, in and of itself.
Hey Mallard, this specific campaign is a counterfactual match from an individual Anthropic employee. If less than $100k is raised, less money will be provided as matching. You're right that in other contexts, this page describes how matching works: https://www.farmkind.giving/support-bonus-fund
"a charity predicated on strict utilitarianism" -- not sure what this is based on?
Upon clicking the donation link for the campaign, the window that comes up states explicitly: "I agree to FarmKind’s terms of use and I understand that the bonus funds are already pre-paid by bonus fund supporters. Learn how our bonus system works here." with "here" linking to: https://www.farmkind.giving/support-bonus-fund.
So at the minimum, your website seems to contradict your claims about the nature of this campaign.
The point about utilitarianism was an aside, specifically delineated from the substantive points, so I won't respond to that, as it would only further distract from the substantive issues.
To your first point -- yes that's our standard T&C checkbox which we lacked the developer capacity to change for this campaign (we're a tiny org). Nonetheless, I can confirm that this match is contingent.
I've been writing a bit about being a writer who is poking around for various kinds of work, which has been just interesting to just enough people to keep doing it. Today I want to talk about long-shot jobs in the field, and what going for them looks like.
For the purposes of this post, I'm also going to be talking about Dwarkesh Patel a lot. Stay tuned to see why!
For definition's sake, let's define a long shot job as follows:
1. Good compensation, similar to what a position one step more junior than a given backend programmer would make. There's a lot of people who want to write, it depresses the payrate, and even someone like me (top percent performer among people seriously trying to do what I do) sometimes struggle to maker serious adult money.
2. Possibility of skirting normal job requirements. Most writing jobs that pay well pay well for reasons that don't necessarily have to do with value added; there's a whole socially-influenced backdrop to justifying the dollars that skill alone doesn't touch. If you are me (minimal education, getting by on moxie and ability) you jump at these opportunities any time they pop up, since most people are trying in "and goes to a school whose name I like" to the job requirements.
3. Fun work. Usually these jobs are unusual in some way, which makes them intellectually fun to contemplate doing.
Enter Dwarkesh, and to a lesser extent David Perell. Perell is the last guy who offered a "long shot" good job in my line-of-sight, and he's important for context here because his job offer was something like this:
*I want very good writers who are smart and very good at writing, the best outlier crazy-good guys in the word. I want to pay them an awful lot of money (I won't say how much) to do work (which I won't decribe). Also they need to have high-end degrees and you shouldn't even apply if you don't have one.*
The writers David gets are going to be the survivors who got through repeated rounds of selection before actually getting work, which is mathmatically going to get him worse writers, but I only actually care much about because I can't go for those jobs.
Now take Dwarkesh, who says this:
*I want people to do a weird kind of writing work for my podcast, but here's my best effort at saying what that work is and what we are looking for. I'm paying 100-220k-ish. Apply here. I want good writers."
I don't know Dwarkesh outside of pinging him for a corporate interview a long time ago (I think) that went nowhere (I think), but I like him. Look at him advertising for a writer and actually caring that they know what the job is, and then apparently also having gone "Oh, what do I want in a writer? Writing ability, I guess?"
That might seem a common enough reaction, but Dwarkesh might have been the only person to ever hire a writer who has ever had it. It's insanity; nobody actually says this. The fact that he's advertising mid-to-senior-dev pay on top of that is also crazy, but believe it or not actually takes a backseat to the other thing.
So you apply to this, obviously (seriously, go apply to this. I'm not getting the job anyway, probably, more on this later). But what are your chances of getting it? Let's say you have the minimum qualifications. What kind of competition are you facing? Well, a million people applied for that. Let's say half a percent are qualified writers who want crazy-ass writer money. That's still five thousand (count 'em) applications, and there's now a good chance he won't even see yours.
My guess is the way he'll get around this is by feeding the big document containing all the apps into an AI and having it do the heavy lifting of giving him a short list, based on either the same qualifications (which were mostly unstated) or another set of quals he's keeping secret. And, I hate to say it, AI's like on-paper qualification. Most of them are little credentialists, at heart.
NONE of this is a criticism of Dwar; I'd love him forever just for this job posting not mentioning degrees. He's a shining diamond just for that. But the first bit you have to grapple with is, statistically, you aren't getting this job, and it's the best one you will see in years.
The second thing to fight with is something like "Can I do this work?". Dwarkesh is hiring someone to do a portion of what he does, and his job is (can't get around it) bizarre. He interviews AI guys and does ad-rolls for a living. That's oversimplified, but to get this job you'd be very well advised to actually have a lot of existing art in the "talks about AI a lot" bin, and hopefully some non-conventional marketing chops that you can show on paper in a short air-table interview setting.
Even if you have those, though, you have to understand that someone who is hiring for this job and willing to pay a lot of money is usually talking about a job function that's important to them, and then ALSO usually talking about something they've done themselves up until now. In Dwar's case, he's also been incredibly-unlikely-outlier-bellcurve good at it, arguably better than anyone similar. There's no way he's doing it "normal".
So now you have a new burden: You need to get Dwar to notice you, you need to get him (or the AI, likely) to shortlist you, you need to be personable/oddly specifically qualified enough to beat the other few hundreds who got this far, and THEN you have to have a long, ethically-mandated conversation about what he's actually expecting and if any living human can actually do it in the way he wants.
And if all this goes well, congratulations. But realistically, it's not going to do that. You should have been a plumber. Everyone needs plumbers! Given that you've already made the mistake and aren't that, this is probably the best you can do.
[Note: Someone does have to get this job, eventually. If you are a writer type looking for this kind of work, I counter-productively urge you to go apply. It seems like a good gig.]
What do you make of The Conscious Wager? I argue we only have a small window of opportunity to build AI due to current existential risks and we should accelerate building it in order for even a slight chance to make consciousness eternal. Even if it increases current existential risks. Because the expected gains are infinite!
https://morologia.substack.com/p/ai-to-infinity-and-beyond
That article's good food for thought. Left a comment that boils down to "the wager's prompt is not descriptive enough of the near-term probability of extinction to be close to my personal assumptions regarding x-risk".
Thank you, I sense that is the problem with the wager as well. I need to make the case that our window is too narrow to delay AI as it is, and that eternal consciousness is worth the bet, even if it means humans must go extinct (since we will go extinct anyways).
Why does AI have to be conscious in order to be useful for conscious beings to extend the lifespan of sentience in the universe as a whole?
I was arguing with an anti-vaxxer on Twitter and accidentally got him excited about a preprint from my labmates lol
Background: nuclear DNA damage and other stresses are responded to by p53 (the famous "guardian of the genome"), but DNA leaked out to the cytoplasm where it doesn't belong is responded to by cGAS-STING signaling, which causes an inflammatory reaction.[1] The system is interesting; bats have less of it, which may be part of why many viruses that kill us don't kill them (less inflammatory over-reaction; this in turn may be an adaptation to flight, which is metabolically intensive, so they don't overreact to their own self-generated DNA damage).[2] My labmates found that armadillos are missing cGAS-STING signaling entirely, suggesting that they're probably doing something else, which may turn out to have to do with their getting little cancer.[3] Although cGAS-STING handling DNA leaked to the cytoplasm is good, chronic activation could rewire things and be oncogenic.
The latest anti-vaxxer fear (a possible one, but unproven, but treated by them like it's proven) is that vaccines containing DNA plasmids (e.g. DNA vaccines or contaminated mRNA vaccines) go into the nucleus and integrate into the genome, or stay long-term in the cytoplasm chronically activating cGAS-STING (or both?).
So anyway I was arguing with an anti-vaxxer about something else (old other-vaccine-and-autism claims, not new plasmid-vaccine-and-cancer claims), and he dug around and found the armadillos paper, and tweeted it as a "gotcha" like "see your own lab agrees that chronically activating cGAS-STING would be bad!" (like that was the contentious part and also on topic). It then turned into an argument about whether upregulated genes that are downsteam of many different possible activators including cGAS-STING are proof of cGAS-STING activation (no) or whether the intended inflammation soon after a vax is proof of long-term inflammation (no).
I think he also thinks my labmates proved armadillos losing cGAS-STING made them not get cancer? (We'd never claim such a big proof without a lot of functional experiment evidence!)
But I'm glad he liked their paper, I told my labmates and they were amused too :)
[1] Zhang et al. (2025). Regulation of the cGAS-STING pathway. Annu Rev Immunol. https://doi.org/10.1146/annurev-immunol-101721-032910
[2] Xie et al. (2018). Dampened STING-dependent interferon activation in bats. Cell Host Microbe. https://doi.org/10.1016/j.chom.2018.01.006
[3] Schaffer et al. (2025). Loss of the cytosolic DNA-sensing genes CGAS and STING1 in armadillos (Cingulata). bioRxiv (preprint). https://doi.org/10.1101/2025.05.13.651073
I've enjoyed looking at Michael Burry transition from hedge fund manager to Substack grifter. He charges people for a newsletter to share his doomer perspective on the stock market. Being a fear mongerer is his stock-in-trade. I think this will do wonders to his personal wealth, and it might spur more hedge fund managers to pivot from the dying single-manager platform to selling their opinions to the public. Watch this space!
He supposedly has a net worth in the $200 million range, I don't think that writing a Substack newsletter is going to meaningfully add to the pile.
A better model for him is that he's writing a Substack for the same reason I'm writing comments; because he's a grumpy dude with a brainful of opinions that he really wants people to listen to.
In your model, why does he charge for subscriptions and paywalls the majority his content?
https://www.youtube.com/watch?v=g4Uv4ftekaI
Why would you think it to be more profitable to write a Substack than to work in the giant $100 trillion money machine that is the US financial system?
Well, when you sell opinions, you make money whether they’re right or wrong, whereas in finance you lose money when your opinions turn out to be wrong. At the very least the earnings from selling an unstick are much lower variance than from actual trading. People who are wrong at least don’t have to pay their audience to listen to them.
> whereas in finance you lose money when your opinions turn out to be wrong.
Only if you're betting with your own money. Hedge fund managers take a cut whether the fund goes up or down.
Either way, given his net worth, he didn't seem to have any trouble with his old job. I guess entertainment is a sector more resilient to stock market crashes, but still.
>whereas in finance you lose money when your opinions turn out to be wrong.
That has been soundly disproven in 2008. The new reality is: If you are sufficiently wrong, the government will bail you out.
Tens of thousands of paying customers is a more stable way to make money than it is to short Palantir and Nvidia for nominal values of $1B, at least for now.
I'd like comments from programmers on this please! Are claims of AI making programmers more productive basically bullshit? What exactly is going on here?
https://open.substack.com/pub/mikelovesrobots/p/wheres-the-shovelware-why-ai-coding
At work, my productivity is bottlenecked by meetings and security theater and other nonsense. At free time, my productivity is bottlenecked by the lack of free time, and the lack of energy when that free time finally arrives.
AI made me more productive in the sense that I did some very small projects I wouldn't have enough time and energy to do otherwise:
http://bur.sk/2024/treasure/
http://bur.sk/2025/quest/
One theory (take it with a mountain of salt) I just came up with that can explain why we're not seeing productivity impacts in the aggregate level:
The vast majority of developers are useless or barely productive. Most software is produced by a small share of actually good developers.
AI is not yet good enough to be clearly helping this share of good programmers.
Where is the shovelware? Easy, it's sitting on my phone in the Claude app. Making a simple mobile game in a few minutes over a beer is a great party trick!
And any time I have some half-formed thought of a game idea, I can test it, iterate on it a few times, and... rapidly figure out it's garbage, typically, but without a sunk cost of having spent a week coding leaving me sentimentally attached. Eventually I'll hit upon one that doesn't suck, maybe.
Claude is enabling me to do things I wouldn't otherwise be able to do at all. I'm a sysadmin by trade, in the pre-AI era the most work-related coding I would do is powershell scripts. Certainly there have been times when I've got to the end of a Claude Code session and realised I could have hand-written the script faster, but those are few and far between. But also, I can do so much more - for example, had a feature request for a SharePoint site for something that couldn't be done out of the box. That would mean hiring a SharePoint developer at great expense to code it up in SPFx... or just ask my good friend Claude. I had for several years considering skilling up as a SharePoint developer as demand is there, glad I didn't waste time on that!
So many solutions coded up across disparate platforms, languages, API's that I just do not have time to learn myself. Perhaps an experienced SharePoint developer could have coded up that SPFx app faster. And an experienced full stack Azure/React developer could have made [business iPad app] faster. And an experienced C#/Azure dev could have coded [business logic app] faster... if they already knew the Dropbox API inside-out. Maybe.
It's like having a team of junior developers at my beck and call, who know every language and architecture. Sometimes they make dumb mistakes, and sometimes have rocky handoffs to a different junior (ie when context fills up and things just aren't the same). Right now, I still need to know how to code to occasionally go in and help on tricky problems, but the writing is on the wall, 100% hands off vibe coding is coming.
And now we have Opus 4.5 who feels less like a junior. It can even make convincing 3D models out of primitives in Godot!
But the shovelware he is talking about, is not about you being able to produce more. You're only one data point.
I am not in any sense a programmer - I write papers in LaTeX, and that’s about it. But with AI assistance, it’s now really easy for me to include diagrams in my papers that I wouldn’t have included in the past. And that article says “where are all the Tetris clones?”, but with a small homework assignment I got a dozen students with no coding experience (one of whom had never even heard of Tetris!) to code up Tetris clones (and the one who hadn’t heard of Tetris before told Claude to make hers run slowly and she now has it on her phone as a relaxation tool). I’ve also used it to get clones of some of the nytimes word games so I can send one to my partner after I’ve finished it, even though the actual site doesn’t let you restart the game once you’ve solved it.
Obviously none of this is anywhere near impacting the productivity of an actual coder, but it is evidence that we can get lots and lots of minor apps written if non-coders start learning to think about what problem a program could solve for them.
Whether it makes an experienced programmer more productive is a different question to whether it makes a newbie more productive.
I'm recently retired so I haven't really used it in a production environment, but it's definitely made me vastly more productive on hobby projects. The colleagues I've kept in touch with agree and their opinion of stuff like the above is that only good engineers actually find it useful. Mediocre-to-bad people get themselves into trouble with it and just wind up with a pile of code that they don't understand. I think you need well-developed instincts for modular design in order to be productive with LLMs. You can't just point it at a codebase and say "implement feature x". You have to architect the solution yourself and just have AI solve small bits at a time.
This is how I approach using the AI as well. When I need some function that does one well defined thing that I could figure out in 15 minutes, instead I spend 1 minutes writing a good prompt and get my function. I have vibe coded a few webapps (I don't know the first thing about web development) and usually I end up with about 1000 lines that do mostly what I want that I don't understand and don't really want to spend time figuring out, good enough for a dopy little webapp but not how I would do something at work.
An EE I work with has learned that the AI is great at reading hardware datasheets for you though and quickly answering questions. I find that pretty slick.
hi, I have a lot of opinions on this. @Loominus Aether posted one of my substack posts on the subject below (https://theahura.substack.com/p/averaging-10-prs-a-day-with-claude) but I've also written other things on the subject, e.g.
https://theahura.substack.com/p/cathedral-builders-probably-shouldnt
https://theahura.substack.com/p/were-all-senior-engineers-now
https://theahura.substack.com/p/coding-agents-can-manage-other-coding
The short version is: you need to parallelize.
Most people who are frustrated with coding agents use them *very* poorly. The correct way to use them is to farm tasks out to them and then go do other things -- often, spin up other coding agents to do other tasks / features. If you cannot get comfortable with that, coding agents just aren't for you. Which is totally fine, but it's a critical distinction.
Beyond that, you need to put in the work to get your coding agents to be better at software engineering tasks. They need to write tests. They need to read and write docs. They need to do research and create plans and get approvals (where you, the senior engineer, are in the loop). Coding agents are really best thought of as untrained interns, so you have to train them.
Soft plug: my company, Tilework, has spent a lot of hours thinking about this exact problem. We are pretty opinionated about how to use coding agents and have gotten some fantastic results across a few different companies right now. If you want to give it a try, check out https://github.com/tilework-tech/nori-profiles, and feel free to reach out at amol@tilework.tech
>Coding agents are really best thought of as untrained interns, so you have to train them.
Do you have an impression about how much they have changed (presumably improved?) over say the last year?
My feeling is yes the agents got better (there barely were any before 2025) but the main driver was the LLMs. For the agents that can switch LLMs, it's very obvious how much of a difference the LLM makes, in my view much more than the agent. Supporting infrastructure for sure got better, too (sandboxes, MCP, skills, etc).
This is kind of supported by the fact that the big coding benchmarks (SWE-bench etc) seem to use very simple harnesses (100 or so lines of code, I believe) and show huge differences between LLMs.
Many Thanks! I haven't used agents myself yet.
>This is kind of supported by the fact that the big coding benchmarks (SWE-bench etc) seem to use very simple harnesses (100 or so lines of code, I believe)
Agreed. While I'm speaking from ignorance, my gut reaction is that a plan-and-execute loop, _given a sufficiently reliable LLM_, should be very simple. Not much more complex than
while(goal not achieved)
{
query LLM for a plan to get from the current situation to achieving the goal
execute the first step in the plan
check to see if the step succeeded or ran into problems - add to LLM context
}
( with a lot of tool-handling complexity buried under "execute first step" )
But, generally, I'd expect almost everything to depend on how good the plan from the LLM is, which echoes your
> show huge differences between LLMs.
What is a coding agent? My only experience using AI is using the prompt window that comes up when you go to https://chatgpt.com/ and type things into the window. Not clear to me how you "make agents".
So just to take a step back it's worth understanding what an "agent" is. Formally, an agent is 4 things:
1) a large language model. This can be any of the popular ones -- chat gpt, Claude, and Gemini are the big 3
2) a set of tools. This is usually a set of scripts that can be called with the right parameters. If you understand what an API is, you can think of tools as an API exposed to the agent system
3) a prompt. This is generally a few paragraphs of text that explain what the goal of the agent is and how it should behave. Normally this is split into a "system prompt" and a "user prompt", but you don't really need to worry about that distinction
4) a loop. The core infrastructure of the agent is a simple loop where the LLM output is fed back into itself as input. At each step, the LLM decides if it needs to call one of the tools it has access to. It will keep going in this loop until the model decides that it is done
From here, you can get a bunch of different kinds of agents. Most of the models you interact with on the web are agents. For example, if you ask Claude to do some research for you, it will go look up a website and then respond. That's two turns of the loop. Or you could have a special deep research agent, which looks up many things and consolidates those findings, and then uses what it finds to drive further research.
A coding agent is an agent that is specifically good at writing code. Most of its tools are related to reading and writing files, the underlying LLMs have been trained on a lot of code, and the prompts are generally code forward.
Hopefully that helps explain some of the distinctions!
OK so if *I* want to set up an agent how do I do it? For the tools, is that something that someone else in the world has made already?
The easiest way is to download a tool like Claude Code, which is out of the box a coding agent. But it lets you modify how it works in all sorts of ways, so that you could make it behave like some other agent. There are also a bunch of startups selling agent builder tools, but none that I would care to recommend. OpenAI also let's you make "custom GPTs" which are kinda the same thing
It's all in how you use it. At least one commenter here is getting 10 PR per day out of it:
https://theahura.substack.com/p/averaging-10-prs-a-day-with-claude
I'm trying to mirror his technique, but even before then I'd say I am at 3x productivity
I could love to see a bunch of actual prompts, and results. Just as 2 examples. I asked some LLM about jj (jujutsu), the new "hot" git replacement. It failed 9 times in a row. It gave me commands that didn't work and I'd paste the error message "no such command `jj foo`" and it would apologize and then proceed to make up some other wrong answer. I get jj is new, so is the majority of software I write.
A non-code example, Google's Thanksgiving Day Doodle lead the an LLM prompt on how to plan a Thanksgiving dinner to be ready on time. The answer it gave looked nice, but if you read the details it had various inconsistencies. When questioned on them it just dug itself into worse holes.
Positive: I asked an LLM to make a single file "hello triangle" in OpenGL on Linux. That worked. I asked for the same for DirectX12 on WIndows. That took several iterations but it worked. I asked for the same in Vulkan. That failed. The LLM tried 12~15 steps and finally gave up and said "please search online for a working example"
I had another today where it told me googletest (an open source C++ testing framework that's been widely used for 10+ years) threw exceptions when skipping a test. It does not. It just hallucinated that.
In any case, maybe if I searched the net enough I could find someone who documented what they made and what each prompt was. I would certainly love to be 3x faster.
https://12gramsofcarbon.com/p/i-successfully-recreated-the-1996
I think there is no solid consensus on this because there are several very different experiences people are having:
-pure vibe coding with no experience
-vibe coding with no direct experience in the specific thing, but as a generally competent SWE
-having the AI do simple tedious work
-having the AI do legitimately impressive but narrowly scoped work
-having the AI do impressive sprawling work
-experienced people using the fancy autocomplete mode in IDEs for cool-but-not-earthshaking speed gains
Those all have quite different failure and success cases. (To be clear, I'm saying that all of them can have both successes and failures, just all looking different and some succeeding more commonly than others).
Furthermore, there's prompting style. Some people are typing "make the webapp better i want the user to understand it better" and some people are typing multi-paragraph essays.
From my perspective, the biggest gains are available when you need to interface with a big complex system in a relatively straightforward way, and just correctly using it *at all* is a huge challenge. You can get a working example running on your own machine, semi-tailored to your particular circumstances, on absolutely anything. That's a really big deal.
The worst failures come from piling vibe on top of vibe. If you are not periodically stepping in with your human judgment for a little refactoring and cleanup, then your vibe coding is going to quickly collapse into misery, no matter how impressive it started.
I feel like the main way I use AI isn't listed there. I find it very useful when working with a library I'm not familiar with. Sure it might be possible to find everything I need in the documentation, but AI can easily tell me stuff I missed, and I can easily check if it tells me something that isn't true.
And another big thing is that I tend to get stuck and know several things I *could* do, but not be willing to pick one and move on. I'm not sure how good the AI is at knowing the best option, but even flipping a coin would probably be an improvement over what I had been doing.
https://xkcd.com/1445/
I recently vibe coded a relatively featureful guitar fretboard learning app in a couple days of unfocused, occasional prompts. This isn't a case of 25% more productive. It's hard to measure, but would have to be well, well north of 1000%, because as a mostly backend person with a lingering, instinctual revulsion for javascript, I could not have created that app manually in any amount of time. I would have lost focus, gotten bogged down, and given up. It absolutely has enabled new applications. That said, this kind of greenfields, "make a self contained app" is a domain where current models really shine.
(Here's the app, if you're interested in seeing the results: https://fretu.de )
I've been tinkering with it and I think it's tough to say. The most useful thing it's done for me is the times it caused me to stop procrastinating, independent of how helpful it was. The biggest issue I have with it is that it often takes long enough for me to get distracted, at which point it might be a negative.
Cursor's tab-completion thing, which is basically just a smarter auto-complete, works pretty well, for the types of things that I might otherwise do a quick regex find-and-replace to do, but it's a bit error-prone at times.
This is a really good point. You'd think getting used to "maybe the AI can solve this for me" would be bad for motivation, but actually, that being an option has gotten me through a couple of really nasty debugging walls that I might have otherwise unhealthily avoided for quite a while. (With the AI outright solving it for me in some cases, and just getting me productively thinking enough to do it myself in others).
I don’t have any metrics on this, but I have high confidence that good use of AI makes me much more productive.
The AI autocomplete means that once I have sufficient code such that there is a style and pattern established, the completion is correct >80% of the time, and the remaining 20% it is simple to correct. If the area I am working in is too complex or novel to use the autocomplete, I disable it.
The “vibe coding” style is more hit or miss, and there is skill in recognizing from the outset if AI will be able to actually solve the problem. My general strategy is to define the overall structure, then review each point interactively, one at a time, until I sign off on it. Once all that detail is in the context window, I can often tell the AI to proceed with the work to great effect.
That being said, for entirely new features I find getting AI to do quality work is near impossible. So I don’t waste my time on that. Because AI constructs features in a one-shot manner, it has no preference for incrementalism and thus no preference for modularity. Anything that requires design work should not be left up to an AI- they might be better than a junior engineer at it, but far worse than senior let alone staff engineer.
So the productivity gains are all about playing to its strengths.
I've found AI is handy as a "micro coding" aid. By that I mean not microcoding, but in effect an enhanced web search whose result clearly summarises and explains a few lines of code to achieve a small part of a larger program.
This is especially useful for tricky or obscure APIs which lack up to date (or in some cases any) documentation. A recent example was packaging a binary file into an OCI (Open Container Initiative) document for storing in a Docker repo (well actually a Harbor repo). It turned out that instead of struggling with the intricate code for "hand rolling" the document layers etc, the simplest and most reliable way was to use a utility called oras, and AI searches gave examples of the command options, which I couldn't readily find with simple web searches.
I found GitHub Copilot particularly helpful for a particularly unintuitive mocking library in my unit tests.
I think for me I'm being like 25% more productive, and it has fundamentally altered my problem solving strategy. It's often faster to go to the AI than to start googling. That is not an empirical metric however, although, I don't know how empirical you can get, since there is no standard unit of work for software developers.
I have a friend trying to build a business around the idea that projects that previously took $50k and a month or two to complete, can now be done for $3k in a week, due to AI. He's pitching it as 3k 3days as I recall, something like that. He says the idea is working, which would completely invalidate that study.
What does your friend this of this Substack essay I posted? Where's the "shovelware", as he calls it?
I don't know if my friend bothered to read the post (I did), but I told him the gist and here's his take:
"Easy to test. Task two devs with making a simple AppScript script that looks for files in a google doc folder and sends them to slack. Neither has touched AppScript before, or the Slack API. One uses AI, the other doesn't. Obviously the dev who has AI write the script will finish much faster.
Now, if you tell Claude Code to fix a driver in the Linux kernel, probably the graybeard with 15 years experience is much more productive than Claude Code, and Claude Code probably won't help you much."
I agree with him, AI really shines when you're handling some tech you have never used before. Recently I finished a pretty complex task about integrating a PHP backend to read and write from Azure Blob Storage, no idea how I would've done this without AI, since there are no libraries to use (Microsoft used to have one, but they deprecated it). It would've been a real grind working out how to talk to that API by reading the docs.
I think I disagree with my friend about the Linux kernel driver case, the Linux kernel was in the training set for the LLMs, I bet Claude Code would do quite well on that task (depends how messy the Linux kernel is).
I think the strongest argument in the post is that we don't observe that the rate of software production has increased dramatically. But then, AI doesn't make development trivial, just easier. It's not gonna change the amount of devs that have the time and energy to start churning out apps and sites as a side-hustle. I personally find it difficult to code in my off time, even with AI, coding is hard and tires out my brain.
*think of this Substack post
(No edit button)
Edit is hidden in the … menu
It doesn't show up for me. It has sometimes though. It might have to do with how I access substack.
How do you measure your productivity and come up with this 25% metric?
Purely a vibes based gutfeel, like I said, it's not rigorous. I have no fucking clue how I could measure my productivity. Hmm, we use fibonacci story pointing, maybe I could figure out how to look at that, but I don't know, I don't try to maximize my productivity, if I feel like I did enough for the day I just stop. So even if AI did make me more productive, I'm not sure it would show up there. And I don't use AI on every task.
I’ve always seen sports as a kind of mate-selection thing, with men and women competing and the winners signaling their genetic quality. Maybe people don’t consciously think of it that way, but in the end, they behave as if that’s what’s happening. So, allowing trans women just breaks that original purpose.
This doesn't seem right to me. Wouldn't your theory also suggest that we ban gay people from sports?
I loved sports, soccer, basketball... Except for in high school, w/ moms and cheerleaders, I don't remember many female observers. AFAICT we were all just playing for fun. If you want to say it's guy on guy competition then I couldn't disagree with that.
I think a simpler explanation would be that the main purpose of women's sports is to allow females to compete despite not being as good as males. Cf the paralympics, weight classes and age classes.
Is the argument here that people participate in sports to find romantic/sexual partners? Or watch sports to do so?
So the argument here is basically that people kind of look at physical fitness as a factor in mate selection. In other words, having men’s and women’s categories in sports fits neatly into this kind of preconceived mental category we all have. And when you break that by allowing trans women to compete in the women’s category, it kind of disrupts that symmetry that people have on a subconscious level. So of course, they might not really understand exactly why it feels off to them, but I think it shocks them because something they’ve internalized on a very deep level suddenly gets broken.
I don't understand the logic pipeline from "people select partners partially based on perceived physical fitness" to "trans women competing in the women's category shocks people... because they can't get pregnant?" Wouldn't this also predict people being uncomfortable with competitive categories for post-menopausal women, or any mention of athletes being unable or unwilling to reproduce? Wouldn't they also be uncomfortable with non-sperm-producing trans women in men's sports?
My intuition is that the "intuitive take" arises from a feeling of "I want to see tits on the telly but the trans women don't have them, which makes me uncomfortable."
Trans women often have nice boobs actually :) But yeah I always wondered if trans acceptance will go up as the transition process get better. But the pushback over trans women in sport was especially strong. This why I think there is something more going there.
OF COURSE acceptance goes up as the transition process gets better. The entire controversy is driven by transgenders who do not pass or do not pass easily.
There’s this kind of intuitive pipeline in people’s heads: they look at women’s sports partly as a way to kind of subconsciously assess who has strong genetic traits. And that doesn’t mean they’re literally sitting there thinking about mating, but it’s just an underlying instinct. And when a trans woman wins, people kind of feel like, “Hey, that doesn’t fit the pattern,” because they might think that someone who has, let’s say, a Y chromosome or a different starting point has an advantage that doesn’t match that intuitive idea of genetic fitness. So people feel like the person winning might actually not have a “better” genetic makeup than the person losing, and that just feels off to them. So that’s kind of the whole pipeline: people are looking for who’s got the best genes, and when that pattern breaks, it just feels really uncomfortable.
Huh, this is interesting. You're arguing that our perception of "fairness" in sports is tied to honest signals of genetic fitness. This explains the trans & intersex athlete controversies pretty well, and maps onto the weird ways we've drawn anti-doping regulations.
In what ways do people behave like it is?
Successful athletes sure do seem to be having a lot of sex with high-quality partners.
The Olympic Village famously needs 300k condoms for 15k athletes over 3 weeks:
https://www.nbcphiladelphia.com/paris-2024-summer-olympics/olympic-themed-condoms-olympic-village-paris/3920406/
My impression is that successful male athletes are generally considered attractive whereas successful female athletes are not generally considered particularly attractive. Would you disagree?
> Would you disagree?
I disagree! On a personal aesthetic level, naturally, but also on the facts.
There's a (now ancient) Oktrends post that pointed out that being polarizing is great for dating. People rated average or below, because half hate them and half love them - so even if ~70%-80% of men are intimidated by women above a certain fitness or muscularity, you can still do pretty well if the remainder loves it.
And indeed, I've had a few exes who were IFBB pros and / or influencers, and they get *constant* male attention, every time they step out the door (and of course, on any of the apps that's multiplied a hundredfold).
>I've had a few exes who were IFBB pros and / or influencers, and they get *constant* male attention,
That's just because they're fit and attractive, not because of the social status the achieved by virtue of winning competitions. Look at the WNBA: it's not like those women are dating movie stars.
Men need competition to signal status, women's status comes purely from their easy-to-observe physical attributes.
So would you expect pairings of successful female athletes with highly attractive males from other fields (e.g. music and modelling and acting) to be as common as the other way around? I'm trying to think of how this could be tested.
> So would you expect pairings of successful female athletes with highly attractive males from other fields (e.g. music and modelling and acting) to be as common as the other way around? I'm trying to think of how this could be tested.
Actually, as near as I can tell, it's the other way around - the hotter and fitter and more extraordinary a female athlete is, the more likely their husband is to be a misshapen potato man.
You can see this firsthand if you ever watch American Ninja Warrior - all the partnered top women (Flex Labreck, Meagan Martin, Sandy Zimmerman) all have basic potato man husbands, in the sense that the women are 10/10 on fitness and capability, and their husband is "just some schmuck" with essentially zero visible redeeming physical aspects.
Women don't optimize on looks to nearly the degree that men do, they optimize on status, and how a given man makes them feel.
And while I'm not quite a potato man, I'm generally a decade or two older than my paramours and I'm certainly not visibly turning female heads as I walk down the street, yet I end up with them too, so...one more point on that side of the ledger, I guess.
I don’t think that’s contingent on placing well in any given competition the way this implies. Excellence in most any field is attractive, not just sports, and athletes tend to have uh athletic builds which I’m sure helps. The competition itself being a mate-selection thing doesn’t really scan with my view of elite athlete psychology at all.
Do last-placers also have a lot of sex in the Olympic village? Is there a way to know?
> Do last-placers also have a lot of sex in the Olympic village? Is there a way to know?
Yeah, I think your view is probably directionally correct, that excellence is itself the attractive factor rather than literally winning.
Forget last-placers - even an Olympic Qualifier who doesn't actually end up competing is a literal paragon of human excellence and capability, very likely to be the fittest and / or strongest person any regular person has personally ever seen, and have no difficulties attracting attention from the opposite sex if seen in their domain.
I never Qualified (although I count a few qualifiers in my social circle), and was only ever regionally competitive at my sport. Even today, more than a decade later, I still turn heads and get attention every time I go to the gym, doing ~60% of what I ever did while competitive, because that 60% is so far above what everyone else is doing.
In my opinion, one of the most serious problem for human beings today, especially in first world countries, is sleep problems. Sleep deprivation, insomnia, and nightmares. All other basic physical needs are obeyed. We can have comfortable temperature (or adequate clothing to protect us from weather conditions), and most people have enough food, including all the necessary ingredients (proteins etc.) But most of us don't have enough sleep.
My impression is that a lot of the problems with "enough sleep" are related to the use of artificial light (including screens).
Yes, maybe.
Absolutely. I think a lot of people would make radical changes to their career/lifestyle if they realized what an impact it can have. Most fitness regimes will be a sideways move if sleep is sub standard
I find this unlikely… in the past people would have to wake up several times a night to put wood on the fire, or would sleep miserably in the heat.
Also people have less kids… which wake you up.
But then again, I sleep like a baby, so maybe it’s just me.
> In my opinion, one of the most serious problem for human beings today, especially in first world countries, is sleep problems.
Isn't this just the direct corollary of "80% of americans are overweight with 50% obese," because food is so solved that approximately everyone is fat?
The reason everyone is fat now is that food companies have had literal Phd's working on junk and fast food for decades, making them as delicious and "moreish" as possible - superstimuli, in other words.
Similarly, every company with a major app has thousands of Phd's on the other end of that app, modeling their customers and optimizing the content to attract as many eyeball-hours as possible. Remember Neflix CEO Reed Hastings saying "our competition is sleep?" This is why, just like 80% of people are fat, people spend 10-11 hours a day on screens now:
https://imgur.com/a/PipMWEf
This is why everyone doesn't get enough sleep - they're so addicted to their Skinner boxes with attention superstimuli on the other end that they sacrifice sleep quantity and quality for one more hit.
The solution is the same one as the food one - don't put yourself in fully adversarial situations with thousands of Phd's optimizing against you. Don't eat fast and junk food, and don't look at screens for 10 - 11 hours a day.
It’s not just that - white collar job lifestyles encourage people to sacrifice sleep to do extra work, often leading to spirals.
I had not enough sleep even when I was a child. There was no internet, and we had no TV. So no, it is not this. At least not only this.
I think that people were deprived of sleep even before the internet epoch.
Also, if you are adult (or even a high schooler), you cannot just stop looking at screens. Today you need internet for too many things, some are impossible to obtain without one. If you want to stop looking at the screen, you need a time machine, or you need somebody else to take care about your issues, or you need to die.
I was also an insomniac as a child, and while we did have a TV, I think I was only allowed to watch PBS when my family first noticed this.
I've recently found that when I've been reading books to my kid for long enough, my mind can start wandering. (Let's ignore for now whether that might be a bad thing). I'll suddenly realize I've been on flawless autopilot for several pages. This is surprising because for me, "mind wandering" means internal monologue - basically the same kind of cognitive process as reading. And sure enough, when I deliberately try to think my own thoughts while continuing to read, the thoughts barely get a half-second foothold before being washed away. But when it just happens, I can have long, coherent chains of thought, including planning things for the future. Probably relevant that I've read these books many times, but also I definitely don't have them memorized - this is happening even with non-rhyming stories with several dozen words per page.
I'm not sure there is much productive insight to be had here, but I just thought it was a really fun little cognitive quirk, the sort of thing that ACX readers might find interesting. I'm also interested in hearing about any similar experiences.
(Also, I wonder if the "second screen experience" lifestyle is going to make the younger generations really good at this kind of thing?)
Thanks for the replies; exactly the sorts of things I was hoping to hear about! :)
I had this funny experience in school, at the age of 14 or so. I was in history class, which I found hopelessly boring*. At some point we're going through the next part in the book, and the teacher says "hey Anon, go ahead and read out loud this couple of paragraphs."
So I go on reading for 2-3 minutes, and my mind gets SO distracted, and I'm focusing on the tone of my voice, the particular text font, how to pause for punctuation, all while the words flow automatically out of my mouth, and I am so glad the teacher's asked me to do that so he'll probably be okay with my "engagement in class" for the day and stop bugging me with stupid questions, so I finish the page and look at him with a smile and a nod, and he goes with a strict face:
"Good. Now, in your own words, how would you summarize this excerpt?"
And I'm like shitshitshit how fast can I speed-read this and mumbling while realizing that "sorry, I wasn't paying attention" won't sound very good as regards my attitude in class, and it felt like ages until he got so mad at me and asked someone else.
It felt so weird that my classmates were all convinced I was deliberately making fun of him and being a jerk because, dunno, I only liked maths and wouldn't just answer a bloody question in history class out of spite.
* I got wiser with age
hahaha oh man, that's hilarious to look back on, but I am definitely sorry you had that experience.
When I was a child, my parents would read Lord of the Ring to me while I was learning vocabulary for French. I knew the book so well by heart that I would call them out for any reading mistake that they would make, but afterwards I would also know the new French words.
Probably it wasn't the most efficient strategy, but I was very good at school and it was good enough for me.
This also happens with making any repetitive announcement; you say the words as reflex, without thinking about them. It can be a problem when the current announcement has a slight variation from the standard.
This happens to me when I'm reading a book for myself, so I can imagine it'll get worse when I read to a child. Usually I realize I've been thinking about other things after a couple of paragraphs on autopilot, and either I pick it up from the top or I decide the mind wandering was a signal that I should stop reading.
I had the same experience and was quite surprised, when I noticed it for the first time.
> Try another book? Or maybe you're anxious about something and not able to enjoy this experience.
Very young children _love_ being read the same book for the umpteenth time. Parents do this for their children's benefit, not for their own enjoyment.
When I feel like making EA-minded charitable donations, I usually give $500 to the Malaria Consortium. Should this 50% match in Item #2 make me donate to Rwandans instead?
I will definitely not be donating to the shrimp, though. The only thing that I think about when I see pro-shrimp substackers is "Mmm, I'd love a shrimp po-boy right now."
It depends on if you think the donor wouldn't have donated if not for the matching drive.
I don't think the match is enough to make you change your donation.
"One common defense of letting trans people play as their chosen gender in sporting leagues is that - although trans women may have a biological advantage over cis women, and this might make them win so often that it stops being fun for cis women, most sporting victories involve biological advantages"
This really isn't an issue I'm following closely, but isn't the whole premise of women's sport that they don't allow the biological sex advantage specifically? This just sounds like you don't accept women's sport as a category in principle.
Which makes sense I guess, it does seem strange to have a sports contest especially for a demographic that's worse at sports. But I don't think sports fans are going to accept "sports are just generally irrational" as an argument.
The whole theory discussion seems to ignore what is, in numbers of participants, easily the largest part of organized sports: amateur/recreational leagues.
The number of adults who at least sometimes play in organized sports leagues has been booming in the US (as everyone who's been doing it for a decade or more knows firsthand). Estimates vary but this category now includes somewhere between 15 and 19 percent of US adults.
That entire huge ecosystem of leagues -- which includes both team and individual sports -- is structured as divisions or leagues that are restricted across a range of different ability levels. It would obviously be ridiculous and unenjoyable if a 25 year old who played high school varsity tennis was in the same brackets with a 65 year old who'd dug out an old racket after decades without touching it. So that's never been how it worked and nobody thinks it should.
Hence for the millions of Americans who pay to participate in organized amateur sport, the idea of a 6'3" former male who played mens college ball joining a local women's basketball (or whatever sport) league has always been prima facie ridiculous. The apostate progressives on this topic who I personally know, which is several, are all either longtime players in such leagues or married to one. Multiply that practical-real-world-experience reaction by millions of voters and this is why the specific question punches so hard now at the Democrats.
> This just sounds like you don't accept women's sport as a category in principle.
Scott is arguing *against* this argument in the post.
> it does seem strange to have a sports contest especially for a demographic that's worse at sports.
Parasports are a whole thing. The main difference is that there are way more women than there are people with any specific disability, and that it would be considered offensive to imply femaleness is a disability, even though in the context of athletics it essentially is.
(You could argue weight classes in combat sports also count as this, though I think it's not quite the same).
I endorse Ozy's post on this topic (also paywalled): https://thingofthings.substack.com/p/on-intersex-athletes-in-the-olympics
>Scott is arguing *against* this argument in the post.
Scott is arguing that since most sporting victories involve some sort of physical difference, it should be okay to allow trans people in women's leagues, since we shouldn't really care about physical differences. This argument implies that we shouldn't have women's leagues at all, since the whole reason that there is a separate women's league is physical differences.
Scott is arguing "against" this argument in the sense that he doesn't accept the implication, but his argument still *has* that implication.
And that implication is also the answer: If you want there to be women's leagues at all, you've inherently decided that physical differences do matter, so you can't just say the opposite.
Well said.
You can go watch Claude Opus 4.5 playing Pokémon Red:
https://www.twitch.tv/claudeplayspokemon
It sucks so hard, it's quite the contrast with the predictions of imminent superintelligence. Seedier corners of the internet are jibing that Claude Opus 5 will require the entire energy output of the sun, which will enable it to progress to getting stuck in the Safari Zone (this one is seemingly stuck in the Team Rocket hideout, which would be around the midpoint of the game).
I feel like we weren't groping forward so blindly in previous scientific breakthroughs. For the nuclear bomb, for example, Szilard worked out on paper that a nuclear chain reaction is possible (then it took the Manhattan project to actually build a nuclear bomb), but here, we don't know that superintelligence is possible, or that LLMs can get there. Will LLMs get there? To AGI at least? Will they peter out? Will we need a new paradigm instead, and if so, how long will it take to discover that paradigm? So many question marks.
I needed a guide to get through that bastard Team Rocket hideout as a kid, but eventually attained general intelligence (well, it could be debated) with a few more years of brain growth, so I'll let Claude off the hook for now :)
If you've ever tried making a model that can play pokemon, you'll see there's a tradeoff between more upfront information and less generality. You can tell it what a pokemon center is or what the end state looks like, but then it will not learn that for itself which means it will not be able to play other games you dont already know how to play. The only true general model is abstract and works on abstract things like direct memory access or pixel-level analysis, it must not be poisoned/biased by human knowledge unless it has discovered it itself. I don't think you will ever get LLMs to play arbitrary games unless it already has the knowledge for them, or the architecture has a way to use human knowledge abstractly and independent of the actual medium.
Far from knowing whether superintelligence is possible, I think we don’t even have a good theory of what it is. There’s a lot of rough ideas people throw around, but no model of intelligence is anything like a model of fluid dynamics or a model of gravity.
LLMs have been a huge improvement over the previous paradigms of intelligence that people were working with for most of the past 80 years, but they’re clearly missing lots of things, and they’re an unusual model that can’t really be understood, not even at the level of a differential equation.
Yes Claude has terrible vision, because Anthropic isn't trying very hard to make it better. Gemini has better vision and does better on these sorts of tasks.
Anthropic is basically all in on making the best possible machine learning and coding agent so they can achieve self improving AI. To the exclusion of everything else (math, vision, video models, image models, world models etc.). Time will tell who was right.
> I feel like we weren't groping forward so blindly in previous scientific breakthroughs. For the nuclear bomb, for example
IMO that's more a feeling from hindsight than it probably felt at the time.
Like the Manhattan project simultaneously pursued three different enrichment technologies because it wasn't sure which (if any) would pan out, and I believe part of why we had three bombs at the time of Hiroshima and Nagasaki (rather than immediately use the first bomb we made) was because there was a real risk that one or more of them wouldn't work. Also famously that whole "will we light the entire atmosphere on fire" concern, (though I'm not sure how realistic a fear that ever was).
And I think the history of technology is filled with "we need a new paradigm but don't know what that will be, or if we'll ever get there" - aviation is a good example of something that was theorized for millennia with no clear path to how to get there. (An infamous New York Times editorial published in 1903 claimed it would take "one million to ten million years" and even Wilbur Wright at one point predicted it would take 50 years)
>Like the Manhattan project simultaneously pursued three different enrichment technologies because it wasn't sure which (if any) would pan out, and I believe part of why we had three bombs at the time of Hiroshima and Nagasaki (rather than immediately use the first bomb we made) was because there was a real risk that one or more of them wouldn't work.
I agree in part and disagree in part. The Trinity test was necessary because they indeed weren't sure that a Pu-239 implosion bomb would work. After that test, the Fat Man Nagasaki bomb, which was the same type of Pu-239 implosion design, was a pretty sure thing. The Hiroshima, gun-type, U-235 bomb was such a sure thing that it was used without a test. Come to think of it, they must have been remarkably sure of even the _yield_ of that bomb, to be reasonably sure the bomber crew would survive.
Earlier in the project, Fermi's first reactor relied on knowledge of the delayed neutrons to know that the reactor would be controllable, even with just natural uranium. If all fission neutrons were prompt, the doubling time constant for the reactor would have been far too fast for human reaction times to control.
For LLMs - well, we understand the low level processes - the individual perceptrons/attention layers well enough. And we have the scaling laws for _part_ of the high level behavior. But a _lot_ is still not understood, as far as I can tell. AFAIK, we don't know the answer to, e.g. "What do we tweak to cut hallucinations by a factor of 2 without losing any accuracy on GPQA Diamond?"
I think AGI is almost the reverse situation of the Manhattan project. There we knew the physics - i.e. do this and you'll get a superbomb - but no one believed you could separate enough of the right isotope of Uranium. Building the bomb was about finding a method, not about whether the underlying concept was sound. But with AGI we have a method, but no one knows for sure if it will result in AGI, or if such a thing is even possible with the current approach. We have a method, but we don't know that the concept is sound.
Scott, did you see this from Sasha Gusev? “The missing heritability question is now (mostly) answered”
https://theinfinitesimal.substack.com/p/the-missing-heritability-question
I'd be interested if you care to comment on this here or in a future post.
Yes, I will be writing something about that paper.
DeBoer launched some online rage against Gusev's post but aimed it at Erik Hoel who he'd apparently mixed up with Gusev, and then when Hoel politely pointed out the error and asked for a correction, DeBoer's response was to Hoel was,
"Of course, all of the questions in the piece apply to you just as well, and you will studiously refuse to answer them because it's easier to impute racism onto anyone who disagrees with you."
Thought Hoel's reply to that was about perfect:
"All you needed to do to understand my beliefs [about heritability], Freddie, was (a) not get confused and think I wrote something I didn’t, and (b) read the grand total of 2 posts I’ve [ever] written on the subject.
Like I know that these are good internet tactics, to attack and downplay and deflect and move on to more arguments as if nothing matters but…can you be a human being for a second? And not…whatever you’ve become over the years of doing this for so long?"
I've read deBoer regularly some time ago, and he has always had an unhinged edge to him that would regularly come out when he gets agitated. Afaik there also was a full-blown psychotic episode at one point in the past. The ratio between insightful and completely crazy in his recent posting has been getting worse for a while again, imo.
You know, we could always just stop talking about him. Being ignored hurts a lot more than being hated.
I do think there could logically be a 6 ft and below basketball league. There are weight classes in boxing and wrestling and age classes in youth sports for a reason. To make competition between competitors more fair. Nba and NFL are the leagues of freaks!
As I child I was really into two sports which heavily favour height: basketball and fencing. I am not tall and hit a wall with both. If there had been height classes rather than weight classes I might have continued.
This is probably easier for non team sports.
That 3x3 traveling basketball tournament has an under 6 ft category. Plenty of fun to be had as a short basketball player. Getting paid million dollars per year is less likely
Look up videos of Zhang Ziyu playing. She's 7'5. It's like watching someone play against kids.
Sure, there could be a sub-6 basketball league. Would anyone watch it?
My guess is it would get about as much traction as the WNBA
Well, the supply of sports leagues already outstrips the demand. There's any number of sports that people could be watching and aren't, because the people who are interested in sports already have their interest saturated.
That said, "under 6 foot basketball" would be an interesting game which would have a different style of play to height-unrestricted baseball. It would be more akin to something like lighter-weight boxing divisions, where the weight restriction leads to a different style of play, than something like under-15s football where it's just the same thing played worse.
I agree and mention this in the post.
I am somewhat disappointed by Claude Code. I wanted to test its vibe code abilities and I gave him some (locally downloaded, html only) wikipedia pages on various iterations of the same sporting event.
I asked him to write me a python script to scrape me the "List of qualified teams" table and the pre-tournament ranking into a CSV. It struggled quite a bit. I stopped it and trimmed the pages down to the table section and it managed to complete it, even handling some formatting differences across years.
So it was a significant speed up, since I didn't have to manually put the data into an excel-like software but still not a super speed up since I still had to edit HTML.
This decreased my timelines a bit but I am not sure. For once, it was Claude Sonnet 4.5. Also maybe I just suck at prompting? Well, at least I got an internal benchmark out of it. Do you have any similar experiences?
I've gotten better performance when I use it with workflow tools. I would recommend you take a look at this: https://github.com/obra/superpowers
This seems really interesting. I'll try it out
I use Claude Code with Sonnet 4.5 in my day job and I quite like it. Off of the back of a business idea that a friend of mine came up with, I'm planning to transition to being an independent consultant that can radically cut costs due to heavy AI usage. Like a vibecoder, but I actually know how to code, which I'm pretty sure means I can use AI for coding much more effectively than any vibecoder.
He was a game changer in my day job. I can use it to solve some simple, mostly pattern matching PRs (e.g write a migration to change a column or create a table) even on languages I don't fully know
Good luck on your enterprises!
I’ve been trying to do this but struggle to find clients. I need a vibe marketing tool.
Hmm, what have you been trying specifically? I'm kinda thinking listing oneself on freelancer or even fiverr could work for this (I have 0 experience on trying to find work on either platform, but it's what I've heard).
The main thing I’ve gotten from listing on platforms is… other users from outside the US who want to use me as a beard (i.e. they pay me a percentage of earnings in return for me bing the front man). I could do a better job of it, I suppose, but I’ve gotten more than a dozen requests from people wanting me to do that.
How does that work, they talk to the client, but you do the work? It doesn't pay well, I guessing.
Wage arbitrage. As an American I can (theoretically) charge a lot higher rates than they can from overseas. With them doing the work, in theory I could take on more projects too.
To clarify, did you have Claude Code write code that does the scraping, or did you have it try to take the HTML and produce the CSV directly?
Ah! Sorry for this. He had to write a python script that ingested HTML and generated a CSV file. It also had to be sufficiently general to work on ~4 iterations of the event (although they were similarly formatted)
I think the main thing I'd suggest (possible you did this) is making sure it can iterate, run its solution, see if it gets (what looks like) correct output. This for me is the biggest thing with these tools - if you just take it's first attempt its a crapshoot but if you give it a feedback loop it tends to do a lot better at getting to a good solution.
Telling it the actual answer of course defeats the purpose, but if it knows the result should be a list of teams and a numerical ranking, then it should be able to sanity check its own results and iterate that way.
---
Also I'm not sure but, it also might do better if it had actual browser access rather than looking at the raw HTML? I've found for a lot of things the AI does better if it can open a real browser, e.g. via playwright (incidentally playwright is probably what I'd be using for the actual code, too) - it might be able to do queries on the content, rather than reading the entire raw HTML.
To be honest it was still iterating when I stopped and decided to give him the trimmed HTML tables. He was on about the third version of the code. I got scared of wasting all my usage as I might have needed to use today on work
I think you might be right about playwright as well. Originally I just gave him the link, but wikipedia didn't allow the requests (got 403s) so i decided to play nice and download the html only page instead. This might have primed Claude in the wrong direction, and he tried to solve it with only beautifulsoup
I don't code but I have heard many people say 4.5 Opus is much better than 4.5 Sonnet. See https://thezvi.substack.com/p/claude-opus-45-is-the-best-model
I want to try it, but my subscription only covers Sonnet for now. Maybe in a few months when token prices go down again
Is this an API access limitation? I just have the lowest paid tier for Anthropic, and I was able to run my benchmark-ette on the Opus 4.5 chatbot. ( report in this Open Thread at https://www.astralcodexten.com/p/open-thread-410/comment/183118905 )
l think it was a free trial kind of deal, now when I run "/model" it shows me Sonnet, Haiku and " Want Opus 4.5? Run /upgrade to get the Max plan or /extra-usage to pay per use"
Ah, that explains it, Many Thanks!
The recent trans sports post is based on a false premise that cis women would "have no hope of winning." Sports leagues already had rules for hormones, and cis women win much of the time.
E.g. the anti-trans activist swimmer Riley Gaines (cis woman) didn't lose her championship to Lia Thomas (trans woman), like anti-trans activists want you to think. Riley and Lia tied for fifth place behind four other cis women.
The war on trans sports isn't about sports, it's astroturf manufactured by Christian nationalist groups like ADF. Linking my comment on that post here: https://www.astralcodexten.com/p/against-the-omnipresent-advantage/comment/183003939. For accessible primers on ADF and on trans athletes, John Oliver has some good segments: https://www.youtube.com/watch?v=sCAuHH5EYnE (ADF), https://www.youtube.com/watch?v=flSS1tjoxf0 (trans athletes). You can also find out more from civil rights groups like SPLC.
Do you believe that women's only sports leagues should be allowed? If yes, why should we bend the rules so that transgender males identifying as women are allowed to compete in them?
>The war on trans sports isn't about sports, it's astroturf manufactured by Christian nationalist groups like ADF.
If the arguments make sense, it doesn't matter who makes them.
The arguments don't make sense.
(As I mention in the longer comment, sometimes they're outright fabrications. The ADF president had her young daughter testify—without disclosing that she was the ADF president's daughter—that her softball team's heart broke when they realized they were playing against a trans girl full of male biological advantages. Trouble is there was no trans girl: it was a cis girl with short hair. And, of course, there's the widespread and deliberately encouraged false impression that Lia stole Riley's swimming championship.)
If you think the arguments don't make sense, I question why you even think it's relevant that the organisations making them are even affiliated with the ADF. It just seems like a completely irrelevant point.
You don't care that sophisticated propaganda groups are working overtime to mislead and manipulate you and your society, with evil intent? You want to avoid finding out when that's where a particular piece of disinfo came from? You want to avoid practice recognizing their tactics and falsehoods in future?
I can promise you sophisticated propaganda groups are not the reason the vast majority of Americans are anti-trans people in sports.
I'm sorry – were you talking about trans activists, or the people opposed to them?
From where I'm standing, the people being economical with the truth when it comes to sporting events are nowhere near as grave a threat to our society as the people campaigning to house convicted male rapists in women's prisons or carry out mastectomies on healthy teenagers below the age of majority.
I've spoken to a trans man that got a mastectomy as a teen. He competed on his high school's boys' swim team, and his teammates were under the impression he had gynecomastia, which is fairly common. He decided to get surgery early, same procedure he would've gotten if he were actually a cis boy with gynecomastia and for the same reason too (gynecomastia is not typically medically dangerous, and treated for self-image/discomfort reasons). For some reason, this is way more controversial.
As I suspected: you're just mad that I exposed ADF because you're on their side and wanted their lawfare operation to remain sneaky.
I bet you also want to ban puberty blockers, even though half the point of PBs for trans boys is so they *won't* need a mastectomy.
When I look this up, I see https://www.outsports.com/2024/12/6/22948400/transgender-trans-athlete-championship-national-world-title/ . Are their examples false?
There seem to be two opposing forces: first, AMABs are X amount better at sports than AFABs; second, trans people are extremely rare, such that in the set of (all cis women + trans men), it's possible or even likely that the top athlete is a cis woman, even if the cis male average is much higher. I think this means that discussion of a certain level of trans victory in sports isn't interesting without knowing what percent of the athletes are trans, why that is (ie whether existing norms are keeping other trans people out), and how that's likely to change over time (as norms and the % trans changes).
I'm not sure how I feel about a world where AMABs are some large amount better than AFABs, but AFABs still usually win because trans athletes are very rare. I guess it matters less? Still seems like however unfair it would be if there were a very large advantage, it's some fraction of that unfair, as opposed to zero.
One thing I'd add to this is that the top athlete isn't picked randomly from all eligible. athletes. Some people are more athletic than others, you know it ahead of time.
Most *specific individuals* who are trans women athletes pose no threat to be the top athlete.
One of them could ask - "forget the abstract, why should *I*, specific trans girl HS senior with 0 athleticism, not be allowed to ride the bench on the girls' JV team rather than the boys' JV team? I'm not trying to cheat, I just want to be on a team with other girls, have camraderie and shit. I'm recognized as a girl in all other contexts, but now you're gonna single me out based on some hypothetical when actual me can hardly avoid bouncing the ball off my foot".
29 trans women out of how many women total who have won titles in the reference class surveyed?
Skimming the list, it looks like the reference class includes lots of state and regional competitions, national competitions from multiple countries (including one in Belgium), several narrowly age-bracketed competitions, several competitive levels (high school competitions, "Division II" NCAA tournaments, etc), a very wide range of sports (track, golf, roller derby, billiards, BMX biking, etc), several events per sport in many cases (e.g. 500 meter dash within the umbrella "track" category), and at least 17 years of data (examples range from 2008 to 2025) although most of them events are from the past 10 years.
A quick fermi estimate: 70 scopes of competition (50 states + 10-ish regions + national competitions in 10-ish countries) * 10 relevant combos of age bracket and skill bracket * 20 sport/activity categories at least as relevant as billiards and BMX * 5 events per sport * 10 years of data = 700,000 potential titles.
At least a couple of the trans women on that list won multiple titles, so the numerator should probably be more like 40 or 50 instead of 29 to match the units of the denominator. That works out to about 0.007% of women's titles being won by trans women. Which is somewhat less than I'd expect if the standards for trans women participating in women's sports were perfectly calibrated, since about 0.8% of the general population is trans.
I'm actually surprised by how low my estimated percentage of "titles won by trans women" is. I had been expecting around 1-2%, based on a prior that the "Testosterone levels must be below X ng/dL for Y period of time before the event" standards that most events seem to use would screen out most of the advantage from having had male hormone levels in the past but not all of it, depending on how well-calibrated the values were for X and Y.
This seems like a helpful estimate.
One important point here is that professional sport competitions don't need to be regulated in the same way as casual sports. Many pro-trans-sports activists don't care about the Olympics so much as they care about a trans teen being allowed to play soccer with her friends at school.
Keep in mind that in many places, high school sports aren't "causal" in any sense of the word. People take them extremely seriously, and parents spend thousands on professional training. I can see many of them raising hell about a trans girl competing, especially if she does well.
Really? I would have thought the reverse. In the pro-sports context, there's financial rewards incentivizing cheating, as well as a business need for at least the appearance of fair competition. Hence all the rules about doping, what weight you can and can't be in certain classes, how often you can practice and what you can and can't do to scout competitors, and so on.
In a casual school-sports context there isn't really anything at stake beyond fun, camaraderie, and the life-lessons learned from competition. Those don't just disappear if the middle-school swim team has one trans kid on it, and without the perverse incentives that accompany financial benefits to the winners, your population of trans soccer players is going to be confined to the fraction of the 1% of trans kids that happen to be interested in soccer, so it seems like it would be hard for that to impact at a level that would break the system.
I think you misread my comment. I'm saying that people in *favor* of trans girls/women in sports often care more about the casual sports than the professional ones, because those impact more trans people. Yet people trying to ban trans women often focus their rhetoric on the Olympics, while targeting kids in casual k12 sports.
Yep. Misread on my part - thank you for clarifying!
haha same at first read
Haha at first I misread you as saying the opposite of what you said. But anyway...
In theory, in another historical context, I could sort of agree. Yes, I care more about kids's basic right to play sports with their friends than about the Olympics, even if I care about that too, and in a golden age of LGBT rights in general the latter might be less urgent.
But in the current context, neither is about sports; both are just precedent-setting lawfare attempts by Christian nationalist groups like ADF, designed to get us used to excluding trans people from society (itself part of a divide-and-conquer strategy to eventually exclude all LGBT people from society), so we can't give these fascists an inch. Even if the war on trans kids is even more stomach churning than the war on trans adults, both are part of the same overall strategy.
One of the main points of Scott Alexander's advocacy is for us to look at the substance of people's arguments instead of attacking each other on partisan grounds.
I agree with this mission and I am disappointed to see you fighting against it.
Unclear what part of this you see as against that mission.
I've made clear that I don't want trans women kicked out of women's sports, on the merits, including at a professional level. In a utopia of LGBT rights I'd still think that, but prioritize it less. In the real world, where *in fact* Christian nationalists are deliberately weaponizing sports—this isn't an ad hominem, this is recognition of their actually-existing political strategy—it is a priority.
(slightly) strained analogy time. A paramilitary group's code phrase, signaling to fellow travellers that they're gathering recruits and getting ready to blow up Metropolis, is "The cock crows at midnight."
We start hearing "The cock crows at midnight" said more than usual. I give everyone a heads up that that's a paramilitary group's code phrase.
I get tsk tsked for my ad hominem, instead I should have focused exclusively on whether or not it's true that roosters crow at midnight
It's also a Toxoplasma of Rage issue, and one which tends to be framed in a way that implies a binary choice between a "No trans women ever in any women's sports" and "Any AMAB willing to check the 'F' box on the entry form can enter any women's competition at any level of play".
The actual status quo of many/most high-level competitions having rules of the form "You must have bloodwork showing testosterone below X ng/dL for the past Y period of time" tends to get ignored, partly because a lot of people don't know about it, and partly because arguing that X or Y should be a higher or lower value is boring, technical, and doesn't work well as a proxy fight over the legitimacy of transgendered people.
Yup. Those who actually cared about sports were fiddling with decimal points, not sharpening pitchforks.
>"You must have bloodwork showing testosterone below X ng/dL for the past Y period of time"
A non-starter because
1. Even in instances in which this is the official policy, it's often done on the basis of the honour system, and hence useless at keeping out bad actors (https://quillette.com/2022/09/28/is-this-the-lia-thomas-of-disc-golf/).
2. Reducing one's testosterone levels has minimal impact, if any at all, on one's bone density or lung capacity, which are effectively baked in during puberty.
1. "Often" is doing the heavy lifting here. If the honor system isn't working, they can change the procedures to require documentation, like many top-level professional men's sports do for steroids.
2. I am very skeptical of both the accuracy and the relevance of this. Bone density definitely changes after puberty in response to hormones, which is why osteoporosis is a problem for postmenopausal women. Lung capacity (in terms of the raw volume of the rib cage) may be baked in, but it's one of several things that affect vo2max. Things I know off the top of my head that are hormone dependent are training capacity and hemoglobin levels.
This study (tiny sample size, but it's the only one I could find) actually found trans women athletes had lower lung capacity (as measured both by vo2max and exhalation volume) than cis women.
https://pmc.ncbi.nlm.nih.gov/articles/PMC11137468/
I don't see how anything that happens in women's sports affects an ideology that last came to power in interwar continental Europe.
I don't agree with this strategy of approaching anything. It's bad behavior on principle, and is an ineffective way of getting what you want in practice.
Yes, there are probably people within the ADF who are consciously acting on the strategy you describe of trying to normalize excluding LGBT people. But most people who oppose trans women in women's sports don't endorse this. We should try to judge issues on their own merits, not on the most repulsive views of their most repulsive proponents.
Edit: from your other comments it sounds like you're in favor of some restrictions on trans women athletes, so I'm confused by what you define as "can't give these fascists an inch."
i might answer better later but:
1. It's not "people within ADF," it's ADF, as an organization. This isn't a church or social club, it's a strategic litigation group whose reason for being is to strategically litigate and turn the US into a Christian Nation.
(Strategic litigation is a tool that can be used for good or for bad. NAACP used it to end racist school segregation; Lambda Legal and ACLU used it to win gay marriage; it was used both to gain and then later to overturn abortion rights. ADF also defend discrimination against gay marriage etc.)
2. The organized anti-trans-sports activist movement isn't a grassroots movement led by well-meaning normies who are fine with trans people but don't really get it. The organized anti-trans-sports activist movement *is*, in its essence, a lawfare movement. It *is* Charlie Kirk and Riley Gaines lying to normies that Lia Thomas stole Riley Gaines' championship, to trick the normies into getting used to eroding trans rights. It's not a radical fringe, it's the radical vanguard party leading the whole operation.
The most important thing to do in this context is make sure people know what's actually going on: that the leaders aren't good faith, they don't care about fairness, this isn't about sports, this *is* just a new attack on trans people after the bathroom-bans attack failed.
> It *is* Charlie Kirk
Nothing "is" Charlie Kirk anymore, as he was assassinated, apparently over trans issues. At most, something possibly "was" Charlie Kirk.
Men, and post-puberty boys, are not allowed to play in sports leagues set aside for women. I have never heard any objections to this. It allows women to play high school, college, and professional sports. Do you take issue with making these opportunities available to girls and women? If not, then it follows that someone who has the biological advantages of male puberty that led to separate leagues in the first place should not be allowed to compete in women's sports. Incidentally, maleness is a much greater advantage than height in basketball. There are lots of tall women, but none of them could make a pro team. If 6' men were allowed to play in the WNBA they would dominate.
To be clear, many sports leagues *did* allow trans women to compete if they met certain hormone requirements set to keep things fun and interesting and sporting. This was fine and cis women were still winning plenty.
If we agree that it's fine if certain sports leagues choose to allow transsexuals, will you also agree that it's fine if certain sports leagues choose to disallow transsexuals?
If so I think everyone in the thread is going to be in agreement. And it breaks apart one big issue into thousands of small issues (should the Moose Jaw Regional Women's Lacrosse Association allow MtF transsexuals?) on which I don't take any particular issue.
My understanding is that yes, leagues choosing their own rules was the recent status quo. Trans bans are a recent development; I don't think there were any laws requiring trans inclusive sports leagues. I prefer the status quo to bans.
(I would want federally funded stuff to be trans inclusive in principle, but still with hormone level rules etc to be allowed as is appropriate to the sport.)
Meeting certain hormone requirements has minimal impact on one's bone density, lung capacity and other traits which are essentially baked in among males who've gone through puberty.
Perhaps. And yet cis women were still winning plenty.
(Re: bone density, I might—this is speculative, I haven't looked into it—make an exception for something like boxing, where it might be about safety not just about who wins. Though tbh boxing is inherently always dangerous and it's kind of odd that we allow it even if fights are exciting. An illegal punch put Prichard Colón in a coma despite his bone density.)
I suppose we need to try and determine which of these three worlds we are in:
A) Trans women have an overwhelming advantage, such that they win many times more often than cis women (although given the comparative numbers, cis women probably still win most events)
B) Trans women have some small advantage, but that is comparable to other minor genetic factors (and so will win more often than cis women at lower events, but the effect is largely undetectable at the professional level where every major contender has genetic factors)
C) Trans women have no significant advantage over cis women, and you'd expect them to win with the same frequency as cis women (possibly slightly higher if there are selection effects meaning only the absolute best trans women bother to compete given the hostile environment, meaning the average competing trans woman is better than the average competing cis woman despite having identical population-level skill)
This is further confounded by massively-shifting goalposts around what is a trans woman in sports and what should be the metric for allowing trans women (if any). It's reasonable to expect different outcomes if the criteria are basically "anyone who says they are a woman" vs "anyone who has openly and persistently lived as a woman for X time" vs "anyone who identifies as a woman and has been on hormone therapy for X time, such that it meets Y condition for trying to objectively assess advantage". Furthermore, several people on the other post and this open thread have conflated intersex people with trans women, or even CAIS people where they have an XY chromosome but due to a total inability to process testosterone appear physically female to all but the most thorough tests (and likely didn't even know they were anything but a normal XX-chromosomed cis woman). These are all different cases that should probably be treated differently.
Personally, I favor the "has to be on hormone therapy and meet some objective measure" criteria which I suspect will result in measured outcomes in World B (or even World C for some sports where the male advantage is smaller to begin with). I am fine with living in World B, and think there is definitely *some* non-zero degree of advantage trans women can have that still rounds off to fair and creates a reasonable world where everyone can have fun with sports.
My understanding is that "has to be on hormone therapy and meet some objective measure" was the status quo for many/most high-level athletic competitions before the current push for a blanket ban on trans women in women's sports commenced. The objective measure was generally of the form "Must have blood tests showing testosterone below X ng/dL for at least Y period of time".
CAIS is included in most definitions of intersex, it's just a different condition from more prominent ones like 5αR2D, which is what Caster Semenya has.
>"Are their examples false?"
In the present context, this seems like an inversion. Instead of "trans people shouldn't be allowed because cis women would have no chance of winning," should it now be "trans people shouldn't be allowed unless trans people have no chance of winning"?
In its own context, this is irrelevant to my point. It's showing that there have been 29 trans champions, not less-than-3 like some guy on instagram claimed sometime in 2022. I didn't say there have been less than 3 trans champions, I said cis women often win.
I'll point out that this article's subheading agrees "most trans athletes aren’t winning major titles."
otherwise:
>"I'm not sure how I feel about a world where AMABs are some large amount better than AFABs, but AFABs still usually win because trans athletes are very rare. I guess it matters less? Still seems like however unfair it would be if there were a very large advantage, it's some fraction of that unfair, as opposed to zero."
That is the world we live in, if advantages are large. Trans people are rare and will remain pretty rare even with increasing social acceptance. Otherwise my view is:
That conversation was already had by sports leagues, and they already got to a generally fine resolution. Any tweaks were and should have remained marginal and technocratic among sports leagues committees, not the subject of ADF impact litigation, or Charlie Kirk marching Riley Gaines out on TPUSA Faith and telling everyone that Lia Thomas is an abomination who stole Riley's championship, or misrepresentations that trans women make it impossible for cis women to win.
I’m curious as to if anyone else has a “bristle” response to anything, sort of like the opposite of a phobia. You get braver than is reasonable when you encounter a certain kind of stimulus.
I cannot even in my imagination be afraid of anything with feathers.
If literally anyone threatens my life (for some reason just violence doesn’t count and I can be rational) my first thought is “try it and see what happens.”
> I cannot even in my imagination be afraid of anything with feathers.
Ratites can be quite large and have razor-sharp talons. The largest to ever exist are extinct though.
I'm with "Some Guy" on this one, I know intellectually that a Cassowary could fuck my shit up but when I look at pictures of them, all I can think is "lmao I'd punch that stupid face so hard"
When a group of friend suggests a dangerous trip/trail/whatever, I get brave real quick and I can follow up on it. There's just something about doing dumb shit with your friends.
I think this is almost correct but if a phobia is a pool of stimulus that is “always flight” I don’t know what the word is for “always fight.”
Many have written about how abortion prohibition is dysgenic: abortion patients are disproportionately low-income, uneducated, and unmarried, conditions that correlate with low intelligence. Most just leave it at that: it’s not like the pro-life side has any counterargument. But there’s a more fundamental reason why abortion prohibition is dysgenic, rooted in the same principles conservatives intuitively understand when it comes to most of life, the principle that voluntary exchange produces better outcomes than coercion.
Businesses that must convince customers to buy their products work better than those that can sit back and collect government subsidies. Employees who fear being fired work harder than those who know that laws make it difficult to fire them. Renters who know landlords can easily evict them behave better than those who know eviction is difficult. Landlords not subject to rent control have an incentive to maintain their properties and thus compete with other landlords on quantity. Many public schools remained closed during the COVID lockdown; private schools, which relied on parents voluntarily forking over tuition, reopened as soon as they were legally permitted to.*
The same logic that applies to business applies to family life. Conservatives understand this, they’ll tell you how the welfare state damaged the American family. In a world without welfare, women know they must choose a mate who will be responsible and willingly support their children, which means putting in the effort themselves to attract such a mate. Likewise, men know they must signal their own responsibility; women want a husband, not a sperm donor.
The corollary to this is that when you have a business nobody will shop at, a worker nobody will hire, a tenant nobody will rent to, a single person nobody will marry, you know that there’s something wrong with the business/worker/tenant/single person.
Now apply this same logic to abortion. The mother does not want to voluntarily support the fetus, which tells you there’s something wrong with the fetus. Maybe it has Down syndrome, maybe it was conceived as a result of rape. More likely, the woman doesn’t want to have children or doesn’t want any more children. In these cases, the man who conceived the child almost always agrees. You might be tempted to say this is a problem with the parents, not the fetus, but the fetus got its genes from the parents. The genes that made the parents not want to support a child or another child are likely present in the fetus. Forcing the child to be born is selecting for low-investment parenting. It is dysgenic.
Eugenics should be about more than just IQ. Proponents should also desire a world where more people have a loving and responsible attitude toward their families. This isn’t something that can be measured in a two-hour standardized test, but it’s vitally important and eugenicists should reject policies likely to undermine it, particularly when, like abortion bans, they’re also offensive to human liberty.
It is at this point where pro-lifers usually accuse me of misunderstanding or strawmanning them. Their argument is about morality, they say. Baby murder! In effect, they say the eugenic implications are unimportant to them, so therefore they must be unimportant to me. That is not how political discussion works. And I think that even adopting their moral framework, their position is still unjustified. The fact that you’re a human being does not entitle you to support at the expense of other unwilling humans. Conservatives understand this when it comes to the homeless man’s “human right” to be fed and housed at government expense. I’m not opposed to using tax money to support the mentally ill and disabled, but even there, I wouldn’t say the latter have an absolute right to receive the money. If 90% of society were disabled, there shouldn’t be an obligation on the remaining 10% to work around the clock and live in hovels to keep as many of the former alive as possible.
Conservatives have ideas that are both good and appealing: the superiority of free enterprise, the value of liberty, the importance of responsibility and family values. But then they shoot themselves in the foot by embracing abortion banning, alienating potential allies (particularly among upper-class people) and undermining the ideological force of their ideas. It’s hard to argue for limited government when you think government should be involved in people’s most intimate decisions. It’s hard to argue against a culture of freeloading if you think one class of “persons” has an absolute right to freeload off of another. Pro-lifers will tell you that abortion bans and right-wing politics are somehow intrinsically linked, but a simple look at other countries shows that’s nonsense. In most Rights in most countries, abortion is a non-issue.
We should make it one in America, too.
* The negative effects of coercion don’t apply when the parties voluntarily agreed to a contract, as in a contract between two businesses, or a marriage contract that does not allow the parties to desert one another without compensation.
You make a lot of good points, but you make them in such an outrageously provocative, edgelord manner that you are literally hard to read.
It's like a YouTube series that is interesting and informative, but every time you watch one of the videos, somebody stomps on a puppy somewhere.
Chance, I don't always agree with you - but in this instance you are completely correct.
There's sense in there somewhere, but it's very hard to access due to the slathering of nonsense over the top of it.
I read a decent amount of Yudkowsky's Sequences not so long ago, and I did not realize to what extent Yudkowsky DESPISES deliberately inflammatory, provocative language. He really HATES it. And over and over, he writes that people who do this are "unenlightened" and, by implication, worthy of contempt.
Did not realize he felt that strongly about it! I always assumed that carnival barking was a perfectly acceptable part of online discourse. I've been choosing my words more carefully lately, since I know this forum is full of Sequence readers. (although predictably, it seems that most of them do not like to be explicitly attached to the label "rationalist")
Moderating one's language seems like a very sensible approach given the local customs. Especially if someone wants others to take their ideas seriously.
Signed: a moderately spoken, rationalist-adjacent individual!
> The mother does not want to voluntarily support the fetus, which tells you there’s something wrong with the fetus. Maybe it has Down syndrome, maybe it was conceived as a result of rape. More likely, the woman doesn’t want to have children or doesn’t want any more children.
Most fetuses with Down syndrome are aborted, but most aborted fetuses don't have that or anything specific to them, rather than the mother. Bryan Caplan recently blogged on the empirics of a "natural experiment" in denying some women an abortion while permitting it to others https://www.betonit.ai/p/the-tragic-hysteria-of-abortion
Probability he does an Unz arc is now 10%.
I haven't kept up with Unz recently, so what arc?
Unz became a Holocaust denier.
I would say there is significantly less than 10% chance that Caplan does that.
I think it's too unnatural to refer to fetuses as freeloaders, and I view conservatism as being quite strongly about believing the naturalistic fallacy is mostly correct (which if true, wouldn't invalidate the naturalistic fallacy, as that's about nature being always good). I'm not sure if the European right are good representatives of the right. Conservatives are also about tradition, I would also argue both tradition and family values would lead one to be anti-abortion.
The naturalistic fallacy would lead one to be pro-choice, in the state of nature animals can freely decide not to provide food to their young, and often do exactly that.
No, it's not natural to humans that grew up in a Western context to think that way. Hmm, maybe Western is too broad, it's not natural to conservatives in America to see things like that, that's not their cultural conditioning, their tradition.
For east coast megameetup! People should (a) sign up to run activities and (b) sign up for the ones I'm co-running (writing workshop on Saturday and Roman empire wargaming on Sunday). Also if you're interested in either taking a childcare shift or depositing your children at one let me know.
For West Coast megameetup, what's the status of activities? Would people be interested/have room for a clone of one of my two east Coast sessions (especially the imperial wargaming)?
Why is the West Coast megameetup happening so early in the month? I'm a college student, and it will come right as I begin my final exams, so I definitely won't be able to attend (but would be able to go to a Berkeley meetup at the same time as the East Coast one).
They deliberately do them on different weekends, and I think the bay wanted to avoid collision with Christmas (once again, all our problems can be blamed on Ray Arnold). TBH it would've been much easier for me too to have had them the other way around.
If you suffered a concrete, economic injury from the credit report going down you may recover it from this suit.
Otherwise, and I can tell you aren't going to like this, you're out of luck. You're trying to step up and say, "This is the law, they aren't following the law and the regulators aren't enforcing it so I've filed this suit to make them follow the law" I'm sorry but that won't work.
If you get mugged, you can sue your mugger for the money they took out of your wallet, you could sue them for physical injuries you suffered, you could potentially sue them for emotional damages, what you cannot do is sue them to make them follow the law from now on. This is the case even if the police are completely failing at their jobs and letting them off with nothing more than a warning.
In terms of personal risk, (aka sanctions and being labeled a vexatious litigant) they're actually very low. Pro se litigants receive a great deal of leeway and judges are often sympathetic. The trick is to be humble and listen to the judge, do that and you're very unlikely to suffer any consequences.
This all sounds like it would wind up costing a lot more than the original ten grand.
If I were you I'd suck it up, pay the ten grand, and resolve to be more careful in the future.
> I’m the one who brought the case, after the $10k judgment was wiped off the books – not the guy hiding from a valid bill.
How'd you wind up dealing with a debt collector in the first place then?
> Yes, I owed a debt. That’s not the part I’m running from. I dealt with it the way people are supposed to:
> I negotiated a payoff with the debt collector’s attorney outside the courtroom as I was instructed to by the judge.
No, you're supposed to pay your debts long before collectors or judges get involved.
Your best option is to get a lawyer.
You say you were a marine veteran; I'm going to assume you made it past E3, or else most of this will be lost on you. As an E5 or E6, did you ever have that mouthy Lance corporal show up who was positive he knew everything better than you did? The one who would seemingly misinterpret every single order, or quote some arcane and incomprehensible part of the ucmj or whatever to demonstrate why he was always right?
That's you right now. There is a whole host of formal knowledge that you lack and informal process understanding you're ignorant of that I guarantee is pissing off everyone you interact with. That will make everyone to include opposing council and the judge less likely to give you a break.
You aren't going to be able to accomplish anything pro se. Get a lawyer, humble yourself to listen to their advice, and put this behind you. This will get increasingly worse for you the longer you try to contest it yourself.
Having read through this entire thread so far, it sounds like (1) you got screwed over by the debt collector's lawyer (2) you got screwed over because you didn't have a lawyer representing you who knew the kinds of tricks they play. You represented yourself, I don't know if that's because you couldn't afford a lawyer, but this is what happens when you represent yourself.
So yeah, "get a lawyer" is good advice here, otherwise you're just throwing good money after bad chasing this.
I'm not a lawyer (and certainly not *your* lawyer), but I have a little experience the debt collection space. If I'm understanding things correctly, the $10,000 judgement has been vacated and they are no longer pursuing you for collection? If so, that is probably the best that you can *realistically* hope for.
If they are still trying to collect, disregard the above.
Either way, I agree that not being able to find a lawyer is a very, VERY strong signal that pursuing the case further is not worthwhile, and likely to be one or both of extremely time-consuming or unlikely to succeed. Of course, if they are still trying to collect and generally making your life miserable, you might not have a choice but to keep fighting them. If you want to keep trying to find a lawyer, talk to legal aid groups. They likely won't be able to help you, but at least they might have better referrals.
Fair enough. Several additional thoughts:
- Have they offered to settle yet? If they have that's a very positive sign (and also it is probably economically correct for you to take the settlement).
-- Settlement will likely only be possible if you drop the demand for a ruling/agreement about future use of "Agreed" Judgements. I also frankly still do not see any path where a pro se litigant successful gets such a broad ruling in their favor.
- The longer you continue to be successful in your case, the more interest lawyers will take in it. Surviving summary judgement (I don't have the stats at hand but I've seen them and a large, large percentage of Pro Se cases are dismissed via summary judgment) in particular is a very strong signal.
- I don't know about the legal aid infrastructure in TN in particular, and maybe what you say is true. In other states (including my state), legal aid groups are overwhelmed, but many do maintain robust consumer rights units, especially as consumer rights cases are (comparatively) quite fast and easy to litigate ("can you show chain of title to Jane" "no" "then why are you suing Jane" "I would like to postpone the case indefinitely"). There are also some national consumer rights groups that might be able to help (though I don't know if they have Tennessee specific resources).
- It sounds like a large part of your problem finding lawyers is that the network of local lawyers is small, and they all know each other. Have you tried asking lawyers in *other* parts of the state if they are interested (because they wouldn't have the local reputational costs)?
If you can't get any lawyer to be interested in taking your case, it suggests the case is not worth their time. Are you sure its worth your time?