Post of the year award: 2025
Winner: Richard Hanania
Richard Hanania explains the based ritual: any objection to right wing opinions has to be due to being too right wing. “Repealing the 19th is bad because nobody should vote anyway” or “hating minorities too much is bad because that diverts hate away from women”.
Anonymous woman describes her dating experience in the bay area on Scott Alexander’s blog, and describes a few types: “the man who is not”, “the man who is”, “the founder”, among others. A little fluffy (to be expected), but conceptually dense, raw, and original.
Deep Left Analysis argues that the “housing shortage” is more accurately described as a “desirable area shortage”.
Phoebe Arslanagic writes a Works in Progress article on the low fertility of South Korea. She argues the problem is the career/motherhood conflict + no marriages + education system. I’ve also seen people make the case that conscription + career/motherhood conflict contribute to an unusually intense gender war.
Baazaa on the decline of socialisation.
Sectionalism Archive argues the internet is a hellish space, where everything is evanescent yet permanent, devoid of value yet enthralling, and corrosive to the construction of identity. It’s also a review of Serial Experiments Lain, one of the best anime ever written.
Uncorrelated on how much AI use affects productivity for software engineers — about 0.5 SD, or 7.5 IQ points. This might seem small, but at scale, this is massive.
Paper on ability correlations between almost a million family members in Norway was published, detailing 16 different relatedness correlations. Even genetically unrelated cousins exhibited a small correlation in IQ (like 0.01), indicating that environmental factors can be concentrated in families as well, not just households.
State of peak oil in 2025: looks like it will happen, but we don’t know when.
New theory of aging: aging reflects a loss of goal direction. Definitely not my field of expertise, but it’s an exciting theory.
Maxim Lott finds that the IQ of LLMs has been rising across time, GPT seems the most dominant across the various rankings and time periods.
Inquisitive Bird finds that cohort-based fertility projections differ from normal ones, as changes in the TFR can be affected by changes at the ages at which individuals have children over time.
Zagrebbi’s advice for young, ambitious right wingers: networking is a meme, just become friends with people you like; send more cold emails; and go to more conferences/parties.
Scott Alexander dives into the data on the vibecession, noting that public sentiment on the economy has never been worse, but on paper, the economy looks fine. He tries diving into a few theories as to why, and doesn’t come back with much. Personally, I think the economy is fine, and that people’s frustrations are more related to hyper-credentialism, a desirable area shortage, and a rough job market.
Dmitry (from the dosage makes it so) writes biofoundationalism: the moral genotype, a post on how political attitudes and values are biologically ingrained.
Cremieux on how to do scientific criticism properly: it must be correct, relevant, meaningful, and aware of the specifics of the field.
Nate Silver exhaustively dives into the data on mental health and political beliefs.
Katherine Dee on the anti-smartphone grift. I am conflicted on the smartphone question. I think they are bad. But I also disagree with everybody on why they are.
Walt Bismarck makes his case against moral realism, which is bad, but I think that the path beyond is creating a new morality, not mere nihilism.
Deep Left Analysis defends us GDP bros.
Deep Left Analysis and Sectionalism Archive defend… The age of consent. Not exactly a hot take, but contracontrarianism is an unappreciated art.
More contracontrarianism: Deep Left Analysis argues… Sexual abuse is traumatising.
Caribbean Rythms 192, where BAP platforms Nick Land.
Scott Alexander reviews and defends embryo selection.
David Pinsof on why AI doomerism is noise.
UBERSOY argues LLMs replace the function of reddit.
Sectionalism Archive on why Muslims tend to be less intelligent than non-Muslims even when controlling for race and location. He mainly chalks it up to lower SES Christians and Jews converting to Islam in Muslim areas that enforce jizya, selective migration into the Ottoman Empire, and incest. He discounts the theory that Islam is a “stupid religion”, as the same could label could be applied to any other.
Bronze Age Pervert on the compact magazine DEI article — on one hand, white men have become increasingly excluded from positions in film and media for a long time. On the other, the author, who was denied a position as a screenwriter or something like that, displays no initiative or ability to do things on his own.
Uncorrelated on the rise of incels: it’s a US-specific phenomenon, and does not generalize to other countries like Australia and Italy.
Cremieux writes an 8623 word article on why national IQs are valid.
Scott Alexander dives into the missing heritability problem with help from a few geneticists.
Caribbean Rythms 196, where BAP talks to Mr Star, a pro-tech right wing twitter account who talks about China, birth rates, and society.
Walt Bismarck on what he learned this year.
Emil Kirkegaard on the uselessness and low quality of medical research.
Dr Monzo on the Faustian Work Ethic, summarized as optimizing for working less hours and making more money.
Sectionalism Archive’s review of Code Geass: “Instead of being a pozzed show that to any normal eye is not pozzed, it is actually an extremely apolitical show hiding behind a mask of politics.” He rated it a 9/10.
Nuance pill on why self-reports of sexual activity on surveys can mostly be trusted.
East Hunter defends hereditarianism.
Immigration research, unsurprisingly, is biased.
Mr Star on America’s new national grand strategy: it’s an explicit, open documentation of America’s ambitions, though its framing of America being an exploited or defrauded country is wrong. China’s rise as a power shouldn’t be seen as something to prevent or fight against, but to cope with and manage.
Aella posts on whether there are birth-order effects for homosexuality. Seems pretty clear that there are, where younger male siblings are gayer than older ones.
Yglesias on why Iraq decided American politics for 12 years.
The typical man disgusts the typical woman, by Bryan Caplan. A nice, self-explanatory title.
Cremieux finds evidence that Columbia is still discriminating using leaked data.
Arctotherium on why immigration does not solve population decline.
Worst Boyfriend Ever details how he prostituted himself on a gay ranch.
Scott Alexander gives a lukewarm defense of national IQs.
Aella writes on how we are delusional about how hot we are.
Misses from 2024
The Machiavellians summary of Nietzsche’s theory of decadence (video, written). Essentially, decadence is when human beings do things that are bad for them. Humans have drives which make them do things that are good for them, and then humans layer consciousness and morality on top of these drives to be able to live together. Repression creates internal division, and the will to power finds itself unable to exert its influence on the world because conflicting instincts waste energy.
Baazaa’s entire blog on managerialism, work culture, and the internet. Best content I’ve seen on the internet in a long time.
Bitcoin is the worst cryptocurrency, and that is why it will win.
Other things I have read
I read far less than I should. I concede most of the anti-book arguments: that most books are not worth reading, even popular ones, and their thesises could be distilled into articles that are 10% of the length but contain 90% of the message.
That said, most of the extensive, specialised knowledge that I have accumulated — in genetics, intelligence, philosophy, and climate change has either come from reading long articles or books. I’m not sure why this is, but I think some of this comes from the time investment allowing me to eliminate bad intuitions that seem initially plausible but crumble under more careful thought.
I read Zero to One in one day, and it is the best book I have ever read. It explains why some businesses generate more revenue than others, why it is possible to consistently succeed without the intervention of luck, and what kind of people tend to be successful entrepreneurs.
I’ve also started reading Beyond Good and Evil, and it’s fantastic. I was previously reading the Will to Power, which was mostly nonsense with a few extremely good ideas thrown in, and the Antichrist, which was better but repetitive. I finished Twilight of the Idols before that, which was good read, but was mostly stuff I already knew about Nietzsche.
Beyond Good and Evil, in the 60 pages I have read, has barely spoke about morality at all. Nietzsche, instead, attacks a lot of our assumptions about seeking truth and the will, which is clever — if you take away rationality and unidimensional model of the self, then the case for moral realism is rather thin.
I read Isabel Myers’ magnum opus on the MBTI, Gifts Differing in 3 days after that. It’s also great. The TL;DR is that:
The MBTI first began as a philosophical model of human types which was developed by Jung, and later refined into a formal and testable model by Isabel Myers and her mother, Katherine Briggs.
The MBTI is not just a model of personality. It’s a psyche model which evaluates how a preference (introversion/extraversion), perception style (intuiting/sensing), judgement (thinking/feeling), and orientation (judging/perceiving) create different archetypes.
All of these facets interact to create the eight functions: extraverted feeling (preserve social harmony), extraverted thinking (what works), introverted feeling (what I care about), introverted thinking (what is true), introverted sensing (patterns from the past), extraverted sensing (what is occuring), introverted intuition (one vision of the world), extraverted intuition (evaluating different visions of the world).
These types then value these functions at different levels. For example, in INFPs, the rank order is Fi > Ne > Si > Te . Functions need their opposites in order to create balance; Fi + Si just leads to endless narrow thinking based on current values and past experience, rather than taking in information from the external world (Ne, Te).
On one hand, there is function dominance, which is what I explained before, and then there is function development, which is defined as how effective the function is used. Typically, people develop their most preferred functions first in their youth, and the other six are either developed in adulthood or not at all.
Some statistics I recall: 75% of Americans have a sensing preference, but only ~10% of Rhodes scholars, and ~20% of national merit finalists do. Extraverts were more likely to be matched on the other 3 facets of the MBTI. The occupational hierarchy looks is as MBTI-stratified as IQ-stratified; finance, banking is full of STs, law is full of Ts, science is full of Ns, manual labour is full of Ss, and whatnot.
All of the conventional criticisms of it (pseudoscientific, low predictive validity, astrology-lite, needless dichotomisation, just the big 5 but 4 factors) are wrong or don’t understand what the model is trying to do.
I might write a macro-post on how it works + how it differs from the big 5 (which works, but I don’t consider a real or interesting model of personality).
I’m glad that you appreciated my articools. Have a good new years!