Does anyone have a good guide on OPSEC for running a pseudonymous online profile? And what do you think about Anatoly Karlin’s contention that AI will make the whole exercise futile?
DanielHobson
9mo ago
You can notice almost nothing else without noticing race. It has no cost of acquisition.
I also disagree that race is low benefit; with gender, it probably carries the most information density of any heuristic. Race+gender would cause dramatic updates in estimates of height, weight, temperament, intelligence, etc. etc. You’d be hard pressed to find another single piece of information that informed as many predictions outside of the sequence of one’s DNA.
If I were to rank what I think are the most valuable pieces of information to make accurate predictions about an individual, they would be, in order:
- Race
- Gender
- Age
- Socioeconomic Status
- Place of Residence
Again, I agree with the general premise that the salience of race is probably too high with Mottizens. We incorporate so many different data points about individuals that a Bayesian should lower the weight placed on race, but your notion that it isn’t the single, or even one of the most valuable data points seems detached from empirical reality.
I guess this all depends on how you weight different predictions you’d want to make, but it’s hard to imagine what blend doesn’t end up with race near the overall most predictive single factors.
I am sympathetic to lowering the salience of race, but totally ignoring it just willfully deprives one of relevant information. One should approach things as a good Bayesian and updating their priors as new information becomes available.
Let’s consider your example of walking down the street. Yes, I would try to avoid that person regardless of their race, but that’s because they are over-the-top signaling that they’re trouble. Not all trouble is the same, though.
Rather than the binary “should I or should I not avoid this person,” consider trying to assign a probability that they are going to rob you. Regardless of race, this percentage exceeds the threshold necessary to justify avoiding them, but that doesn’t mean the probability is equal. Unless some deity tells you exactly that number, there is signal in race.
The only cases in which race, or any other available information, should not be included in your Bayesian assessment of probability is a hypothetical case of absolute certainty, which just doesn’t exist.
So, yes, be a diligent Bayesian who actually updates priors based on other available information, but there is signal in race, as there is signal in everything.
DanielHobson
9mo ago
Yeah, I’ve read Scott’s two pieces from back in the day. I thought the piece you linked to was great, while the anti-NRx FAQ wasn’t his best work and did a disservice by not attacking his steelmanned version. Regardless, these pieces are nearly a decade old. I’m wondering where all the neocameralists went (Prospera?) and how the movement developed since then, considering it’s now basically disappeared as it’s own entity.
DanielHobson
9mo ago
Does anyone know of a good history of NRx, discussing the emergence of different strains of thought and where they’ve settled?
I suppose much of the culturally right-wing, nationalist branches were integrated into the alt-right and national conservatism movement, but what of neocameralism and the libertarian ethos of competitive governance? Peter Thiel, Patri Friedman, charter cities, etc.? Yarvin simply writes too many words for me to bother reading outside of certain exceptional pieces.
DanielHobson
9mo ago
I want to write up a thread on education that’s only tangentially related to the culture war titled: What Should I Look for in my Child’s School?
While it’s inspired by recent developments in public schooling, it’s much more than that, looking at course offerings, child wellbeing, social networks, values, etc. Essentially a rationalist dive into what schools produce the best overall outcomes. (A lot would have to come from the contributions of others.) I just don’t really know where to post. I’m not going to post it in the CW thread on a Saturday where it will get buried, don’t have a substack to link to, and so am considering a top level post and hoping it follows the rules.
DanielHobson
9mo ago
Of course it’s a history exercise; building a credible case for a counterfactual relies on deep knowledge of pre-existing context and trends. You cannot explain how Rome would differ without understanding how Rome was and how Rome did change.
A counterfactual without utilizing the factual would be shoddy work.
When I claimed that there had been no substantial disputes, I was referring to the fact that every criticism I had seen to that point had presented alternative mechanisms that could be at play without testing any of them or seeing if they were addressed in the paper. There’s a big difference between hypothesizing various different factors that could affect the data as opposed to seeing if one of them is actually a valid alternative cause.
Today, a rebuttal that met those criteria was posted. I can’t speak for how damning it is of the Clark paper, but wanted to acknowledge its existence and share.
Update: Cremieux has replied to this rebuttal and found it wanting.
DanielHobson
10mo ago
This is the first change that is actually going to majorly impact my Twitter use. Only being able to load 600 posts a day is about 5 minutes of activity on Twitter. There’s so many replies that get loaded every time you click on a tweet, and even refreshing your timeline probably results in loading several hundred. Even the paying users will be severely throttled. Between this and Apollo for Reddit shutting down today, I might have a lot more free time!
I’d be interested in learning or reading more about this. My impression is that it really was just: “ignore all evidence, science doesn't apply here and you just have to accept what the societal elite say the science is" until Arthur Jensen and William Shockley attempted to remind everyone that hereditarianism is true, and that the blank slate consensus was fabricated.
The post war period had things like the 1950 UNESCO Statement on Race, but that really wasn’t evidenced based. It is a pure moral document (that many scientists refused to sign because it lacked evidence).
Only in the late 70s and 80s did figures like Lewontin and Gould emerge to try to claim that it was evidence-backed. This could be a blind spot for me, but that’s my understanding of the timeline.
DanielHobson
10mo ago
No need to worry about me doubting your sincerity, this came off as a good faith question.
I would say “apply,” with regards to observed phenomena by considering genetics and biology as at least a perfectly reasonable cause exploration, and probably as the default causal explanation.
To walk through a real world example, we observe that in America, whites are 5’10 on average, blacks are 5’9, and Asians and Hispanics are 5’7. Say we wanted to explain why this is; our first area of investigation should be whether this is just a genetic quirk. If we have reason to believe it’s not, like finding different degrees of heritability across groups, then we might look to cultural or economic factors, like malnutrition rates, or a cultural practice of amputation at the knees (obviously this is exaggerated, but for height, I couldn’t think of another cultural mechanism).
Height is pretty low stakes, of course. Still, HBD sees this as the most effective way to model and understand the world around them, and the topic being controversial, like crime, intelligence, or social status, has no bearing on the use case of the model.
I would define it as:
The belief that individuals and groups meaningfully differ in nearly all measured attributes for biological and genetic reasons.
That’s really the extent of the belief. The point of contention is what is off limits to scrutiny through that lens. No one would object to the idea that skin color, eye color, or hair texture differ between individuals and populations for genetic reasons.
HBDers (alternately “race realists,” “hereditarians,” “scientific racists,” etc.) are those willing to apply that lens to all human traits.
Is this really true in the case of HBD? Almost all of the new evidence is validating what were previously commonplace beliefs.
This paper is just strong evidence in support of Ronald Fisher’s additive genetic model. Plenty of hereditarian work is just validating Galton.
The intellectual arguments were widely accepted, and then rejected following World War 2, despite the evidence still supporting them. I wouldn’t be so confident that theories rejected while having the weight of evidence will be adopted because the evidence, which was already pretty clear, becomes even more overwhelming.
DanielHobson
10mo ago
Gregory Clark published The Inheritance of Social Status: England, 1600-2022. You can find breakdowns of the results and methodology by geneticist Alexander Young and Cremieux in Twitter threads. The main takeaway is that a model of genetic inheritance and assortative mating nearly perfectly explains social status across nine different measures.
This builds on previous findings that dramatic changes in social structure or wealth transfers are often only temporary setbacks for elite families. In China, the Cultural Revolution, perhaps the single biggest upheaval in social structure and wealth redistribution in human history, saw the pre-communist elite families spend one generation below median income/education before outearning and outlearning other households by 16% and 11%, respectively, in the second generation. A similar phenomenon is seen in the American South following the Civil War, where it took antebellum elite families one generation to regain equal footing, with the second generation surpassing their counterparts in income and education.
Critics of the hereditarian hypothesis have posted critiques of the study, but, to my knowledge, no clear alternative hypotheses or explanations for the genetic model fitting basically perfectly.
DanielHobson
10mo ago
This ended up being correct. For some reason my adblocker removes the upvote button and literally nothing else. Upvote for you!
DanielHobson
10mo ago
How do I use this site’s upvote feature? I know it has one, and I can see it when logged out, but don’t have the button logged in. I’ve just been reporting exceptionally good comments as quality contributions, but there has to be a way to access the in-between step.
DanielHobson
10mo ago
This is both false and uncharitable. The linked article references:
-
The widespread admonishment of Nick Bostrom among EAs after his comment on factual group differences was leaked
-
TIME article from disillusioned women in EA making questionable claims of sexual assault (to which the CEO of Open Phil replied, not the organization itself, as you suggest)
-
Open Phil making donations towards criminal justice causes without any evidential basis for their effectiveness
-
A highly upvoted post on the EA forum titled “I’m a 22-year-old woman involved in Effective Altruism. I’m sad, disappointed, and scared.” This post then goes on to critique EA for placing too much emphasis on rationality and not enough on emotion.
-
Highlights two cause areas (global dysgenic trends and the power laws of crime) that are ignored by EA as taboo.
There are more instances outside the article’s scope that I could list, but figured that captured the main body of issues, and further examples would just be further evidence of these specific trends.
DanielHobson
10mo ago
There seems to be a distinction here between individual women and feminine structures. Elizabeth I, Catherine the Great, Marie Curie, Rosalind Franklin, etc were all women, but they operated in structures organized along masculine norms. As such, I don’t think it’s necessary to exclude the Hermione’s from the levers of power, but to require masculine norms in their operation.
I may be naive in thinking the maintenance of such a precarious balance is possible. The referenced women existed in environments where they were a fraction of a percent of the total population. Effective Altruism, despite selecting for women with the most masculine of norms, is succumbing to feminization with only a quarter of survey-takers being women. LessWrong and ACX survey-takers are about 12% female.
Of course, very little of this is actionable, but it’d be interesting to identify the threshold at which it snowballs. As a side note, I agree with Helen Andrews that feminization will be disastrous for the rule of law, so we have that to look forward to. While the influence is already clear, it will do nothing but strengthen.
DanielHobson
10mo ago
I am generally inclined to agree with your assessment of “luxury beliefs” and who practices them.
A great recent example is the mayoral election in Chicago, which is one of the most direct elections on the issue of crime. Looking at only white voters (to account for demographic confounding), the wealthy voted for Vallas, while the status-poor voted for Johnson. If you read between the lines, it’s well broken out in this FiveThirtyEight article. Their Bo-Ho is essentially what you characterize as the “Outer Party.”
Henderson’s original conception does better capture their insulation, though. Bo-Ho Chicago crime has risen dramatically since 2020, and their votes have contributed to that, but it would be unfair to say that they are the ones paying for it. The costliness of the signal is dramatically reduced by their geographical isolation from the areas devastated by their policies. 2019 saw 25 white homicide victims, while the peak in 2020 was only 32. In contrast to black homicide victims going from 415 to 627, it’s clear the white Johnson voters aren’t really “reaping what they’ve sown.”
DanielHobson
11mo ago
No alternatives means nothing and is entirely up to discretion. If an alternative results in 50% more diversity hire but 1% less efficiency, is it viable? What about 10% more diversity for 80% less efficiency? I doubt there are many alternatives found that result in increased efficiency, and if there are, the firm that doesn’t implement them will be punished by the market.
Thanks for sharing the Palladium article. It’s a death spiral that I remain more pessimistic about than Hanania, and my ideations have shifted from how best to change it towards how best to avoid the catastrophic consequences.
DanielHobson
11mo ago
I knew about the history of judicial decisions, but was unaware of that legislation. Thanks for correcting me! I think represents a meaningful challenge to the vision presented by Hanania in the Federalist Society speech to which I linked. He operates under the impression that only Griggs need be overturned. In my estimation, after reviewing what you’ve provided, is that the ruling of Griggs is now enshrined into law and no longer reliant on precedent.
What then is the path forward? My initial reaction would be a wide-reaching ruling that recognizes that intelligence is the single best predictor of job performance, and so any semblance of g-loading makes a test or requirement meets the standard “that the challenged practice is job related for the position in question and consistent with business necessity.” Standardized tests, IQ tests, leetcode, etc. would be de facto protection for all hiring on merit, with disparate impact damned.
Does that seem like a viable path forward in your estimation?
DanielHobson
11mo ago
No aspect of the CRA would actually need to be repealed to achieve most of what Hanania wants, which is the elimination of disparate impact doctrine. That is not enshrined in the text but was instead created through bureaucratic EEOC decisions, executive orders, and legal decisions. One president who takes interest in the issue, along with a favorable decision from the Supreme Court, could eliminate the legal basis of the ideology.
If you think that isn’t enough, I’d probably agree. I am in favor of giving people the freedom to discriminate however they want. The only durable discrimination is that enforced by law. Unjust discrimination really is too unprofitable in a free market. If Mormons want to live in a neighborhood that bars residence of non-Mormons, they should be allowed to. They will pay for it through reduced home prices, but I have no desire to infringe upon their ability to make that choice.
Further, without repeal or amendment, businesses are still liable to bogus workplace discrimination claims that receive outlandish payouts.
So DEI would not disappear, it just wouldn’t be compelled by law. That would still be a meaningful step forward from the current state of affairs, with no legislation required.
DanielHobson
11mo ago
The killing is only tangential, and likely accidental. If someone is presenting a credible threat of violence, is it right to physically restrain that person, even if the restraint carries a chance of death?
It is almost impossible to have a physical altercation with a 0% chance of lethal outcomes. Punches kill people all the time. Seemingly non-lethal chokes (like a blood choke that you release when struggle stops, which is what the video seems to show) always carry a risk.
The marine does not snap his neck or do any other obvious action expressing intent to kill. So is he justified in doing any physical restraint?
DanielHobson
1yr ago
The banning of Juul, an e-cigarette, was an eye-opener. No legal pretext, no bans of similar/identical products, just one company that was chosen to be crushed by an unaccountable bureaucracy. It’s really not that different from the common conception of the CCP’s level of market interference; all American companies and products exist at the will of bureaucrats.
It seems like there should be ways to work around this, such as by having an LLM rephrase your text. This isn’t very good for long-form content, but good enough for a Twitter account where specific ideas are more important than style.
More options
Context Copy link