Links 9/15: Linker Tailor Soldier Spy

Little-known footnote to the Scramble For Africa: Russian Somalia lasted one month before being conquered.

Latest big meta-analysis finds “little evidence” for Roy Baumeister’s ego depletion hypothesis, describes itself as “strongly challenging” the idea of willpower as a limited resource.

Why have murder rates suddenly shot up in St. Louis and Baltimore? The obvious thread linking those two cities doesn’t seem to be it; the rise appears to have started before the recent round of shootings and riots. Hopefully this doesn’t represent a halt or reversal of the secular trend towards decreasing crime – but if it does, then it could be another clue to what’s going on.

This week in neat methodologies: evidence in bias at Olympic games. Winners from last year gain a greater premium in subjective contests (eg figure skating with judges) than objective contests (eg the hundred meter dash), suggesting that judges are biased upward by an athlete’s previous reputation.

Headline of the week: The Weather Channel Reports Firenado On Kentucky Lake Brimming With Bourbon. I hope Sharknado 5 is about a fire-bourbon-shark-nado.

Also this week in neat methodologies: we don’t fully understand the genetics of intelligence, but we do know a couple of genes with small effect. Our model for natural selection on intelligence suggests that selection pressure should affect the various intelligence genes equally. That means the few genes we know offer what could be a representative sample of the entire structure that we could use to detect or refute such selection. And that in turn means that we can test for a genetic basis of intelligence differences between populations by checking how they differ on the few genes we know. Davide Piffer tries this and finds a correlation of 0.9 between genetically-predicted intelligence and actual IQ on the population level. But he gets correlations almost as high between randomly selected genes and actual IQ, which could either mean those randomly selected genes are linked to intelligence-related genes (his theory) or his methodology sucks (eternal default theory). Also, I am suspicious because you never get a real correlation of 0.9 with anything; even measurement error should be worse than that! Also, the author doesn’t do so well with my usual test of “does viewing his Twitter account increase my confidence in him as a person?”

This is probably a metaphor for something: Chimp Wielding Stick Takes Down Drone. Next time they should hold out for a craft with more experience dodging sudden giant animal attacks

The Marines recently did a study showing that women performed less well than men in combat, but as best I can tell it was fatally flawed; since the military has only recently started accepting women, all the women were new but many of the men were experienced. That just means experienced people do better than new people, which is unsurprising. Come back when you’ve got equivalent groups.

Bisexual women are more likely than straight or lesbian women to have an eating disorder, confusing everybody regardless of their theoretical perspective.

David Friedman’s Law’s Order starts off by explaining why there should be a higher penalty for armed robbery + murder than for armed robbery alone, or else armed robbers will have no incentive not to kill their victim and eliminate a potential witness. A recent Global Post article argues that the Chinese have screwed this one up by making a law that if you run over someone and they live, you have to pay for their medical care; but if you run over someone and they die, you have to pay a flat fine which is often much less. Therefore, Chinese drivers who accidentally hit someone will try to run them over again to finish the job. Snopes isn’t buying this and can’t find any sources, but I vaguely remember reading articles before in the Chinese-English media that were sort of like this.

A person who is not Ozy nevertheless analyzes the evidence on the results of gender transitioning, finds best studies show it can improve mental health.

The original Sokal hoax – critically acclaimed avante-garde artist Pierre Brassau turned out to be a chimpanzee.

Harold Lee is always thought-provoking and worthwhile. The current thing of Harold Lee’s that is thought-provoking and worthwhile is on the Manchus’ self-conscious attempts to avoid decadence in Imperial China and how it relates to modern immigration.

Clever in a Rube Goldberg sort of way: cows keep eating metal, which damages their digestive tract. The solution: feed them a big magnet that stays in their stomach and prevents metal objects from going further.

Doing your arithmetic in base phi (ie 1.618…) is less crazy than you’d think.

In 1968, Britain was skimping on celebrating the 50th anniversary of the RAF for political reasons. An RAF pilot decided to take matters into his own hands and fly a fighter jet around the city of London without permission, including passing through the Tower Bridge. Everyone had a big laugh about it and he was allowed to leave the RAF on “medical grounds” without penalty. Simpler times.

The street dogs of Moscow have adapted to their urban environment, including learning how to use the subway train system to get where they need to go. Also some speculation that the packs enforce on their members a policy of not defecating in highly-trafficked areas in order to avoid conflict with humans, and that they are gradually switching from a pattern of strongest dog as pack leader to one favoring the most intelligent dog. Also, they seem to all be well-fed and to avoid competing with each other for food so that everyone can get their share. Actually, is there any way we can put Moscow street dogs in charge of the government?

Study on the effect of investing in schools finds “very precise zero estimates” of achievement effects.

People complain that mining the Alberta oil sands is environmentally damaging. But the original plan didn’t involve the sort of inefficient mining techniques being employed today. The original plan was to detonate a hundred nuclear bombs over the oil sands, in order to boil the oil to the top. This is maybe the most 1950s idea I have ever seen.

Very nice small study with neat methodology finds that more intelligent people are more honest.

78% of the variance in national income today is predicted by the variance in national income in 1500 AD. Much of the effect remains even as far back as 1000 BC, insofar as we can measure that sort of thing.

I Have One Of The Best Jobs In Academia; Here’s Why I’m Walking Away. The first half kind of makes me want to punch the guy, but the second is a really good analysis of how we ended up in the higher education bubble and what we can do to stop it.

In case you hadn’t heard, Jeremy Corbyn did in fact win the Labour Party leadership. Some discussion of his policies on Reddit, which range from “actually sounds pretty good, not sure why everyone’s so worked up over this” to “legitimately terrible idea, but not obviously moreso than everyone else’s terrible ideas”.

In 1956, Mao Zedong declared a policy switch toward greater freedom of speech, and invited party officials and intellectuals to criticize him and his leadership so he could learn from their advice. In 1957, Mao declared that it had all been a trap, and killed everyone who had spoken up.

“I don’t think any currently published translations capture the original [spirit of the Iliad]. The Iliad we know today is the written form of an entire oral tradition unto itself, which ancient epic poets whose names and numbers we’ll never know invented on their feet, their genius and imagination driven forth by musical accompaniment that kept the dactylic-hexameter beat. The modern listener should find the subject matter familiar as well: drink, booty, fly ladies, stacks of treasure, macho violence, and ostentatious modes of transportation.” Lula writes the Iliad as a rap and IT IS AMAZING.

Ben Henry going over some problems with Vox’s article claiming that Donald Trump made no extra money over what he’d have earned by putting his inheritance in an index fund. Quick summary: it ignores consumption (!), doesn’t really account for taxes, and in one case suggests that he should have magically picked the best moment to invest beforehand.

On the other hand, Vox’s test for political bias is actually really clever and really good. It asks you a bunch of factual questions about politics, like “Do studies show capital punishment decreases crime?” and then sees whether your errors form a consistent pattern that flatters your preconceptions. Both authors are part of the broader rationalist/EA community.

A social worker with more experience doing actual therapy than me reiterates that not-having-trigger-warnings is not the same thing as exposure therapy and shouldn’t be billed as a mental health good.

I’m not sure I’m understanding this article right, but if true, it’s…different. Half of all Japanese universities will close or scale back their humanities and social sciences departments after the government orders them to “serve areas that better meet society’s needs.”

Homeopathy Conference Ends In Chaos After Delegates Take Hallucinogenic Drug. Didn’t something like this happen in a Robert Anton Wilson book?

Last week I wrote an article on politics where I claimed that the left was moving leftward and the right rightward. I got a bunch of angry comments, half of which thought I was crazy because both parties were obviously moving rightward, and half of which thought I was crazy because both parties were obviously moving leftward. Over at his own blog, Free Northerner makes the case for a general leftward shift. Suggest you take your comments there instead of inflicting them on me again.

Harvard University declares it “can’t afford” closed access journals and asks its researchers to support the open access system. Despite the fact that their reasoning is obviously a lie (Harvard could afford journals printed upon platinum leaf in ink made from the blood of endangered white tigers) this will hopefully be a big boon to the open-access research movement.

The 2016 presidential polling has some notable anomalies. And not just the fact that people are still supporting Trump.

The amount of First World inequality attributable to education is very low, and education-related policies are unlikely to impact inequality very much.

Inside the Chinese city where the average man has three girlfriends.

More really neat methodology: when it’s dark, it’s hard for police to see what race a driver is, making them less likely to racially profile. By comparing ratios of black drivers to white drivers pulled over, and how the ratio varies during different lighting conditions (summer, winter, daylights savings time, areas with more streetlights), researchers find the odds of a black driver being stopped relative to a white driver increase 15% when police are better able to distinguish their race.

US companies are less likely to be publicly-traded than in the past.

I FINALLY FOUND THE STUDY I HAVE BEEN WANTING SOMEONE TO DO FOR AGES. The Greater Male Variability Hypothesis suggests that although men and women have similar average IQ, men will end up overrepresented in high-IQ occupations like “physicist” because men have higher variability (and therefore more men will be at both the high and low extremes). Some studies have shown preliminary support for this idea, but of course IQ is really hard to study and frequently confounded by cultural stuff. But the hypothesis says this variability should exist for every biological trait (it’s because men have only one X chromosome, so all their recessive mutations are expressed). So the best way to study it would be to study a bunch of easy-to-study obviously biological traits and see if men have more variability among most of them. And they mostly do, in categories from blood bilirubin levels to 60-meter-dash-running times (women are more variable along other parameters, including BMI and thyroid hormone levels, but there are fewer in this column). Overall the paper seems to provide significant albeit inconsistent support for greater male variability.

If you’ve ever wanted a really perfect way to intuitively compare the size of two countries/states/continents without worrying about things like Mercator projection distortion, is what’s been missing from your life until now.

Mitochondria have always been creepy. But now it’s starting to look like they talk to each other. Let’s hope they’re not chatting about how eukaryotes suck and it’s time to take over.

In these times of Syrian refugees, did you know that most of Haiti’s wealthiest families are Syrian immigrants?

Good (?) job opportunity for American women – make $100,000 + as a surrogate mother for a Chinese couple.

Speaking of which: in The Gambia, babies born in September have seven times the mortality rate throughout their life as babies born in June. Probably because in the first case Gambian mothers eat way more vegetables at the appropriate seasons. Unclear whether this is just folate or something more important; very unclear what lessons first-worlders can learn from this.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

724 Responses to Links 9/15: Linker Tailor Soldier Spy

  1. Bruce Beegle says:

    Four days late, and I’m surprised that nobody has mentioned Wyoming Knott (from The Moon is a Harsh Mistress, 1966):

    But hosting is a good profession for Free Woman. It’s high pay. Some Chinee families are wealthy and all my babies have been Chinee–and Chinee are smaller than average and I’m a big cow; a two-and-a-half- or three-kilo Chinese baby is no trouble.

  2. John Ohno says:

    > Didn’t this happen in a Robert Anton Wilson book?

    Yes — in the Illuminatus! Trilogy. But it was LSD, it was an american political convention, and everybody ended up getting along better at the end (rather than everything descending into chaos). (And again, in the same book, several different people spiked the water supply at a large outdoor concert in Germany with LSD, but people were expecting it.) Also, in Masks of the Illuminati, Aliester Crowley spikes the drinks of Albert Einstein, Carl Jung, James Joyce, and another character with some unnamed hallucinogen in Germany.

    I guess I’m saying it’s a pattern.

  3. Abigail Novick says:

    The ego depletion meta-analysis is probably wrong. Here’s a good critique of their statistical methodology:

    News of Ego Depletion’s Demise is Premature: Commentary on Carter, Kofler, Forster, & Mccullough, 2015
    Michael Inzlicht, Will Gervais, Elliot Berkman

    Carter, Kofler, Forster, & McCullough (in press) conducted a bias-corrected meta-analysis of the so-called ego depletion effect to determine its real size and robustness. Understanding the size and robustness of the ego depletion effect is of some import, given how foundational the phenomenon has become to our general understanding of control and effort over time. Carter and colleagues’ analysis is an improvement over past meta-analyses for a number of reasons, most notably because they addressed and corrected for the inflationary effects of publication bias using multiple advanced statistical techniques. Despite our enthusiasm for their work, we are skeptical of their conclusion that the ego depletion effect is not real (d = 0) for three reasons. First, the statistical techniques they used to draw their conclusion relies on a faulty core assumption; second, they performed many of their tests on an inappropriately small number of studies, leading to an attempt to correct small study effects with small studies; and third, and most critical, we demonstrate that the statistical techniques they used are themselves biased and insensitive to the presence or absence of effects. When we apply a far more sensitive biascorrection technique, the p-curve, we uncover a meta-analytic effect of moderate size (d=0.55) [0.45, 0.59]. Despite our admiration for this line of meta-research, our analyses suggest that Carter and colleagues’ correction attempts might themselves be in need of correction.

  4. Deiseach says:

    Scott, you downplayed the probability of a serious meteor strike hitting Earth in the AI risk post, but now experts (well, an expert) says we are all (possibly) going to die!

    Professor Robert Walsh, executive director of research at The University of Central Lancashire, is of the opinion that threat of a meteor strike shouldn’t be ruled out completely even if NASA has said that there is no danger.

    NASA stressed in the official statement released in August that the claims of such a strike are completely without a scientific basis and that if there were such a large asteroid hurling towards, its Near-Earth Object Observations Program aka “Spaceguard” would have detected it. “There are no known credible impact threats to date — only the continuous and harmless infall of meteoroids, tiny asteroids that burn up in the atmosphere,” NASA said at the time.

    However, Professor Walsh cited the vehicle-sized meteor that exploded in the atmosphere above Chelyabinsk in Russia and said that NASA had actually completely missed this meteor. He adds that the US space agency was in fact oblivious to it hurling towards Earth and if the explosion would have occurred over major city like London, the effects would have been “devastating”.

    Citing the history of Earth, professor Walsh said that annihilation of dinosaurs from the face of our planet is one of the most perfect examples that such devastating impacts do happen and that we should be prepared for it.

    You see? Instead of diverting money and effort to AI risk, we should all be working on the imminent meteor annihilation threat! 🙂

    • Scott Alexander says:

      It’s easier to miss a vehicle-sized meteor than a city-sized meteor.

      I’m not sure how “devastating” it would have been if the Chelyabinsk meteor hit London. My guess is double digit death toll, which is less than some landslides. Also, I think something like 0.1% of Earth’s surface is city, so the chance of it hitting London is minimal.

    • drethelin says:

      2 things: there’s no obviously useful charity to donate to here working on the issue, it’s just something that astronomers and the military are already doing, ie scanning the sky.

      And the second is that if we solve FAI the AI will be able to help stop the asteroids 😀

      • Doctor Mist says:

        there’s no obviously useful charity to donate to here working on the issue

        If someone actually wants to make a donation toward averting the imminent meteor annihilation threat, there is always the B612 foundation (

  5. zslastman says:

    So I David Piffer’s methodology is odd, but I can see a pretty obvious source of bias there and I don’t think his ‘corrections for population structure’ would do a good enough job of correcting for it.
    Your ability to find an allele’s effect in the population is dependant on it’s frequency – you’ll never identify it as affecting IQ if only 1 or 2 people in your sample have it. Thus our IQ alleles are pre selected for being high frequency in the populations we found them in. As are random SNPs from a database in fact. This would explain his results I think. (wouldn’t swear by it until sitting down to read the original methodology paper).

    • zslastman says:

      Also the snps need to be causal in both the populations, as opposed to interacting with another variant or just marking another nearby causal variant. The author doesn’t seem conscious of the difficulty in ascertaining this.

    • Scott Alexander says:

      I’m not sure that’s right. The easiest way to “find” an SNP is if it’s 50-50 in the population you’re looking in (right?). That means that if you look at another population, there’s equal chance it will be higher and lower.

      Also, I assume most of the SNPs were found in European populations, but some of Piffer’s populations are higher IQ than Europeans and others are lower IQ than Europeans; in both cases the SNPs seem to go in the right direction.

  6. Who wouldn't want to be Anonymous says:

    I was wonder what take any of you (Scott?) might have about this Paxil business.

    • Scott Alexander says:

      Not too sure yet as child psychiatry is far from my area of knowledge, but some points worth considering:

      1. Paxil sucks in general and is strictly worse to other antidepressants except in very specific cases.
      2. Paxil is not approved for adolescents in the United States; other SSRIs like Prozac and Lexapro are.
      3. This study is interesting iff the findings generalize from Paxil to other SSRIs, which they very well might
      4. David Healy is a well-known anti-antidepressant activist with a strong axe to grind. This doesn’t make him wrong, and he’s earned some respect, but I’ll still be more likely to believe this when I see people other than him finding it.

  7. Alia D. says:

    After I did the quiz on political bias, I was reading the part about how people that were getting paid for right answers did better on certain “tests”. A confounding problem with “knowledge” surveys could be confusion at an unthinking level about whether the respondent was taking a test with right answers or if the responded was being change to vote on what the consensus answer should be. Certain polls with questions fairly similar to some of these get treated by advocates as “evidence” that they ask politicians to respond to. Some of the questions on the quiz are ones where there is an ongoing debate with at least some plausibility to some evidence on more than one side. Whether you are trying to evaluate where the balance in the argument lies for your own private self or are you trying to weigh in on your side of the question could lead to you giving different answers

  8. Arthur B. says:

    The X chromosome may might be the “how” of greater male variance, but the “why” is polygamy, competition with other men has a convex payoff.

  9. Shenpen says:

    >The Marines recently did a study showing that women performed less well than men in combat, but as best I can tell it was fatally flawed; since the military has only recently started accepting women, all the women were new but many of the men were experienced.

    Yes – but they get the very best of women recruits, while they get perhaps upper-mediocre male recruits as the best men will go to some kind of Special Forces, paratroopers, green berets, whatevers etc. are todays elite male high school athletes likely to enlist as a mere Marine?

    • John Schilling says:

      I don’t know how things work in your country, but the United States Marines are generally considered an elite force, roughly equivalent to paratroops. We’ve got an entire Army for people who want to serve their country, collect a paycheck, learn job skills while putting up with military BS, etc, etc. And then we’ve got the Marines, for people who have different priorities.

      • HlynkaCG says:

        Pretty much.

      • Sastan says:

        Nah dude. Marines are basically Infantry. I know we have this inter-service rivalry, and the Marines put a lot of effort into selling the public on how hardcore they are, but there’s not much to that. Same with Paratroops. They’re Infantry that learned to jump out a plane. Marines are Infantry that learned to deal with the sexual advances of the Navy while on ship. Which is more “elite” depends a lot on who is currently in command, and what raft of NCOs they have.

        Their basic is longer, and if my uncle is to be believed (he was in both services, Marines first, then Army) a bit tougher, but nothing like Ranger School or Selection.

        They do seem a bit more strict about the D&C aspects, but that doesn’t make “elite” troops. In fact, the reverse is usually true.

  10. Phil says:

    The most 1950s idea ever is clearly Project Pluto. It’s just no contest: who wouldn’t want to build an unshielded nuclear powered low altitude ramjet that you can’t stop once you start it up?

    The US managed to spend $2billion (2015 inflation adjusted $) on this monstrosity before someone with enough clout stopped to ask the obvious question: “How exactly are we going to test it without killing everyone for hundreds of miles in every direction?” Follow-up questions like: “This thing has to fly at low altitude to get under the Russian radar: Surely irradiating every friendly nation on the flightpath between us and the USSR (plus bonus shockwave damage: this thing is very supersonic) is a really bad idea?” were just nails in the coffin lid really.

    • It’s a crazy idea to try to use Pluto on the Earth but I can’t help feeling it would be a pretty reasonable way to explore Venus or the gas giants.

      • Nornagest says:

        A nuclear ramjet would make an awesome atmospheric probe if we could get it there, but getting it there is tough. There’s a lot of mass on the SLAM — partly because it was designed to be stuffed full of hydrogen bombs, but even without that issue you can only make its reactor so light. And once you have a rocket capable of schlepping a city bus full of bricks to Jupiter, you might want to start thinking about other things you could be doing with it.

    • FacelessCraven says:

      Project Pluto! The very coolest of cold-war nightmare weapons!

      …And obviously, we launch it from some godforsaken corner of Alaska, directly into the heart of the Soviet Union itself.

    • Sniffnoy says:

      I think Project Orion has it beat. Project Pluto would have been propelled by a nuclear reactor; Project Orion would have been propelled by setting off atomic bombs behind it.

    • Who wouldn't want to be Anonymous says:

      Project Pluto may be one of the most 1950’s ideas ever, but it has absolutely nothing on the total waste of money on something that will never fly that is the F-35.

      • Nornagest says:

        If they’d stopped wasting money on the F-35 before they got to the point where they had something flying, we’d be many billion dollars richer.

      • John Schilling says:

        The F-35 has completed over ten thousand hours of flight testing, and entered operational service with the United States Marine Corps earlier this year.

        Overpriced, yes, but this is not one of the forums where hyperbole makes you seem wise or persuasive.

        • Who wouldn't want to be Anonymous says:

          Operational” is where the hyperbole is.

          Or better yet, let the Director of Operational Test and Evaluation tell you himself. “The event was not an operation test, though, in either a formal or an informal sense of the term. Furthermore, it did not–and could not–demonstrate that the Block 2B F-35B is operationally effective or suitable for use in any type of limited combat operation, or that it is ready for real-world operational deployments, given the way the event was structured.”

      • FacelessCraven says:

        I dunno, it sure seems like the Osprey is up there in terms of fantastic boondoggles.

    • AlexanderRM says:

      My favorite part is the line on the Wikipedia page, “The nuclear engine could, in principle, operate for months, so a Pluto cruise missile could be left airborne for a prolonged time before being directed to carry out its attack.”.

      In fairness, the Wikipedia page also says “cruise in circles over the ocean”, which I assume would mean the Arctic ocean for the most part, so that’s environmentally irresponsible and would send a fair amount of radiation throughout the upper atmosphere, but both of those are probably not massive in comparison to the actual impact of a large-scale exchange of nukes between the US and Soviet Union.
      …actually reading more of the Wikipedia page pretty much ruins the humor; they were in fact aware of the irradiation and shockwave, and it was actually closed down because the Pentagon felt it was “too provocative”.

      (Incidentally: As a result of a video game, I associate the phrase “cruise missile” with conventional chemical weapons as distinguished from nuclear ones, so for a moment I thought the project was to build a missile with nuclear propulsion to deliver a chemical warhead. XD)

  11. Setsize says:

    NIMH director Tom Insel blogged: “The biomarkers for depression and psychosis and post-traumatic stress disorder are likely to be objective measures of cognition and behavior, which can be collected by smartphones.” This was posted two weeks before Insel announced that he would step down from NIMH and join Google Life Sciences.

    • Deiseach says:

      The biomarkers for depression and psychosis and post-traumatic stress disorder are likely to be objective measures of cognition and behavior, which can be collected by smartphones

      My immediate reaction to that starts with “F” and ends with “off”. Yes, I’m quite happy to let Google etc. have access to intimate personal mental and physical data! I do not consider they might use this data in any way other than strictly confidential medical applications! I certainly don’t think I’d be spammed with ads for “Ask your doctor to prescribe you new BrainBlasta from PzillySK” if my smartphone monitor detected I was down in the dumps!

      I didn’t even give them my real name when I had to sign up for a Google gmail account, why the hell would I let them have access to actual real data about my current state of mind and body? (Yes, I am that paranoid. Or careful. Or hate people that much. Take your pick).

      As to this, I take leave to doubt it. 100% accuracy? Really? Unless the speech sample in question was “The goblins under my bed are telling me to burn down the school and murder you all”, I don’t think so:

      One was the publication of results from a collaboration between Columbia University and IBM. The team, led by Gillinder Bedi and Cheryl Corcoran, was looking for a biomarker to predict which clinically high-risk youth would convert to psychosis over a two- to three-year follow up period from an initial interview. Rather than depend on a protein in blood or a brain scan, they used an innovative big data approach to analyze the speech from the initial interview. The approach, developed by Guillermo Cecci at IBM, maps semantic coherence and speech complexity as a window into the earliest stages of disorganized thought. While analysis of previous clinical features have yielded, at best, 80 percent prediction, this automated analysis of unstructured speech was reported to be 100 percent accurate for identifying who would convert to psychosis during the follow up period. This is a small study (34 participants with 5 developing psychosis), but it serves as a preview of what we might see as the power of technology is applied to provide objective measures of behavior and cognition.

      If they can make anything coherent out of normal teenage speech, which tends to go “Yeah, well, right, you know?” *shrug*, then I’m impressed, but I think diagnosis via speech analysis has a long way to go yet. It sounds interesting, but I wouldn’t hold out the idea that we can do away with face-to-face meetings with trained and qualified personnel; I’m doing online cognitive training as mentioned in that post, and it’s going… about as well as you’d expect for someone extremely introverted and avoidant. The basics are probably helpful (I skip the ‘inspirational life stories’ examples) but unless I have a real person there to hold my nose to the grindstone, I know I’m not going to change. No amount of perky once-a-week messages from my online supporter are going to do any different, especially as I am not going to put up online anything more than the bare minimum of personal information (‘deny, obfuscate, and keep your trap shut’ is my motto for all interactions where I’m not protected by absolute anonymity. You lot on here have learned a lot more true facts about me than any real person in meat space).

    • Scott Alexander says:

      Insel’s stepping down from NIMH?

      Oh, the people I know are going to be SO HAPPY.

      On the other hand, five years from now if you try to do a Google search, you will get a lecture on how real searches should be rooted in neurobiology, and there won’t be any option to skip it and go to the page you want.

      • Setsize says:

        He probably deserves that snark, but I thought the interesting part was his apparent change of tune (praising behavioral measures/interventions over biological).

      • Deiseach says:

        I had no idea NIMH was a real place.

        This is a bit like reading Lewis’ “That Hideous Strength” and then a few years on, the UK government develops an organisation called NICE 🙂

        • DrBeat says:

          But, like, the NIMH in the title of “Mrs. Frisby and the rats of NIMH” is the National Institute of Mental Health. They did studies on rat society there. It inspired the book.

        • houseboatonstyx says:

          @ Deiseach
          This is a bit like reading Lewis’ “That Hideous Strength” and then a few years on, the UK government develops an organisation called NICE

          Yeah, every time I see that I feel scared to go to Britain. It’s clearly in the Twilight Zone.

          • Winter Shaker says:

            It gets worse. It used to be (if I remember rightly) the National Institute for Clinical Excellence … thus NICE.

            Now it’s the National Institute for Health and Care Excellence … yet the acronym remains unchanged. Something spooky going on there.

  12. Guy says:

    I’m just going through it, but has anyone managed to find the web appendix to the racial profiling article? They mention it a few times and I’d like to get ahold of it, but can’t find it.

  13. BBA says:

    “US companies are less likely to be publicly-traded than in the past.”

    Depends on what you mean by publicly traded. The number of exchange-listed companies dramatically increased in 2006 when NASDAQ officially became an exchange.

    The article indicates that until 1996 NASDAQ listed a lot of companies that wouldn’t qualify to be listed on any exchange today. It’s still possible to be publicly traded without an exchange listing but that gives you all of the costs of being public (having to file regular reports with the SEC, SOX compliance, etc.) without any of the benefits, so you’re probably better off private.

    And, of course, privately held small businesses have always vastly outnumbered publicly traded companies.

  14. onyomi says:

    I haven’t looked into it in detail, but I am super skeptical of that depletion effect meta-survey, both because of very strong personal intuition and anecdata, and also because I remember all through my life periodically seeing headlines like “couples on a diet more likely to argue, study shows,” all of which seemed highly plausible to me.

  15. onyomi says:

    So, chronologically, Ferguson and Baltimore cannot have caused the rise in crime, but it also seems like it can’t be a coincidence that those are the places where crime is going up disproportionately. Maybe whatever is causing crime to go up also caused (the resentment underlying) those protests?

    • meyerkev248 says:

      Places with rising crime are more likely to see an increased and harsher police presence as people not committing lots of crime try to keep a lid on it?

      At which point, a more ruthless, more active, possibly worse trained/less experienced iff increased crime led to recent hiring police force that’s meeting citizens more often is a lot more likely to have Fergusons.

      Combine it with the bit where AFAICT, most of these police departments were operating under serious anarcho-tyranny (Only solve 1 in 4 murders, but give out a lot of traffic tickets), and Ferguson could be less the start of the problem and more the symptom of a growing problem.

      • keranih says:

        You might find Jill Levoy’s Ghettoside: A True Story of Murder In America an interesting and informative read on the subject of the interaction between the police and the community.

  16. tanagrabeast says:

    On the link about nuking the Alberta oil sands, it should be noted that the plan was for nuclear detonations underground, a much less insane proposition than the one implied with Scott’s wording here (“over the oil sands”)

    • Winfried says:

      Aren’t the oil sands themselves underground?

      It is a little misleading, but over the sands can still be underground or it can be surface level.

      • FacelessCraven says:

        I definately read it as surface detonations, not burried ones, which made it seem a whole lot more horrifying.

  17. Emily H. says:

    On the story about the Japanese universities’ humanities departments: the Times Higher Ed article links to a blog post which links to a story that’s now behind a paywall, but having done some preliminary bilingual investigating, I think they’re dramatically overstating the case.

    As far as I can tell, the Education Minister did send a letter to Japan’s public universities telling them to abolish or reform their humanities and social sciences departments. But the ministry has since walked back from that somewhat, emphasizing a more “mend it, don’t end it” (or perhaps “you can mend it! You don’t have to end it!”) kind of approach. The only reports I can find of programs actually being shut down is with a few schools getting rid of their non-licensing teacher-education programs: normally, if you get an Education degree in Japan, you have to also get a teaching license. But there have been some programs that offer Education as sort of an easy general-education liberal-arts degree with no teaching license requirement. And a couple of schools are getting rid of that program this year.

    It could be that I’m wrong and the Times Higher Ed article is right, but often English-language reporting on Japanese news is really bad because of the language barrier, and I really think I would’ve found more Japanese news about it if 17 universities were cutting programs that broadly.

  18. Robert Guiscard says:

    That political quiz is basically BS. It doesn’t tell you whether or not you’re biased. At best it tells you whether you agree or disagree with “scientific consensus”.

    • Jon Gunnarsson says:

      That’s not how the quiz works. If your disagreement with the scientific concensus isn’t correlated with the scientific consensus going against what people sharing your political ideology tend to believe, then the test won’t pick up any bias. But if there is such a (positive) correlation, that is evidence of bias, wouldn’t you agree?

      • Nathan says:

        Take me for example. I don’t believe that the 20th century warming was mostly caused by human activity and I don’t believe the US stimulus lowered unemployment. I’m well aware that the majority of expert opinion is against me on both of these issues. But those answers get marked as factually wrong and therefore I assume contribute to a measured bias to the right. This means to some extent the test is just saying that right wing opinions are inherently biased.

        I mean I suppose you could argue that if I was truly unbiased then my non-mainstream opinions should be equally distributed towards the left and the right, but I don’t think that’s exactly what the test is attempting to do. I like the idea behind it, but the authors are too willing to substitute academic consensus for objective truth.

        • Nita says:

          the authors are too willing to substitute academic consensus for objective truth

          Where are they supposed to get “objective truth”?

          • Lupis42 says:

            Where possible, stick to questions that have relatively well defined objective answers. For example: “Did US unemployment follow this curve, that curve, or neither.” Where you must rely on consensus, specify “is the academic consensus…”.

          • Nita says:

            Using only questions with objectively known answers wouldn’t allow you to detect a bias in judgement. You would flag the people who have their facts wrong, but not the people who consistently interpret the facts in a way that suggests motivated reasoning.

            (Of course, the “bias” in judgement may also be due to correct, yet controversial priors — but if we knew whose priors are correct, there would be no controversy in the first place.)

          • Nathan says:

            @ Nita: But that is not what the test purports to do! Its whole premise that people are more ready to believe facts that fit their ideological preconceptions. Questions like “What proportion of the US budget is spent on food stamps?” are ideal for this – hardly anyone knows the actual figure so instead they’ll pick the number that “sounds right”. The point is that if the answers that “sound right” are consistently skewed one way that reveals a bias on your part. Everyone has a fuzzy view of the world. The question is to what extent we fill in that fuzziness with ideas that flatter our preconceptions.

            Like I would genuinely enjoy a test like this if it was well made! Like most people I like to think that I am a fair minded and objective thinker. I would really welcome some half decent way to test whether that’s true or just a vain self image. This test has the beginnings of that but isn’t there yet.

          • Lupis42 says:


            If there is no objective truth, it is equally likely that the test takers authors are biased, and the test is not a test of bias, but of bias that they don’t share.

            When there is no known objectively correct answer, it is not possible to measure the gap between any position and the objectively correct answer, and it becomes a test of how much of their biases I share.

        • Airgap says:

          You probably misunderstood John. If you disagree with the scientific consensus 50% of the time when disagreement is a right-wing position, and 50% of the time when it’s a left-wing position, you might be a sharp contrarian or you might be a moron crank, but you definitely aren’t a blind partisan. But if you always disagree with the scientific consensus when it supports the right, and never when it supports the left, it’s far more likely that you’re a unthinking leftist partisan than that you’ve examined all the questions carefully, and only in those cases does the consensus appear wrong to you.

          You could validate the test by giving more sophisticated “How much do you actually know about global warming/the stimulus?” exams to those marked as biased, but I’m willing to bet on average it’s pretty reliable.

          • ” But if you always disagree with the scientific consensus when it supports the right, and never when it supports the left, it’s far more likely that you’re a unthinking leftist partisan than that you’ve examined all the questions carefully, and only in those cases does the consensus appear wrong to you.”

            The logic of that assumes that what the test defines as the scientific consensus is itself unbiased. Large parts of the academic world have a strong left-wing bias. That might mean that if the true probability of a claim is 60%, it gets classified as consensus if the left likes it, not if the left doesn’t. If such a pattern exists, an unbiased and well informed test taker will appear biased towards the right.

            I haven’t read or taken the test, so don’t know if that describes the actual situation or not.

          • Lupis42 says:

            I haven’t read or taken the test, so don’t know if that describes the actual situation or not.

            It does on economics (more precisely, it’s biased towards MIT). If Arnold Kling is correct about Fischer’s dominance over modern macro (I assume he is, but have no way of checking), that bias is pretty dominant by # of economists, and so might not be reflective of bias on the part of the test makers, but it’s still bias, and it doesn’t reflect reality to any significant degree.

        • Limi says:

          It is exactly what the test is trying to do. Notice that you didn’t have a problem with the right leaning questions, like on gun control or capital punishment., only the left. Because of your bias.

          Incidentally, and because I want everyone here to know how super clever I am, I got two questions wrong and 0 bias. It probably (read:definitely – I didn’t go out of my way to answer against what I thought, but it obviously had an effect.) skewed the results that I knew they were looking for bias, but that just makes me wonder how so many people in these comments can have had trouble with the questions.

  19. Alex Trouble says:

    I’m not a huge fan of the vox quiz. Some of their questions are disputable, and others give vague answers.

    • Phil says:

      Yeah. Eg: the ‘does organic farming give greater or lessor yields than conventional farming’ question. Are we talking yield per acre, yield per $? These are not the same thing! Organic farming probably yields less per $ and yields less per acre, but the question should still be precise about it.

      (IIRC, intensive multi-crop agriculture on the same land {permaculture to it’s friends} has been shown to be more productive per acre than conventional agriculture, but IIRC it’s more labour intensive, so costs more in the west, plus we don’t have the institutional knowledge to support it.)

      • keranih says:

        I actually disagee that this question was badly constructed, or that it is under any serious debate.

        Of course, it does matter how one defines “organic”, but by the USDA standards, organic production requires more land input, more labor input, and/or more time (and generally all three) in order to produce the same end product (pound of lettuce, gallon of wine, pound of beef, etc.)

        All three things cost money. Where organic might potentually save money is reduced inputs such as imported (to the farm) fertilizers, pesticides and herbicides, improved crop genetics, etc. In cases where the organic farm is using these types of inputs, they are generally of far less effective quality, so the market price for them is lower than for the conventional inputs which are banned from organic.

        The other way that organic methods yield less is in the degree of loss of the crop. Whether it is rotten apples or dead sheep, organic systems tend to have a much higher loss rate, which accounts for a great deal of the increased price, because so much of the crop (which already cost so much more in terms of land, labor and time to produce) doesn’t make it to market.

        RE: permaculture – You should check out 40,000 Years of Farming, if you have not already. The short version is that permaculture is a mature version of food production where a shortage of land is met with massive labor investments. In the US, we never had that land shortage, and our labor pool was never as concentrated, and we were able to replace the human labor with mechanical and chemical innovation, hence permitting the NBA, NFL, Hollywood, Wall Street, Miami Beach and slum kids wearing $100 sneakers to exist.

        (There are other tradeoffs in permaculture besides labor, but I suspect I am already growing tedious.)

    • brad says:

      I think it is perlucidly clear from the responses here that the quiz is not a good match for a population of knowledgeable pedants. Ideally you’d get many of the questions wrong so that most of your answers could contribute to the measurement of bias.

  20. blacktrance says:

    From the Manchu decadence post:

    But what this argument ignores is that these atheists, and the wider secularizing society as a whole, were nevertheless raised with the values of a Christian society. And despite their de-conversion, they still mostly behave according to these tenets, valuing honesty and courage and humility and chivalry. Sometimes these values are dressed up in new names – instead of saying chivalry they might say “being an ally.” But once you know where to look, the underlying values are unmistakable.

    But like the second-generation Asian immigrants, even while they’re enacting all the object-level virtues, they can’t quite articulate why they should care about them.

    This passes as “thought-provoking and worthwhile”? Has Harold Lee talked to any atheists?

    • Ghatanathoah says:

      I’m often astounded at how badly theists fail the Ideological Turing Test. I don’t know if it’s a bias on my part that they seem to me to fail it worse and more frequently than many other groups. I remember that some theist blog performed the test and a lot of theists passed, but it seems to me that any person, theist or otherwise, who reads blogs talking about ideological Turing Tests would be way above average at passing them.

    • nil says:

      I’m an atheist (Christian parents, but was never theist myself) and I’m right there with him. It’s not like it’s impossible to cobble together reasonable solid grounds for your first principles from a purely atheistic perspective, and you can certainly find people who have done the necessary legwork (especially here!)… but most of the athiests I know act moral just because it’s good to be good (with “good” being defined in a broadly liberal Protestant way). Works just fine for them, but still seems a little baseless on a long-term cultural level.

      I’m curious what about it seems so obviously wrong to you?

      • Samuel Skinner says:

        Because “values of a Christian society” are suspiciously similar to “values of all human societies”. Unless he is claiming specifically Christian values, “don’t murder, commit crimes, cheat, steal” are pretty universal norms (well at least bourgeoisie norms).

        “. It’s not like it’s impossible to cobble together reasonable solid grounds for your first principles”

        Nope. Human morality is inconsistent. See wireheading. Human moral intuitions break down once they stray from prosocial behavior.

        • nil says:

          Certainly there are some basic moral rules that everyone follows, and everyone will will likely continue to follow. I don’t expect Lee believes third-generation atheists will start murdering people just because. But there are more subtle questions that show much more variation over history and society. The big ones are the ones that map roughly onto a spectrum of left/right, sympathy for weak/respect for strong, gift economy/property economy, slave morality/master morality, anti-Molochism/pro-individualism. From my point of view, Christianity’s teachings (at least the NT) push strongly towards the former. Doesn’t mean Christians can’t endorse the latter, but it requires either distortion (e.g. properity gospel) or work (see, e.g., conservative Christian traditionalists struggling with immigration).

          Now if the first poster were to have said that those are all solved by innate moral intuition, material situations, or gut bacteria, I’d understand but disagree (and probably wouldn’t have posted). If he/she said that it all boils down to values taught by parents and society without any meaningful referent to philosophical underpinnings, I’d understand and be inclined to agree–after all, there’s a great deal of the above diversity within Christian societies/history, and I’m inclined towards a view of human nature as “extremely plastic meme machines” rather than a bunch of ethics professors. But “has he even talked to atheists” struck me initially as either just another restatement of “but contemporary atheists ARE ethical!” or as conflating the atheist population he/she knows in the “rationalist” community (which has spent a lot of time thinking about the basis for moral stances) with the wider population of atheists (which, again, in my experience are mostly coasting).

          • Samuel Skinner says:

            ” From my point of view, Christianity’s teachings (at least the NT) push strongly towards the former. ”

            The Catholic Church has been pretty outspoken against birth control. There does not appear to be a significant difference between Catholics and others in the use of birth control in developed countries.

      • blacktrance says:

        most of the athiests I know act moral just because it’s good to be good (with “good” being defined in a broadly liberal Protestant way)

        I could say the same for most Christians. They haven’t arrived at their values by thoroughly studying theology and ethics in general and the Bible in particular, but just follow their moral intuitions. It’s baseless, but that doesn’t stop it from perpetuating itself.

        A lot of atheists (not just in the rationalist community) can give a justification for their values that’s at least as solid as that of a typical Christian. It won’t necessarily be well-grounded, but that hasn’t been a problem for Christianity, so why should it be one for atheism?

        • Vaniver says:

          It won’t necessarily be well-grounded, but that hasn’t been a problem for Christianity, so why should it be one for atheism?

          Christians, once a week, bring themselves and their children to a building where they’re publicly lectured on the importance of morality, and what morality means.

          What is the atheist equivalent?

          • blacktrance says:

            Many self-identified Christians rarely go to church, and even for those who go often, church attendance doesn’t cause them to have well-grounded moral views. But in general, the whole discussion is a strange one – there’s nothing uniquely Christian about honesty or courage being virtues (Aristotle was writing about them before Christ’s birth), and it doesn’t take much reasoning to figure out that those things are good.

            Regardless, Lee isn’t talking about church attendance in his post, he’s contrasting how well-grounded the typical Christian’s or atheist’s moral views are. This is a separate question from whether Christians are better at perpetuating their views despite most of them not having a through understanding of the grounding.

          • chaosmage says:

            What isn’t?

            Moral lessons are absolutely ubiquitous. You can’t read any newspaper, read any piece of fiction, watch any TV (narrative or news) or consume any web content that’s more complex than porn, without picking up moral judgements. They’ll usually be implicit: what’s good is what the good guys do, what’s bad is what the bad guys do. But they’re explicit often enough, if only in valuing behavior as “shitty” or “nice”.

            And in anybody who isn’t highly unusually averse to narrative media, they easily add up to more than an hour per week.

            These lessons aren’t very complex, but I heard a bunch of Christian sermons and I don’t remember any better from them. These sermons give simple lessons like “be honest” or “imagine what your behavior looks like from outside” and that’s enough. Turns out that sophistication and consistency and rigor are non-essential to morality for most people’s everyday purposes, as long as there’s a “when in doubt, ask an expert” clause somewhere in your moral ruleset. And atheists don’t lack that.

        • notes says:

          The argument is, I think, that the existence of those who did derive their values through thorough and well-grounded study acted as an anchor for the broader group following their moral intuitions.

          Remove the anchor (or decrease its weight), set a new anchor over on atheism, and the community will drift into a new position.

          Some see it as a feature; the author of that piece sees it as a (potentially dangerous) unknown.

          • Jaskologist says:

            The argument is that atheism doesn’t have an anchor at all. They don’t drift into a new position, they just keep drifting.

          • Randy M says:

            Atheism lacking moral anchor? What you need, friend, is new and improved Atheism+ ™!

          • Nornagest says:

            Atheism doesn’t need a moral anchor, because atheism is not a totalizing worldview. (Atheism+ tried to be, but… yeah.) Individual atheists need a moral anchor, and by and large they seem to have little trouble finding one.

            (Sometimes it’s a bad one. But that’s a problem that theists have too, though the variance is arguably lower there as most versions of theism have evolved reactions against apostacy, heresy, etc.)

    • Scott Alexander says:

      I believe Harold Lee is himself an atheist.

  21. Stezinech says:

    On the Vox quiz, I’m amused that the average score is 6, which is equalivalent to random guessing (given the that there were three options). That’s a tough quiz!

    Also, I intially felt that the quiz contained more correct Liberal answers than Conservative ones. But I totalled it up and it comes out to 8 Liberal, 8 Conservative, 2 Neutral, which makes more sense.

  22. anon says:

    Regarding bisexuality and eating disorders.

    There’s a huge problem with the label “bisexual”.

    I suspect that women who are promiscuous or polyamorous are much more likely to define themselves “bisexual”.

    Women who are effectively bisexual, but don’t have these traits, are less likely to call themselves bisexual, they will call themselves straight or lesbian.

    The first reason for this is that a woman who has had a smaller number of longer lasting partners is less likely to even discover that she can be with both sexes.
    The second reason is that in my experience, a woman who used to be committed to a man and now is committed to a woman, doesn’t say she is bisexual, instead she tends to say that she believed she was straight, but she discovered herself to be really “lesbian”, and lesbian is what she is. Whereas a woman who leaves a woman for a man, will say exactly the opposite.
    This is probably because if she called herself “bisexual” while in a very serious monogamous relationship, she’d risk being perceived as a lacking committment or faithfulness.

    A woman who is polyamorous and/or promiscuous, is thus much more likely to call herself “bisexual”, and this may explain the pattern.

  23. US says:

    As for Syrians, I’d assume that if the Haitian experience is relevant to the United States the Danish experience might be as well. So below a few other data points.

    The employment rate of Syrians living in Denmark is the lowest of that of any country included in our statistics. 22,8% of Syrians in Denmark are employed. In 2011, there were 2440 Syrians living in Denmark, and 1743 descendants, so this is not a small sample problem. An analysis including data on approximately 2000 Syrians arriving in Denmark between 2009 and 2013 showed that only 13% of Syrians were employed after having stayed here for three years. If you think they’ll do better later on, keep in mind both that that 22,8% employment rate above is based also on individuals who have already been here for many years and for example the related observation that immigrants from Iraq, Lebanon and Somalia who had been in Denmark for at least a decade a few years ago were found to all have employment rates between 30% and 41%.

    Among Syrian refugees arriving in Denmark in the time period from 1999-2006, 17 percent (self-)reported having a tertiary education and 5% reported having what in the literature is termed ‘an education leading to a vocational/professional qualification’ (e.g. plumbing); corresponding numbers from other refugee countries are 16% and 10%. The people reporting on those numbers made the in light of the data to me sensible conclusion that Syrian refugees are unlikely to be better educated than other refugee populations. As a side-note, whether you look at Syrians who arrived before 2006 or at those who arrived later, there are fewer Syrians who finish a tertiary education in Denmark than is the case for non-Syrian refugees.

    I recently wrote a long data post on these and related topics on my blog, which is linked to above my name – there are sources, links and further comments there. I’ll probably delete the post later, but I might wait another week or so before I do that.

    • Scott Alexander says:

      It’s a good post. I won’t link to it if you’re going to delete it (why?) but if you don’t mind I may steal some of it in case I ever feel the need to write about a similar topic.

      • US says:

        I know Gwern saved some of the observations in the post on his evernote account and like him you’re of course very welcome to save the data/talk about the observations/etc. as much as you like.


        I generally don’t blog politics or politics-related topics on my blog, so one reason is simply that it probably doesn’t really fit in well with the rest of the content, or for that matter fit in well with the sort of blog I want to have (I’m quite apolitical in general and as a general rule I don’t really like to (‘waste time’) talk(ing) about politics-related topics).

        Another reason is that I live in Europe, and in Europe radical muslims have started systematically killing people for simply drawing cartoons they don’t like – Denmark already has a rich history of murder attempts on cartoonists and authors critical of islam (google e.g. Kurt Westergaard or Lars Hedegaard). Over time I’m sure those people will come up with other reasons for killing people they consider to be their opponents, and once they start emulating the Bangladeshi radicals murdering bloggers left and right I do not want to be a target. I’m aware that deleting the post will not make it disappear completely, but it lowers risk to a more acceptable level (the risk is likely very low, but it’s also hard to estimate/quantify and I’m *very* risk averse).

        Actually in light of you potentially linking to the post I find myself more than a little conflicted now, because I’m often really annoyed with people who obviously don’t already know the sort of things I talk about in the post, and a link from you would lead to *a lot* more people having awareness of the existence of this data than is currently the case. Can I think about this a bit? Either way you’re of course as already mentioned more than welcome to share the observations/save the data/ etc. – you can ‘steal’ whatever you like..

    • Sastan says:

      There are some massive differences between groups that on the surface wouldn’t seem to be that different. Middle eastern Muslim immigrants in the US tend to do quite well, and assimilate. The more recent raft of Somali refugees, not so much. Europe seems to have a lot of problems with theirs.

      I’m sensing the outlines of a theory, but I don’t have much to go on. I suspect it has something to do with the interaction of the dominant ideologies of the host and home nations at the time of emigration, as well as the social status of the immigrants and difficulty of immigration. Refugees will always be a more mixed bag than economic migrants, if the trip is long enough.

      Needless to say, it’s complicated, and I don’t have a firm enough grasp on it to articulate better than that. Perhaps after coffee!

  24. Douglas Knight says:

    The Vox article on homicide starts by claiming that the timing doesn’t match the deaths of Michael Brown and Freddie Gray, but then it proceeds to give a couple of theories to explain that match in timing.

  25. Izaak Weiss says:

    If you thought base phi was cool, check out negabinary

    Also, any utilitarians out there who have read Strong Female Protagonist? How do you feel about Feral’s choice? How would you feel about it if she were forced into it rather than freely choosing it?

    • Saint_Fiasco says:

      I read that story arc a long time ago, but from what I remember, it seemed to me like they were being too careless with the goose that lays the golden eggs.

      Feral shouldn’t be donating organs directly to sick people. She should donate tissues for medical research instead.

      As for forcing her to do stuff, I think that would run the risk of turning her into a more problematic super villain. Better to just use the tissues she loses during battle or something like that.

  26. gwern says:

    the Manchus’ self-conscious attempts to avoid decadence in Imperial China and how it relates to modern immigration.

    Or it’s just regression to the original population’s mean and intermixing. You might as well ask why second-generation immigrants from Africa do not do as remarkably well as the first-generation, or the third generation as well as the second generation.

  27. moridinamael says:

    1) Dogs embody all qualities of human goodness. We should create dog-ems before we create human-ems, make the dog-em superintelligent and just let it decide what to do with us.

    2) What is the best way to publicly stand behind corporal punishment as a superior alternative to the current prison system? How does one avoid looking like a horrible person?

    3) Re: the op-ed on Academia. As a PhD holder, I agree with the author’s assertion that the first and foremost problem is too many people. You can almost derive all the other problems from this fact. For a while now I’ve been of the opinion that you shouldn’t even pursue a PhD unless you come from a wealthy family. I don’t mean “should” in any moral sense, I just mean that it’s a bad idea and you’re going to get a bad outcome unless you’re just doing it, essentially, for fun.

    4) I scored a 0% bias. But I sort of mis-clicked on two of my answers. But only in a way that should have made me seem more biased. So I guess I’m less than 0% biased.

    • Jaskologist says:

      1) Dogs embody all qualities of human goodness. We should create dog-ems before we create human-ems, make the dog-em superintelligent and just let it decide what to do with us.

      Give the man a MIRI fellowship, he’s just solved unfriendly AI.

    • Randy M says:

      “2) What is the best way to publicly stand behind corporal punishment as a superior alternative to the current prison system? How does one avoid looking like a horrible person?”

      I agree with this position. I’d say you should commit a crime then request corporal punishment in leiu of prison, but who would trust a felon? 😉
      In seriousness, I think you have to get people to see that they themselves would probably prefer temporary pain to long term incarceration. Except if that is true, is it a deterrant? It seems to me both more humane and more of a deterrant to those accustomed to a prison lifestyle, but this looks like a paradox, or at least counter intuitive.
      It would be a heck of a lot cheaper, at least, though of course it would only work with a (smaller) penal system for violent repeat offenders.

      • Psmith says:

        The effectiveness of a given punishment has more to do with its speed (how long after offense?) and its certainty (will you get punished?) than with how nasty it is. Mark Kleiman talks about this in his “When Brute Force Fails.”

        • Randy M says:

          Perhaps, but “how nasty is it is” certainly a factor. If getting caught stealing means I might be beat to death some months later, versus I will certainly get an angry later the next day, I’m more likely to consider it in the latter scenario.

          And everyone agrees swift, certain justice is preferable in any event.

        • “The effectiveness of a given punishment has more to do with … its certainty (will you get punished?) than with how nasty it is.”

          I don’t know what data you are working from, but the evidence I saw for this claim long ago struck me as wrong. Doubling the fine had less effect than doubling the probability of receiving it. But the size of the punishment isn’t the fine, it’s the fine plus reputational costs of being convicted plus legal costs plus time spent going through the legal process. Call all of those X. X+2F is not twice X+F.

          • John Schilling says:

            And in some subcultures, the reputational cost of taking a state-sanctioned caning and being seen out on the street the next day doing the same thing you got caned for is likely negative.

        • Psmith says:

          Randy: Well, the theory is that corporal punishment is easier to impose swiftly and certainly than jail terms. If corporal punishment is less severe (by revealed preference) but swifter and more certain than jail, it might still be an effective deterrent. Jaskologist and Esquire are getting at this downthread, I think. (Surely there’s scope for a natural experiment here–Singapore vs Hong Kong, or something.).

          David: now you mention it, I’m not seeing a whole lot of empirical work to support the claim that swiftness and certainty > severity, although I may need to comb Kleiman’s bibliography more carefully.

          John: good point. On the other hand…I think we might be underestimating the actual unpleasantness, and not just the physical unpleasantness, of (say) caning as used in Singapore and Malaysia. See, e.g. (Returning to the first hand again, this might be an instance of Kleiman’s point about how a quiet, peaceful, boring prison might be nastier than the status quo for the sort of people who end up in prison, even though that wouldn’t be true for the sort of people who set prison policy or read Kleiman’s books.).

          • John Schilling says:

            Why are we counting corporal punishment as “swifter” than a prison term?

            If it’s that the punishment is over sooner, that weighs against it as a deterrent. Getting It Over With Quickly is almost universally agreed to be the best strategy for dealing with any transient unpleasantness; presumably criminals are no different in preferring A: not to wait around interminably wondering or fearing what is going to happen to them, and B: to get it over with and get back to their lives, whatever “it” is.

            If the idea is that the punishment is going to be initiated sooner, how does that work? Are we dispensing with trial by jury and the rest of due process? That is particularly perverse, given the irreversibility of corporal punishment.

          • Anthony says:

            If the idea is that the punishment is going to be initiated sooner, how does that work? Are we dispensing with trial by jury and the rest of due process?

            Plea bargaining. “If you plead guilty, you can choose the caning, and be back home tomorrow. If you go to trial, you’ll be in jail for at least a week before the trial even begins, and might be in jail for months or years.”

          • Saint_Fiasco says:

            Are we dispensing with trial by jury and the rest of due process? That is particularly perverse, given the irreversibility of corporal punishment

            Corporal punishment is not irreversible unless it’s really harsh and you end up maimed or something.

            Normal prison, on the other hand, can have more lasting consequences. You lose your job, your relationships with friends and family are irreparably strained. If you work on something that requires constant re-education you will be way behind when you are released, etc.

          • Who wouldn't want to be Anonymous says:


            1. Plea bargaining (in the US, at least, but maybe everywhere) is already completely, utterly, laughably broken.

            2. I don’t have much practical experience, but I was under the impression that you had to enter your plea bargain in court. Which means you need to wait for (and appear at) your court date either way. And if you are getting the sort of corporal punishment that only takes one day to administer (as opposed to spread out over days or weeks to avoid killing you) you’re probably looking at either time served or a pretty short stay after sentencing.

            I am kind of ball parking, though, because I don’t have a good sense of the stroke-to-days in jail equivalency is.

          • John Schilling says:

            You can’t simultaneously claim that corporal punishment is going to prevent people from engaging in lucrative or otherwise rewarding crimes years down the road, and that it has no enduring effects.

            You can’t simultaneously claim that it’s no big deal to punish people without a trial because you offered them a plea bargain, and that it’s OK to threaten them with decades in prison because you offered them an alternative that is no big deal.

            And you damn well can’t claim that the powerful, permanent, and fundamentally unjust effects of applying this sort of punishment to innocent people don’t really matter because the punishment is just an owie that they’ll get over when they realize you didn’t mean it.

            You all are in fact proposing judicial torture because you claim it will be less harmful than prison. And our prison system is dysfunctional enough that you may be right. But when you start advertising as benefits of the new system the fact that we don’t have to waste time with any of that pesky “due process” stuff, and that it isn’t so bad that we will occasionally torture innocent people because it won’t leave ugly scars for the TV cameras, then as far as I’m concerned you just volunteered to be the first ones with telephone wires taped to your genitals.

          • Saint_Fiasco says:

            One problem with the consequences of prison is that they are long term. It’s hard to be deterred by a something bad that will happen to you a few years down the line.

            Not being able to find a job after you are released from prison doesn’t deter anyone, even though it sucks. The more immediate pains of being in prison do deter some people. Can’t we have the pain without the expensive life-destroying long-term damage?

            I agree there should still be due process. Why wouldn’t there be? Are you worried that if punishments aren’t severe enough, people may plead guilty of anything just to get the process over with?

          • moridinamael says:

            @John Schilling and whoever else

            I don’t endorse this scenario where you get caned if you plead guilty but you got to jail if you go to trial. That seems like some really perverse incentives would arise.

            Actually, I pretty much think that the crimes for which corporal punishment would be a realistic punishment are limited to nonviolent crimes, such as drug possession and most forms of theft except mugging. These crimes, if they even go to trial, tend to have short trials anyway.

            Additionally, I can easily imagine a scenario where once the sentence is passed, the convict gets to sit in a cell for something like a week just waiting for the caning to be carried out. That would nicely remove any psychological benefit to “getting it over quickly.”

            So, as a rambling aside, I have deep philosophical problems with the whole Western conception of criminal justice. I think, in fact, that the Western conception of criminal justice is itself confused about what it’s trying to do. It sometimes aims for “justice” but then ends up actually exacting something more like “revenge,” and then other times it seems to have no interest at all in justice and is really just trying to protect society-at-large from criminals by locking them up, and then at still other times it makes noises about rehabilitation but it never credibly approaches this ideal.

            I mention this because corporal punishment seems terrible to me, but it seems orders of magnitude less terrible than the monstrosity we have allowed to exist.

          • John Schilling says:

            Can’t we have the pain without the expensive life-destroying long-term damage?

            No, we cannot. Perhaps you ought to consult with survivors of torture on this. The pain, the memories of the pain, the psychological consequences of the pain, that is the life-destroying long-term damage. If it doesn’t have long-term effects, it won’t work as a deterrent to serious crimes.

            I agree there should still be due process. Why wouldn’t there be?

            Because people are advertising as one of the virtues of this proposal, that it gets things over with quickly. There won’t be time for due process. Because the claim is that people will volunteer for the torture through the plea-bargaining process, the whole point of which is to circumvent all that tedious “due process” stuff.

            Are you worried that if punishments aren’t severe enough, people may plead guilty of anything just to get the process over with

            I am worried that people will be offered the choice between two punishments, both of which are too severe, and that this will be rationalized on the grounds that “Hey, we offered you a choice; if you can’t stop bitching and whining about it, obviously you made the wrong choice, but that’s your fault so you might as well shut up because we’re not listening”. Whichever choice they make.

            There’s too much of that in the criminal “justice” system already. Branding judicial torture as the Happy Fun Punishment that modifies behavior in a socially beneficial manner with no real harm, is going to make it worse.

          • houseboatonstyx says:

            @ John Schilling
            The pain, the memories of the pain, the psychological consequences of the pain, that is the life-destroying long-term damage.

            How severe (or long, or often repeated) a caning are we talking about here? How many, if any, British statesmen and other respected figures of the late 20th Century had not been caned in their school days?

            Of course if they did all get long-term psychological damage, that could explain a lot of things.

          • John Schilling says:

            We are explicitly talking about “corporal punishment” sufficiently severe to substitute for the years-long prison sentences handed out for felony crimes, to dissuade people from performing extremely lucrative or rewarding acts many years in the future, to deter even people who haven’t even been punished and will evaluate the threat of such punishment with the rose-colored glasses of ego and machismo. So no, it’s not twelve strokes with the headmasters’ cane, or even thirty-nine lashes with the cat.

            And we are implicitly talking about punishment inflicted in at most a few days. So it’s not going to be fifty lashes every week while spending the next year in prison, as is sometimes the case in e.g. Saudi Arabia.

            Really, it’s not going to be strokes or lashes or anything of the sort. Historically, the sort of corporal punishment that works for felonies involves mutilation. To avoid the piles of severed hands that really won’t do in the age of youtube, but get the required immediate but enduring deterrent effect, we are going to need to break out the electrodes and waterboards and so forth.

          • suntzuanime says:

            Youtube would just make the deterrent effect more powerful. This sort of punishment doesn’t work if you do it in secret, it’s all about instilling fear in the hearts of the citizenry.

      • Have you read Thinking Fast and Slow? I expect people would remember corporal punishment as being worse even if they suffer much less in total. And the punishment would occur closer in time to the time. So overall I suspect it would be more effective despite the fact that I’d also prefer it to prison.

      • Sastan says:

        Unfortunately, we have a very, very large population of violent repeat offenders. Even Ta-Nehisi Coates is acknowledging this in a recent interview I heard. Nonviolent drug offenses aren’t going to get us out of mass incarceration, because most of the people doing hard time are doing it for very good reasons.

        • Pku says:

          Object-level counterarguments aside, if the high prison US population is due to repeat violent criminals, why do you think the US has so many more repeat violent offenders than any other country? (enough so that it has the high crime rate *and* the ridiculously high incarceration rate?)

          • FacelessCraven says:

            Becuase we’re a warlike people from a warlike lineage, the descendents of rebels and vagabonds since the founding of the country?

          • Sastan says:

            Not to put too fine a point on things, but there’s a reason it was Coates talking about it. Black males are far, far more likely than other groups to be violent criminals. I really don’t much care about the argument of culture vs. racism vs. genetics. This isn’t intended to start a long argument that would be better suited to Steve’s site.

            The US has far more black people than the western european countries we are so often compared to. If you break out crime by race, American whites are broadly comparable to the middling to high crime end of Europe. The way we get to four times the murder rate of places like Britain is that their population is 1% black and ours is 12% black. And blacks in America commit violent felonies at something like six or seven times that of their white counterparts. They are also much more likely to be victims.

          • Nornagest says:

            If in one population you have 88% robbing one bank a week and 12% robbing seven banks a week, and in another you have 99% robbing one bank a week and 1% robbing seven banks a week, you end up with 1.7 times the numbers of banks per capita getting robbed in the first, not four times.

          • Pku says:

            But even comparing white americans to white europeans, we still get more than twice as many prisoners per 100,000 (~320 vs. 150) in the US as in england, so there’s clearly a large part of the US prison population issue that has nothing to do with race.

          • Anthony says:

            Pku (responding to your bottom-level reply) – the homicide-commission rate among whites in the U.S. is right at the high end of European homicide rates, and about 3 times as high as the most peaceful European countries. (2.4 per 100,000 for U.S. whites, while Europe ranges from about 0.7 to 2.7, with most between 1.0 and 1.6.) Assuming that murder is a reasonable proxy for all sorts of violent crimes, U.S. whites are still more violent than European whites – maybe twice as violent. (That assumption isn’t very safe, as apparently criminals in the UK are generally more violent but also much less likely to kill their victims. But reporting for non-homicide violent crimes varies a lot between countries, so it’s hard to make useful comparisons.)

            Therefore one would expect the white incarceration rate to be higher in the U.S., even given the same criminal-justice system. But that’s a false assumption – American prison sentences are longer, and more often applied, than in (almost all?) European countries. Two equally violent populations would have higher incarceration rates under the U.S. criminal justice system than under the UK, or France, or Germany’s.

            Hm – you show data indicating that U.S. whites are a little more than twice as likely to be in prison than in “England” (England, or UK?), and the UK homicide rate is 1.0, compared to U.S. whites of 2.4, so that explains more of the variation than I’d have expected.

          • Pku says:

            I was quoting England (wikipedia’s list of countries by incarceration rates separates the UK into the component countries for some reason).

    • Nornagest says:

      What is the best way to publicly stand behind corporal punishment as a superior alternative to the current prison system? How does one avoid looking like a horrible person?

      You can’t avoid looking like a horrible person to a lot of people, because this is a sacred value for a lot of people. There is no reliable way to reason people out of their sacred values: you can only hope to build a critical mass that doesn’t care if it looks horrible. Sometimes you can play sneaky rhetorical tricks but I don’t think you’ll find that easy in this case.

      • Esquire says:

        I would try making it fight against the sacred value of keeping families together. A major advantage for corporal punishment for society is that kids (and moms) get to have dads around.

    • Adam says:

      Provided you don’t mean permanent dismemberment, I’d take it. Who wouldn’t? Get caught and get beat to shit once in public sounds a lot better to me than get caught, get locked away for ten years and beat to shit by prison gangs and corrections officers.

      • John Schilling says:

        Which means it has less of a deterrent effect, in addition to not materially preventing the offender from committing crimes. And I’m not seeing any rehabilitative effect beyond the limited deterrent. So, as a means of preventing or controlling serious crime, this seems likely to come up short.

        Which is I suspect why the cultures that were big on corporal punishment, did go for mutilation or the like as punishment for felonies.

        • Adam says:

          Speaking just for me, it would have zero deterrent effect. Prison at least has some, since I’m empirically not violating laws I don’t actually agree with, like it being illegal to organize all the street walkers out there, offer them protection and sell them cocaine, and skim their profits. I’d gladly do that if not for the prospect of going to prison.

          • moridinamael says:

            You would really gladly do that if it meant you would be caned? Asking seriously, this is an interesting data point for me.

            If your answer is yes, then have you considered that after being caned once, your opinion might change? Like, are you potentially underestimating how much you might not want to be caned?

            Empirically, lashing/caning has been used effectively as a deterrent in other times and places. That is why I doubt your claim of “zero” effect.

            ETA: There is also an argument to be made for “three strikes” (etc.) policies in situations where corporal punishment is employed. If caning really doesn’t seem to be getting the message across for certain people, then those people get to go to jail.

          • Adam says:

            It’s possible I’m underestimating how bad it is to get caned, but I’m thinking right now that the threat of another pimp killing me would be the bigger fear. If you just straight up offered me the payday and all I had to do in return is get caned, I can’t imagine not accepting that. I might be giving myself too much credit in figuring all the years of military service dulled my reaction to temporary pain while leaving me still very afraid of prison.

          • Sastan says:

            But it’s not getting caned once! If you have an ongoing pimping enterprise, it’s every time you get caught, until the judge decides you’re a congenital shithead and locks you up, in which case you get ALL the bad outcomes!

            The bad bit about caning doesn’t seem to be the pain, it seems to me to be the humiliation of having this done in public. I remember when that dipshit American tagged a bunch of cars in Singapore and the country ground to a halt for several months and expended more energy than we ever did on hostages trying to get them to give him jail instead of four swats with a bamboo stick! So apparently, there’s a lot more deterrent effect than you are granting.

        • Randy M says:

          It may not be as much of a deterrent, but compare that with the effects of prison life on recidivism, being basically college for the criminal career track. Training, networking, room & board, etc.

        • Jaskologist says:

          Part of the potential benefit is breaking up prison culture, such that people who have committed relatively minor crimes don’t get put away with harder criminals and become more like their new peer group.

          The other part of the theory is that criminals are dumb, and more likely to be deterred by a quick punishment than the distant prospect of a worse one.

          • Sastan says:

            If I read the literature correctly, the harshness of the punishment isn’t well correlated with deterrence, but the likelihood of any punishment is.

            Obviously, you can’t fine people a gummy bear and hope to get deterrence, but according to my best read of criminology, we should expend maximum effort to solve crimes and try people quickly, even if that means reducing the level of punishment. Better a smack in the mouth when the kid does something than a spanking four hours later when dad gets home.

        • Esquire says:

          I think if you are a low-IQ hyperbolic discounter it could easily have a much stronger effect.

          To use fancy neuroscience for a second: pain and especially humiliation activate your lizard brain negutil centers much more powerfully than rational worry about about years behind bars.

    • Jiro says:

      What is the best way to publicly stand behind corporal punishment as a superior alternative to the current prison system? How does one avoid looking like a horrible person?

      One of the problems with corporal punishment of criminals is that for humans to enjoy inflicting pain is common enough to be noticeable. There is not a similar prevalence of humans who enjoy humanely imprisoning or even painlessly killing others. And you can’t just bar humans who enjoy inflicting pain from being judges and prosecutors. So at a minimum the relevant professions will attract people for whom the profession is a conflict of interest. (And under some ethical systems there are other reasons against it, like varieties of deontological and virtue ethics.)

      • Randy M says:

        1) Would the Stanford prison experiment argue against that?

        2) I’m sure the Mythbusters can invent a humane canning robot, anyway. Tested on over 25 pig carcasses too.

        • Samuel Skinner says:

          “1) Would the Stanford prison experiment argue against that?”

          It isn’t considered a good experiment due to the scientists rather blatant interference and set up.

        • Limi says:

          The Stanford prison experiment was bullshit, it’s basically a primer on how not to do a psychological experiment. I don’t have any other insights into this topic, I am just on a one man mission to get people to stop talking about that experiment unless they are talking about what a wreck Zimbardo was.

          Beaten by miles, that will learn me for not refreshing from last night.

          • Samuel Skinner says:

            The more people who speak up, the more credibility they will have and the faster citations of the SPE will die.

        • Randy M says:

          I know that there are problems with the methodology, but the mere fact that it was *done* implies that there are people who may well enjoy imprisoning others.

    • ivvenalis says:

      The protagonist of Count to a Trillion used a horse-em in order to avoid problems with unfriendly AI.

      • moridinamael says:

        *throws his rough draft of a story with a dog-em protagonist in the garbage*

        • John Schilling says:

          Count to a Trillion was a mildly entertaining wish-fulfillment fantasy; there’s still room for a really good story with the premise 🙂

          Also, I don’t think the horse-em actually showed up until the sequel.

    • DrBeat says:

      1) Dogs embody all qualities of human goodness. We should create dog-ems before we create human-ems, make the dog-em superintelligent and just let it decide what to do with us.

      If you wish to imagine the future, imagine a metal tongue, licking a human face forever.

    • Nita says:

      2.1. So, are you partial to caning, as in Malaysia and Singapore, or Sharia-style flagellation? Or perhaps you can suggest an alternative method?

      2.2. If corporal punishment is more effective, why do you think the countries where it’s practiced still use imprisonment? (E.g., caning of adults is always combined with a prison term, lashing is used for relatively minor offenses, such as drug use or kissing in public.)

      • Psmith says:

        In re your 2.2: I was surprised to learn that, as of 2011, Singapore promised caning for a good deal more than misdemeanors. See

        Edit: “combined with a prison term”–never mind, link isn’t really relevant then.

      • moridinamael says:

        2.1 I haven’t really gone into the details of what method would be best. My original comment originated mostly from a visceral horror at the current prison system, and a subsequent grasping for any kind of alternative.

        2.2 So, there’s going to be some subset of people who will just happily keep on stealing cars or whatever because they know they’re “only” going to get caned. Some other solution would be needed for these repeat offenders. Also, I expect there are some crimes like murder, armed robbery, etc. which seem to suggest that the person committing the crime is not healthy psychologically and won’t be amenable to the same types of punishment as most people. In my perfect world, people who don’t respond to corporal punishment would be given mental healthcare instead of being thrown into prison, but we’re quite a distance away from that reality in the US.

    • Kevin C. says:

      On #2:

      I’ve got a number of arguments on this issue, but as for the one that makes you look less “horrible”, my best one begins by pointing out that all punishment involves inflicting some manner of suffering upon the criminal; that’s what makes it punishment. And while we have the usual social primate instincts to punish (for good evolutionary/game-theoretic reasons), we also on average have countervailing empathy that makes seeing others suffer uncomfortable. But this latter does not treat all forms of suffering equally. Physical pain is more “visible” than psychological suffering, and whatever our intellectual comparison of their severity, we react more strongly on the instinctual and emotional level to the former than the latter. Consider the difficulty getting emotional and psychological abuse taken as seriously as physical abuse. (Or look at many of the other replies, and the visceral horror displayed, the bare assertions that no matter how bad our prison system is, the psychological suffering it inflicts, however severe, cannot ever hope even begin to match even moderate corporal punishment. This despite, for example, the evidence for the severe and long-term harm inflicted by solitary confinement, going back to at least de Beaumont and de Tocqueville’s work.) Thus, the suffering we (the democratic “We the People” we) inflict on our fellow human beings via our penal system is rendered more abstract, less visible and less able to engage our emotions and empathy instincts, and thus more comfortable for us to inflict than we would be were it physical pain.

      Secondly, the suffering we inflict via prison is spread out over long periods of time, reducing the intensity at any moment but not necessarily the total quantity, which becomes hard to measure. Setting aside the future time orientation of the person being punished, consider outside observers such as us. Whatever sense we may be able to obtain of how much prison sucks at any given moment, attempting to integrate that over the time of a decades long sentence is not readily done, particularly at the emotional level that engages our compassion.

      Third, the suffering we presently inflict occurs at a separation, behind prison walls, out of sight from our lives in general, and out of sight, out of mind. Together, these three factors ensure that our current system makes the pain we inflict distant and abstract, and thus easier for us to avoid engaging our empathy. Corporal punishment, particularly if administered in some semi-public fashion, so that some sizable fraction of the electorate will have seen first-hand at least one time the pain inflicted in their name, would force us as a society to truly confront what we do to our fellow human beings to enforce our laws and rules. Consider a politician running her campaign on tougher sentences for X, Y and Z. When “tougher sentences” means more years in prison, many see her as “tough on crime” and “keeping our streets safe”; when “tougher sentences” means more lashes, the perception likely shifts toward “sadist” and “bully”. I may be a cynical misanthrope, but I’m not so misanthropic to believe that the majority of our population has serious sadistic tendencies, thus, with corporal punishment rubbing in our faces the harm we do via the penal system, rather than hiding it away as prolonged psychological harm mostly invisible behind prison walls, I’d expect a lot of legalization, a lot less “tough on crime” rhetoric, a lot more effort toward restitution, a lot more effort on prevention via moral education as preferable to punishment (as reading Confucians like Xunzi has helped persuade me is preferable to the Han Fei Legalist attitude), and generally more tempering of justice with mercy, more engagement of our primate instincts to compassion to put stronger checks on our primate instincts to punish.

  28. Esquire says:

    I really think the Vox article does not do nearly enough to disprove the “Ferguson Effect” hypothesis. It is not really very clear how much crime was rising pre-Ferguson, certainly not clear enough to dismiss the Occam’s razor explanation of what connects Baltimore and St. Louis, given that both cities are major outliers for crime increase.

    Honestly I think the teachings of rationalism are very clear that we should be more careful than this about dismissing a simple first-order explanation that fits very well in a situation where our emotions are giving us very strong urges to rationalize it away.

    • Wrong Species says:

      What I find amazing is that they open up with obviously it’s not the ferguson effect and anyone who thinks so is dumb then they sympathetically portray the idea that it’s because people don’t trust the cops. Yes, they mention that it has the same problem but only in the last paragraph. The framing really bugs me.

  29. Pete says:

    On the Chinese car/medical costs story, my then girlfriend spent 3 months in China in 2007, and she was told about that back then, although the warning for her was specifically related to taxi drivers, so the myth (if it is a myth) has been around for at least that long.

    I was happy to do pretty well on the political bias test. I’m apparently slightly liberally biased but I think some of the questions I got wrong were due to my lack of knowledge of American issues.

    Why not nuke Alberta? Not really anything of value there except the oil (I kid, my wife is Albertan).

    • Pete says:

      I also knew about the cow magnet. On one of my first days of university my tutor handed us all a magnet and asked us what we thought it was for (non of us got it).

      Apparently he’d been involved in an experiment that required as little magnetic field as possible, so they built it in the middle of the English countryside, only to find that the nearby cows were causing problems.

  30. zz says:

    Re: school investment.

    In the paper Scott linked to (Sept 2015), authors “find little evidence that school capital campaigns improve student achievement.”

    Trying to circumvent the paywall, I find a Dec 2014 paper with the same title by the same authors [pdf], which finds “modest increases in student achievement and attendance, primarily among poor students” and “small but measurable impacts on student achievement, high school graduation, and college entry.”

    Also, a February 2015 version (same authors, same title), agrees with the Sept 2015 claim of “little evidence”.

    What gives?

    • Adam says:

      I’m not sure that’s always the point of a capital campaign anyway. I remember the school district passing a bond measure when I was a senior and what they ended up doing was rebuilding both parking lots and all the hallway overhangs. I doubt that had any impact on anyone’s grades, but that doesn’t change the fact that buildings and roads require maintenance every now and again and it was probably good for our cars and it certainly meant students got wet less when it rained.

    • Emily says:

      That’s an “ask the author” question.
      Edit: for the first author, that’s first initial last name at

    • gwern says:

      If you compare the first version with the final Sep 2015 version, you can see that the effect-sizes for reading & math change and become non-statistically-significant; econ papers can change a lot in preprint form which is why you need to read the final version. Presumably someone pointed out a missing covariate or a non-clustered standard error or some technical detail like that… I tried wdiffing the text version, but there are way too many small changes of phrasing and formatting, and so the output was unreadable garbage hiding whatever the technical changes were. The datasets seem to be the same, so that’s probably not it. Reading through them side by side, the analyses do seem to differ: final has section ‘4.2 Event Study Analysis’ where original has ‘C. Campus-level Analysis’. The justification for 4.2 is in a footnote:

      “The power gain afforded by focusing on students actually affected by capital investments comes not only from
      improved precision of the estimates, which has to do with the number of renovations or constructions relative to the
      number of close bond elections. It also relates to the bond election treatment being diffuse relative to renovations or
      constructions, which make effect sized much smaller in the RD analysis. We return to this issue in Sections 6 and 7.”

  31. Tim Martin says:

    I kinda don’t buy the Lehre et al. study (on greater male variability.) It’s not that I think the hypothesis itself is wrong, but that the paper provides pretty weak evidence in favor.

    Ignoring the results for the psychological measures (those I find relatively believable, as they’ve been established by previous research), the authors took a smattering of physical measures and found that most but not all measures had greater male variability than female variability. This provides some evidence in favor of the hypothesis – but we have to believe that these are the right measures to look at (I don’t have enough knowledge of biology to say), and we have to believe that the results as a whole are significantly different from chance.

    (The blood parameter results, in isolation, are pretty darn unconvincing. Of 31 parameters analyzed, they got 11 with no difference, 13 with greater male variability, and 7 with greater female variability. By chance we would expect 10.33 parameters to fall into each group.)

    Strangely, the paper doesn’t mention where the extra variability might come from. It only mentions “genetics” in the intro, saying that evolution might be responsible for the difference. But first we have to keep in mind that not all genetic traits are adaptive. Second, if we assume that the male/female difference IS adaptive, the question remains, how does it work? The authors don’t mention the fact that males only have one X chromosome until the end of the Discussion section, which makes me feel like this was a last minute consideration, rather than a serious motivation for believing their hypothesis was worth testing. A physical mechanism goes a long way toward convincing a person that a hypothesis is legitimate, and up until the *next to last* paragraph of their paper, Lehre et al didn’t have one.

    So I think this is rather weak (though weak evidence is still evidence). And if greater male variability is caused by recessive genes on the X chromosome, then do we know what traits some of those genes are linked to, so we can test those specifically? Again my paucity of bio knowledge is getting in the way here. Someone please help out!

    • Nita says:

      if greater male variability is caused by recessive genes on the X chromosome

      I think the best way to test this would be to compare male and female variability in several XX/XY and ZZ/ZW species — is the heterogametic sex always more varied, or are males more varied even if homogametic?

      This study* suggests that on average across a bunch of species, mammalian and XX/XY insect males tend to be more varied than females, while the opposite is true for birds and butterflies (ZZ/ZW), at least when it comes to body size.

      * The variability is in the sex chromosomes
      K Reinhold, L Engqvist
      Evolution 67 (12), 3662-3668

  32. Douglas Knight says:

    7x mortality? What is that in life expectancy? 20 years? I don’t believe it.

    I can’t figure out what study the BBC article is about. Here are two papers by one group, from 1997 and 1999 claiming 10x mortality rates, also in the Gambia. The same authors looked at Bangladesh in 2006 and included some graphs about the Gambia.

    • Luke Somers says:

      In life expectancy, it would more or less be when 1/14 of the generation would otherwise be dead. I don’t know when that would be, but it seems pretty reasonable to suppose it’s markedly older than 20.

    • Douglas Knight says:

      Yes, 7x mortality, which is the opposite of reducing mortality by 86% is worth 20 years of life.

  33. Jiro says:

    On the other hand, Vox’s test for political bias is actually really clever and really good. It asks you a bunch of factual questions about politics, like “Do studies show capital punishment decreases crime?” and then sees whether your errors form a consistent pattern that flatters your preconceptions.

    I responded to this.

    • Deiseach says:

      You may be pleased to learn that I am only 5.6% biased about American politics 🙂

      Though I wish they’d included a “don’t know” option; I guessed about half the answers because I don’t have a clue about studies about capital punishment and crime. I suppose they had the idea “If you’re conservative, you’re pro-death penalty, so you’ll think it works to deter crime and vice versa for liberals” but, um, I’m conservative and I’m anti-death penalty (yeah, blame the pope for this one).

      • anon says:

        Apparently there are no right wing Catholics in the Us, or if there are, they are not commonly acknowledged. It’s rather puzzling.

        As an European, I often get the impression that Americans, more than others, see the political spectrum as having only one dimension.

        I don’t want to sound derisive here; I actually think that this is a puzzling and interesting phenomenon that needs to be analyzed. Probably the reason is their rigid two party system.

        • Adam says:

          The Catholic viewpoint does seem to get drowned out for whatever reason. I’d wager most of my family takes basically papal party line positions, help the poor, don’t be materialistic, don’t fight wars except for your own survival, against the death penalty, euthanasia, and abortion. But for whatever reasons of historical contingency caused the marriage of business interests with protestant evangelicals in the United States, the baseline ‘red v. blue’ spectrum thing puts these all over the place and makes it seem like ad hoc nonsense, like it’s oh so obvious from first principles that if you’re against abortion, you have to also be against unions and state welfare benefits because those things clearly have so much to do with each other.

        • Nornagest says:

          There are right-wing Catholics in the US, but not many of them — Catholics are a historical mainstay of the Democratic Party’s moderate wing, and the Republicans have had only limited, recent success in courting them. This may have something to do with the historical significance of anti-Catholicism, which is now basically irrelevant outside of Chick tracts and the half-dozen Protestant fundamentalists that read them unironically, but which was a big deal well into the second half of the 20th century.

          The Red/Blue thing is more sociological than political as such, especially in an SSC context. That is, Red Tribe (in Scott’s parlance) or phrasing like “he’s a Red stater” describes a cultural grouping that the Republicans court, not necessarily a person’s policy preferences. George Bush senior, for example, was culturally Blue, and Bill Clinton was at minimum kinda violet. (Hillary’s solid Blue, though.)

        • brad says:

          One of our most prominent right wing Catholics (Justice Scalia) claims that the Church’s position on the death penalty isn’t infallible and he isn’t buying it. Also, why aren’t all masses in Latin anymore?

        • Anthony says:

          Aside from opposition to abortion and same-sex marriage, Catholics who follow church teachings won’t necessarily take “right-wing” positions. Many economic issues are not matters of faith’n’morals, and so are left to people’s prudential judgement, and therefore there are Catholics all over the place on those issues.

  34. onyomi says:

    Wow, I tend to greatly underestimate the size of countries close to the equator: I mean, I knew Algeria and Libya were pretty big, but… And conversely, you’re not SO big, Russia…

    • anon says:

      Russia is still huge.

      For me the surprise is that Australia, Brazil, Canada, China, an the US are all about the same size.

      I didn’t think Australia and Brazil were that big.

  35. onyomi says:

    I can say that judging in rhythmic gymnastics, in which I was very interested for a little while, and the judging of which is very subjective, is hugely biased in favor of teams which already have a strong reputation. One time I saw Russia (the favorite) give a terrible performance, and the judges waited a very, very long time to produce their scores. Clearly, it was like “oh shit! I had my Russia scores all ready to go and now I’ve got to figure out how to knock them down a bit without also totally knocking them off the podium…”

    Which is sad, because I don’t like the deemphasis on aesthetics and emphasis on pure technical achievement which has taken place in sports like gymnastics and figure skating as a result of the shift to make the criteria more objective, but it seems like maybe it was necessary, since the judging of aesthetics IS definitely wildly biased.

  36. ddreytes says:

    There’s a pretty simple solution to the ‘anomalies’ in presidential polling Kos piece: head-to-head polling more than a year out from the election is basically meaningless and should be treated as such. The mistake was trying to take the head-to-head polling as a useful indicator of the state of the race to begin with.

  37. Speedwell says:

    About numbering systems: I used to support an engineering database where we kept track of design changes using versioning. The first version was designated A, then the next change moved it to version B, then a further change to C, and so forth until Z. The next version after Z was AA, then AB, then AC, up to AZ, BA, BB, BC, all the way to ZX, ZY, and ZZ, whereupon you started over at AAA – ZZZ, AAAA – ZZZZ, and so on. One of the interesting time-kill questions I posted to my training classes was, “what kind of numbering system is this called”? We came up with some ideas, but the major hurdle was that we had no zero. We had a small program that could allow us to tell how many versions were between, say, version DHV and version BK, by brute force assigning a natural number to each possible letter version, but we could not think of any way to actually perform math with the version letters. Finally we basically settled on it being a sort of tally system, just an alternate natural number counting method. Your thoughts?

    • Nita says:

      Hmm, sounds like Excel columns.

    • brad says:

      Without a zero it is not an abelian group, that’s why the resulting algebra is ugly.

    • Izaak Weiss says:

      It’s weird. It’s almost base 26, except you don’t have a zero. Some quick googling reveals that what you have is a Bijective numeration:

    • Luke Somers says:

      Seems to me it’s like Z = 0 but if you actually write a Z out, you need to carry an A into the next column, and aside from that it’s base 26.

      A = 1
      Z = 26
      AA = 26+1
      AZ = 26 + 26
      BA = 2*26+1
      ZA = 26*26+1
      ZZ = 26*26 + 26
      AAA = 1*26*26 + 1*26 + 1

      So, to do arithmetic on the letters… first you do the carrying where everything to the left of a Z gets incremented or created as appropriate. Then treat Z as zero. At the end, decrement everything to the left of a final Z.


      Let’s check. I’ll show incremented letters lowercase and non-incremented upper.

      AZB – CZ = bzb – dz = awb = AWB

      AZB – CZ = 1*26*26 + 26*26 + 2 – 4*26 = 1*26*26 + 22*26 + 2 = AWB

      Looks good. Not exactly rock solid proof!

  38. Fred says:

    On the other hand, Vox’s test for political bias is actually really clever and really good. It asks you a bunch of factual questions about politics, like “Do studies show capital punishment decreases crime?” and then sees whether your errors form a consistent pattern that flatters your preconceptions. Both authors are part of the broader rationalist/EA community.

    Really? Because it reads to me like a test of “Do you agree with the consensus opinions established by people with heads up their own asses?”

    • Sastan says:

      Let’s be fair! The more accurate formulation would be “Do you agree with the consensus opinions established by people carefully selected for their political orthodoxy and slavish subservience to the current liberal establishment?”

      • anon says:

        If you believe that the people with the heads up their asses are only wrong when it goes against your side, you’re biased. So test working as intended.

        • suntzuanime says:

          Not true. Reversed stupidity is not intelligence; even a particularly braindead consensus wouldn’t be wrong about everything. This becomes even more clear when you consider Sastan’s theory, because then you would expect the errors made by the consensus to be correlated; measured relative to a biased consensus, truth looks like bias.

  39. solis says:

    8.33% political bias!

    However the test is flawed. There’s nothing like a consensus that the Obama stimulus package has lowered unemployment.

    • Adam says:

      Hundreds of billions went to highway construction, which trivially employed people. This isn’t really a political question. The political question is whether that’s a good long-term strategy for increasing employment given the economic distortion and method of funding, but it definitely increased employment as a direct short-term effect. They’re also pretty shitty temporary jobs.

      • MichaelT says:

        But it isn’t even clear that it effected short term unemployment positively, considering that unemployment exceeded their worst case scenario models beforehand.

        • Because the Interest on Reserves policy was unprecedented and economists had no idea how badly it could screw things up. As opposed to fiscal stimulus, which has been tried enough times that economists generally have a sense of it works.

          • MichaelT says:

            When has fiscal stimulus ever gotten a country out of a bad slump? The Great Depression slogged on even though Hoover, who very much intervened in the economy contra to most history books, and Roosevelt spent massively to try to get the country out of depression to no avail. Also, Japan used both fiscal and monetary stimulus during the 1990’s to no avail. The research on the supposed Keynesian Multiplier effect of fiscal stimulus is divided at best.

          • Cerebral Paul Z. says:

            Perhaps the question should have been “Would the Obama administration’s 2009 stimulus package have reduced or increased unemployment, in the absence of monetary offset?”

          • Oh, generally monetary offset does tend to make fiscal stimulus meaningless. But I think it’s pretty clear that the Fed was messing up badly in 2008 and I think it would be a mistake to assume they would have done enough to make up the difference if the stimulus bill hadn’t passed.

          • Nathan says:

            You’re right that they were screwing up, but I don’t think that we can reasonably say with confidence that we understand their thought processes and the pressures on them to such an extent that we can confidently predict no difference in their actions in a very different hypothetical. I think it’s at least likely they would have tried QE sooner. At the very least it seems a stretch to say they definitely wouldn’t have and that means the overall effect on unemployment is uncertain.

      • Randy M says:

        I recall some contraversy that the “shovel ready jobs” were not, in fact, ready. By now Im sure the money allocated has, ahem, trickled down. But the larger point made in response would be that “what would have happened with the money taken to employ those contractors if it had not been spent by the government?”

        Wasn’t Joe Biden responsible for tracking that? Did he ever issue a report? (not quite rhetorical, I just recalled that and hadn’t heard if anything came of the VP’s oversight).

        • Adam says:

          Well, some of them were. I had the unfortunate pleasure of driving across the entire country twice shortly after ARRA was passed and virtually the entire state of New Mexico was under construction, though it wasn’t as bad everywhere else. Things other than highway construction were probably not quite so shovel-ready.

          The second thing was exactly my point, though. It could crowd out private investment and lead people into shitty temporary jobs when they would have found better work by waiting a little while, and cause a slower recovery, but it still directly employed people, which lowers unemployment, even if it does so more slowly than doing nothing at all.

          It’s almost like they devised a question to see if you’re not willing to ever say anything that could possibly be misconstrued as flattering to Obama, even though it’s not actually a positive thing you’d be saying.

          • Nathan says:

            The monetary offset critique still applies though. I really don’t think that you can say with confidence that the Fed would have acted identically in absence of the stimulus.

  40. Eli says:

    Scott, British universities are already run by the state. Didn’t you know?

    • AlphaGamma says:

      Debatable how true that actually is. They are de jure entirely independent of state control but receive state subsidies in return for doing certain things the state wants them to (like limiting tuition fees charged to students to a certain amount). However, they are self-governing bodies not branches of the relevant department (currently Business, Innovation and Skills). University professors are not government employees.

      To illustrate this, Oxford and Cambridge periodically threaten to “go private”- that is, to start charging their students tuition at the market rate, in which case the government would no longer subsidise undergraduate education there. Obviously a truly state-run university couldn’t do this.

      Interestingly, this situation has also existed with schools in England in the past. There were the direct-grant grammar schools, which ran themselves but were funded by the state. In the 1970s, they were told that they would have to either stop selecting students on academic ability or stop receiving state funding and become private schools.

      • Pete says:

        Curious about this. I went to a selective state funded grammar last decade. Was this something different from a ‘direct grant’ school. Is it because my school had to follow the national curriculum (which is no longer true as the school is now an academy).

        • AlphaGamma says:

          I think the surviving state grammars were never direct-grant, they were maintained grammar schools. These were completely state-funded (direct-grant schools took some fee-paying pupils) and under the control of the local authority (usually the county). It was up to the local authority whether to get rid of them or not, whereas central government passed a law to get rid of the direct-grant grammars. Some counties (notably Kent) still have something close to the system as it operated in the 60s, others have a few surviving grammars, still others have none.

          School-level education in England/the UK is a weird mess born of the fact that various non-state bodies (churches, guilds and charities) were providing free education to a lot of children before universal free education was introduced, and their schools were absorbed into the new system not replaced by it.

  41. Leonard says:

    Why have murder rates suddenly shot up in St. Louis and Baltimore? The obvious thread linking those two cities doesn’t seem to be it; the rise appears to have started before the recent round of shootings and riots.

    If you look at the graph Vox has of St. Louis homicides, then their assertion that the rise started before Mike Brown was killed is colorable, albeit weak. But notice that their graph covers 2013 and 2014. There are only four or five months post-Brown. Strangely, they don’t say anything about what has happened murder-wise in 2015 in St. Louis. What do you think happened in 2015 that they don’t see fit to mention? Murders up 58%.

    I guess that doesn’t fit their narrative.

    Similarly, the Vox article links to a Baltimore Sun infographic showing data through May 31 2015. And indeed, homicides in Baltimore were up (compared to the same month in 2014) in all of the months through May except January. Fred Gray died April 19. So, his death did not cause killing to be up in Baltimore. But did Mike Brown’s? (Or more specifically, the media frenzy subsequent to it and Eric Garners?) Vox carefully ignores that possibility.

    And what about the three and half months after May 31? Well, my quick googling did not find a convenient month-by-month breakdown, but I did find this tool, which can show all the homicides in Baltimore over several time ranges. According to it, for the year so far, there are 235 homicides. Through May 31 according the the infographic linked by Vox, there were 115. Thus, in the 3.5 months that Vox did not see fit to report on, there have been 120 homicides, an average of 34 per month. This is much higher than any single month in 2014.

    The Vox piece is deceptive. It’s hard to believe that the author did not look for such obvious data as what happened this year in the cities he’s reporting on. Foul.

    Please don’t go to Vox if you are looking for fair reporting about any subject that has any sort of left/right political valence.

    • Urstoff says:

      At what point can I rationally ignore any link to Vox? I think it’s reasonable to do that for, say, Salon or WND. Not that they’re never accurate or correct, but just that there is no expectation for them to produce quality content and thus reading an article is more than likely a waste of time. It seems like Vox has been on a precipitous downhill slide ever since it started.

      • Randy M says:

        I must admit some cognative dissonance around Scott’s affinity for Vox.

        • Adam says:

          I get a little amused at the social justice wing perplexed by Scott’s affinity for talking to reactionaries and then see the same thing reversed when somewhat conservative-leaning people are perplexed that he likes Vox. This quiz was re-published from the EA forum. Scott’s anti-neoreaction FAQ was a discussion between him and the media director of MIRI. Like everyone else in the world, Scott has an affinity for things written by his friends. Apparently unlike most other people, his friends span a pretty broad spectrum of politics.

          • Randy M says:

            There are usually at least a couple links from Vox in each link post. In this one there are three (Ferguson, Bias, & Academic), and a link to a third party article discussing another Vox story. This implies that Scott reads them fairly regularly.
            I have seen Vox accused of shamless sloppiness, including on three of the above pieces. I generally find Scott capable of great diligence in search of truth, thus the cognitive dissonance.
            I’m not saying I’m offended or think he’s going over to the dark side (although I do think Klein and fellow bloggers are coming from the same ‘rational progressive’ sort of self-identification Scott prefers to see himself as) just that I am noticing confusion.

            The fact that Vox publishes ‘freelancers’ as it were may be the explanation–some are good, many are not.

      • Urstoff says:

        I do, of course, make exceptions for people who are generally able to spot content from slop, such as Scott and Tyler Cowen at MR. If they link to Vox, I’ll generally check it out.

      • Whatever Happened to Anonymous says:

        >At what point can I rationally ignore any link to Vox?

        Well, I don’t know, I used to think I could pretty much disregard anything that came from Slate, but then I found out they have a guy called Auerbach (or something like that) who seems pretty reasonable. You can similarly find some pretty interesting stuff in Vox among the “Something Is Not Perfect Somewhere, Here’s How It’s All American, Republican, Straight White Males’ Fault”.

        • Cauê says:

          Slate has a few black sheep (if you were suprised they have Auerbach, check out Saletan some time). I think Vox is a lot more homogeneous.

        • CatCube says:

          Emily Yoffe is usually pretty good at Slate. She also occasionally gets ripped up by her notional allies, like when she advised young women that it probably wasn’t a good idea to get completely shitfaced because it increases the likelihood of sexual assault.

      • There’s lots badly partisan writing at Vox but I still follow Yglesias’s and Matthews’s articles.

      • gattsuru says:

        I’m a very vehement critic of Vox, but I don’t think they’re worth ignoring yet. They’re often inaccurate, wrong, or (most often) so shallow as to miss the interesting story, but because they’re so obvious about it they present a lot of opportunity to dig into deeper questions that you’d normally not have cause or reason to inspect.

      • haishan says:

        Matt Yglesias is often pretty good. Dylan Matthews is often pretty good. Anything related to race or gender is indistinguishable from everything else the media puts out related to race or gender. Everything else varies widely in quality.

    • xq says:

      Strangely, they don’t say anything about what has happened murder-wise in 2015 in St. Louis. What do you think happened in 2015 that they don’t see fit to mention? Murders up 58%.

      Did you not scroll down? There’s a bar graph showing exactly that in the article.

      • Leonard says:

        Thank you for pointing this out. It makes clear that the author does know — or did, at least, when arguing for the defense — that murders are way up in St. Louis and Baltimore. Thus, he is either actually lying by omission when presenting the prosecution, or perhaps is so ideological that he simply cannot report on the issue fairly. Either way, it’s rather damning.

        The graph showing 60% is way, way down in the article. It’s in another section altogether, specifically “Theory: The cities seeing a rise in murders aren’t part of a single national trend”. (In other words, “nothing to see here, move along”.)

        (BTW, I did see that graph when I read it, but I paid little attention since “nothing to see here” is plausible. I do not expect to see evidence for section A in section D, though.)

        If you are trying to be fair, you should mention all of the murders in 2015 in the section “Theory: The ‘Ferguson effect'”, because all of them are as relevant (if not more so) than the cherry-picked-looking data that actually appears in that section.

        • xq says:

          They actually mention the 60% increase in St. Louis in the first paragraph. They definitely aren’t trying to hide it.

          I think interpretation comes down to whether you believe there actually was an increase pre-Brown. It is reasonable to favor monocausal explanations over multicausal explanations, so if the increase predates Brown, that’s decent evidence that the cause also predates Brown. If you think the pre-Brown increase just represents typical variation and all the real increase was post-Brown, then it becomes much more plausible that Ferguson is the explanation.

          So, was the pre-Brown increase real? I don’t know. Need more research.

          • Adam says:

            I didn’t get the impression they were trying to prove or disprove anything. They were presenting multiple hypotheses that were all more or less of equal plausibility, including the ‘meaningless noise’ hypothesis, the point not being that this or that is true, but that which you choose to believe is more likely to reflect your underlying bias than to reflect the true cause.

    • PGD says:

      Even their argument on St Louis is weak — basically, murder rates were up briefly pre Mike Brown, so let’s just assume that Ferguson had no impact on anything that happened after his death? What kind of argument is that? If you look at the pre-August 2014 months in St Louis, from their graph it looks like murder figures were equal to 2013 for 3 of 7 months, lower for 1 of 7 months, and higher for 3 of 7 months. Then after Brown’s shooting they are consistently higher (often much higher) every month in 2014 and 2015. I wouldn’t say it’s slam dunk either way but to claim this as conclusive is ridiculous.

      The intellectually honest thing to do would just be to present the data and say we don’t know for sure. But not only do they draw conclusions, they give a condescending lecture about how you must be ignoring the facts if you think that huge increases in violent crime post-Ferguson have anything to do with Ferguson. Then they have the guts to lecture you about ignoring ‘blips’ in the data, as though their entire argument against the ‘Ferguson effect’ wasn’t based on just that!

  42. Vaniver says:

    A recent Global Post article argues that the Chinese have screwed this one up by making a law that if you run over someone and they live, you have to pay for their medical care; but if you run over someone and they die, you have to pay a flat fine which is often much less. Therefore, Chinese drivers who accidentally hit someone will try to run them over again to finish the job.

    I heard this, with the exact same logic, about New York City taxi drivers about 5 years ago. The economics checks out, so I suspect it’s real, but I also suspect that the effect size is small.

    • houseboatonstyx says:

      If it happened in New York I think we’d have heard about it, with videos. I mean, back up and run over them again!? That sounds kind of noticeable.

      I have seen that advice seriously given in pro-hand-gun circles, though.

      • “I have seen that advice seriously given in pro-hand-gun circles, though.”

        I’ve had the advice given to me by a police officer—in Philadelphia c. 1975. He told me that if I ever shot an intruder, I should make sure of two things–that he was dead and that the body ended up inside my house.

      • FacelessCraven says:

        “I have seen that advice seriously given in pro-hand-gun circles, though.”

        Yup. It’s a bit oblique, as good tactical doctrine tends to be more lethal as well, but the point is usually made that departing from good tactical doctrine is not only bad tactically but legally as well. Various things that might seem like a possibly reasonable gamble to lower the lethality of a shooting are strongly discouraged both on tactical and legal grounds. Using FMJ ammo rather than hollowpoints, for example, would:
        A – significantly increase the chances of your opponent surviving if you shoot him,
        B – increase the risk that he would still manage to do you harm after being shot, and
        C – increase the chance that he’ll sue you for shooting him.
        The chance of B is pretty unknowable, but arguably relatively small compared to the increased chance that the attacker survives; there’s a moderate amount of evidence that the “stopping power” of most handgun cartridges is largely psychological for the person hit. The increased chance of C seems pretty high, and if the attacker chooses to put you in a situation where you legitimately fear for your life, well…. you do the math.

        [EDIT] – and yeah, the “pro-handgun-circles”, ie the practical self-defense shooting community definately includes a heavy contingent of cops as well. As I understand it, the rules for what constitutes a legal use of lethal force are fairly similar for both groups (in theory, if not in practice). Near as I can tell, the advice has only grown more common over time.

      • Vaniver says:

        I mean, back up and run over them again!? That sounds kind of noticeable.

        You’re right; the specific claim I heard was that they would speed up rather than slow down, in the hopes that a collision would be fatal. I didn’t hear the “run over them again” part.

        • houseboatonstyx says:

          @ Vaniver
          You’re right; the specific claim I heard was that they would speed up rather than slow down, in the hopes that a collision would be fatal.

          (I suppose you mean a ‘collision’ with a pedestrian or a biker.)

          That’s more believable, being easier to pass off as accident. It would take very quick thinking though, and quick over-riding of normal habits of safety and morality.

          I think it would be more likely for a witness to report “He deliberately speeded up!” — when in fact it was an accident.* But that report second-hand makes a good urban legend, and if first-hand it makes a reason to sue the driver, so there may be more reports than there ever were such events.

          * (EG wrong pedal, or an attempt to get through the intersection before the victim got into it.)

  43. Vaniver says:

    The obvious thread linking those two cities doesn’t seem to be it; the rise appears to have started before the recent round of shootings and riots.

    The graph in that article looks to me like random variation around one level, and then a shift upwards to a new, higher level, around which there is again random variation. Under such a model, you would expect an increase before the baseline shift about half the time, so that doesn’t seem like evidence against because the odds ratio is 1:1.

    This is the place where Bayesian statistics is super useful; you can figure out the likelihood ratio for the underlying crime rate switching on various days, and then combine them with your prior.

  44. In 1956, Mao Zedong declared a policy switch toward greater freedom of speech, and invited party officials and intellectuals to criticize him and his leadership so he could learn from their advice. In 1957, Mao declared that it had all been a trap, and killed everyone who had spoken up.

    (1) The Hundred Flowers incident was so well-known during the Cold War that I’m startled to see it cited here as something weird or surprising. People used “hundred flowers” as shorthand for “never trust the Commies”. This unsmoothed Google Ngram shows that “hundred flowers” (peaking a decade after it all happened) entered the language to about the same extent as “pattern matching” has more recently.

    (2) I’m no fan of Mao, but it’s not accurate that he simply “killed everyone who had spoken up”. A lot of people got sent to forced labor camps, which may have shortened their lives, but I’ve never heard that the dissenters were executed.

    • Scott Alexander says:

      I’d heard the phrase “let a hundred flowers bloom” before, but didn’t even know it was associated with Maoist China, let alone the context.

      • Tim Martin says:

        I’m curious what your source is for saying it was a trap, or (more surprisingly) that Mao *said* it was a trap.

        Wikipedia makes it sound like Mao was actually interested in criticism originally, but he didn’t expect the Hundred Flowers campaign to escalate like it did. Once it escalated and Mao felt threatened, he took advantage of the fact that he now knew who was critical of him.

        Another online encyclopedia that I checked explicitly says that historians disagree about whether Mao planned his attack from the start.

        • Scott Alexander says:

          As per the Wiki article: “Mao remarked at the time that he had “enticed the snakes out of their caves.””

          • Whatever Happened to Anonymous says:

            In the end, it comes down to whether you believe Mao to be more ruthless (so as to engineer such a situation to scourge the opposition) or petty (where he’d scourge his opposition for being louder than he expected, and then claimed that it was “his plan all along”).

            Being human, he was perfectly capable of both.

          • oligopsony says:

            Variation of the ruthlessness theory: the bait-and-switch was a way of making (the functional equivalent of) a credible commitment not to engage in future speech liberalization, such that future solicitations would not be trusted.

      • Tim Martin says:

        Perhaps your summary was meant to be more humorous than accurate. But factually speaking it looks like a load of Mao Ze-dung!


      • I’d heard the phrase “let a hundred flowers bloom” before, but didn’t even know it was associated with Maoist China, let alone the context.

        No doubt my surprise at people not knowing this is attributable to my advanced age. The Cold War, a constant presence until my mid-30s, is ancient history now.

  45. “Come back when you have equivalent groups.”

    Scott, those two groups will always vary on a huge number of variables. You will never have equivalent groups, and you will never have equivalent results.

    • Sastan says:

      They were all marines who had undergone identical training regimens. The women had undergone special selection and been given special preparation to bring them up to speed. The only differences would be sex and the fact that the more senior male soldiers would have four to eight years of experience to draw on. The structure of reporting squad-level results reduces the impact of this, as there would only be one or two senior enlisted men in each squad.

      • Who wouldn't want to be Anonymous says:

        In particular, there are only one or two senior enlisted men in each squad. No women have the requisite experience to fill those senior positions, so the senior positions are all filled by men in both groups. The difference between the test and control groups is limited to the junior enlisted positions which, as I mention upstream, has a fairly narrow experience variance.

  46. njnnja says:

    Isn’t the one thing we learned from “broken windows,” “crack cocaine,” “lead in gasoline,” and “abortion” debates for why crime declined is that we have no idea what makes crime go up and down? In which case, rather than try to figure out if the “Ferguson effect” is increasing crime rates (which the timing seems to rebut anyways), shouldn’t we be figuring out if an increasing rate of crime causes a higher incidence of police brutality (which is consistent with the timing of crime increases first, then notorious police incident happens)? If that is true, then even if we can’t figure out what causes the uptick in crime, we could identify areas where there is a higher risk of police brutality and proactively try to implement programs to deal with that.

    • houseboatonstyx says:

      Concerning ‘lead in gasoline’, there is a great amount of objective data out there, if it is collectible. Similar localities banned lead at different times, which can be studied along with their crime rates; IDIC.

      • SUT says:

        There’s still confounding factors. For example, some localities are presently banning plastic bags. Others may in the future. Do you think there’s a common factor in localities which are early adopters?

        • Adam says:

          Dallas started charging $0.05 per plastic bag and shopping became really annoying when places that don’t offer paper (especially Target) would try to minimize our bag cost by shoving way more stuff than actually fit into a bag, like we seriously cared about paying an extra $0.05. Luckily, the measure lasted about two months before it got repealed.

          • Deiseach says:

            Didn’t shops in Dallas start selling reusable bags? The plastic bag levy in Ireland started the same way, and now most people have a motley selection of bags to pack the messages in when they go shopping.

          • Adam says:

            Shops everywhere I have ever been in the United States have sold reusable bags for a long time. I was a little surprised that Dallas of all places would try to do this. The only other place I had ever lived with a plastic bag restriction was San Francisco, which outright banned them a while ago. I think Los Angeles has actually banned them now, too, but it’s been a while since I last lived there.

          • Nornagest says:

            Don’t know about Dallas, but larger stores in California now commonly offer a choice between reusable plastic bags for about $0.50 and single-use paper bags for about $0.05. (Single-use plastic bags are banned in most Californian cities.)

            As best I can tell, the only people that actually go for the reusable option are aging earth-mother types. Though I kinda like being asked if I want a bag; previously, I often felt rather silly carrying e.g. a single tube of toothpaste out in its own bag if I hadn’t had the presence of mind to say “no bags, thanks”.

        • houseboatonstyx says:

          Do you think there’s a common factor in localities which are early adopters?

          That wouldn’t matter. We’re looking at Town A’s own crime rate before and after it bans leaded gas.

    • Randy M says:

      You seem to be saying since we don’t know exactly what causes crime flucuations, we should go to where there is the most crime and tell the police to make sure they are extra careful.
      Isn’t that like saying “We don’t know what caused this wildfire, so we’ll make sure to give the fire department safe driving classes before dispatching them, since we do know speeding causes accidents and firemen speed to crimes.”
      Perhaps there’s some good that could come out of that, but it seems a misplaced focus at teh very least.

      • njnnja says:

        I don’t mean to say that we should stop doing research on crime; merely that these interesting new data points about higher crime rates in St. Louis and Baltimore might be part of the answer to a different question. It might be low hanging fruit that has not been well researched and therefore could do a lot of good (at least more than the standard mood affiliation arguments about what causes crime).

        And I’m not sure about firefighters, but the US Army does spend a non trivial amount of resources on basic safety such as driving (motorcycles especially).

        • Randy M says:

          Yes, but not *during the fire*!
          (Except for the old story about first responders being turned way from Katrina until they took sexual harrassment courses. Quite possibly a lot of exaggeration there, but the same principle.)

  47. Hook says:

    About that cow magnet: I never expected gizmodo to cover a piece of technology that my grandfather was using in the seventies (and possibly before). I had a couple of spare magnets just like the ones pictured as toys in the early eighties.

    • Tau says:

      I know, right?

      Yes, rumen magnets are neat and effective preventative medicine. No, it’s not news.

      It’s articles like that one that remind me to put almost no weight on the opinions of most urban Americans about livestock welfare. Because the same people who will talk a good game about “humanely raised meat” or whatever have no actual idea what that means on a practical level – either how to accomplish it, or how to tell if it’s happening.

      • keranih says:


        I mean, I’m all for not being cruel to the things we eat…but the people casting rocks at “poor welfare” should at least know what they are talking about.

        (IE – guess what was the primary cause of “hardware disease”, and whether it was more common in “bad farmers” or “good farmers” in 1950. Or in “pastured cows” vs “feedlot cows” in 2000.)

    • Hari Seldon says:

      Cow magnets were some of my favorite toys as a child. Super strong and made a really awesome noise when you let them roll them down the side of the fridge / shed / tractor.

  48. Whatever Happened to Anonymous says:

    “After the hoax was revealed, Rolf Anderberg insisted that Peter/Pierre’s work was “still the best painting in the exhibition.””

    Well, at the very least, you can’t deny their self-awareness.

  49. Ever An Anon says:

    After I read that article on Chinese drivers intentionally killing people they hit a few days back I was curious as to whether or not it was BS. Given that my Chinese girlfriend hadn’t heard of anything like it, or the supposed saying “It is better to hit to kill than to hit and injure,” it’s a good bet that this isn’t real. The law about liabilities for injuries versus death is real so maybe this happened once or twice but it’s unlikely that it is as prevalent as the article implies.

    I didn’t even bother to read the “city where everyone has three girlfriends” article. That’s Bigfoot Confesses to JFK Assassination levels of unlikely. Chinese culture is getting more westernized but even here that would be taboo. There’s an actual Chinese saying along the lines of “A man who has sex without planning to marry is a criminal” attributed to Mao.

    • Hanfeizi says:

      “I didn’t even bother to read the “city where everyone has three girlfriends” article. That’s Bigfoot Confesses to JFK Assassination levels of unlikely. Chinese culture is getting more westernized but even here that would be taboo. There’s an actual Chinese saying along the lines of “A man who has sex without planning to marry is a criminal” attributed to Mao.”

      And yet Mao flagrantly violated that belief, frequently taking peasant girls to his bed. Many executives and officials in China have one or more mistresses. This isn’t uncommon at all; my first year in Shanghai I taught at a school where it seemed like every other female student I taught was some guy’s kept woman.

  50. Sastan says:

    Scott’s criticism of the Marine study is less devastating than he might imagine. The women who participated in this had all received the exact same basic training as the men had, and most of the ground-combat stuff is covered there. They were then sent to the Infantry School for further training, so they had covered all the aspects in schools. Now, it is fair to say that men who are in the ground combat roles for years are probably more practiced in the technical aspects, but this is mostly going to be a factor for the NCOs, realistically. No one else has enough time in.

    Inexperience might explain why women do worse on weapons qualification, but on climbing over a wall? Exactly how much practice do you need to climb a wall? Many of the tasks the women performed poorly at were brute measures of strength, and this is important, as it suggests a more prosaic biological difference. How does this affect the injury rate, which is higher in women?

    All male units were faster moving, and the difference rose in relation to how much weight they had to carry. This is hard to place at the feet of lack of training, either you can carry 150lb of gear or you can’t. The female marines would have the same experience of physical fitness training as the men except…..their standards are far lower! So imagine that, when the standards for physical fitness are lower for one group, they can’t carry as much or move as fast! Seems straightforward to me.

    All male units were faster at casualty evacuation, which again is basically “how much weight can you carry over what distance how fast?”. Training, beyond basic PT, is irrelevant here. I assume women are just as capable of calling in 9-line medevacs and administering first aid. I wouldn’t be surprised to find they performed slightly better at that aspect. But when you have to throw a 200-lb man on your back, his 150lb pack and your 150lb pack on, and run five miles? I’m not sure “training” is gonna cover the differences.

    • suntzuanime says:

      I bet there are tricks for climbing walls that you would learn through experience. I’m dead certain that experience helps you get injured less often. One of the leading causes of injury is not knowing what the fuck you’re doing.

      • Sastan says:

        There are! You learn them in basic, which all the marines in this study went through!

        As to injuries, the summary specifically states that the majority of the difference in injuries was made up of repetitive-use stress injuries, so unlikely to be a technical failing.

      • alexp says:

        I’m not in the military, but I occasionally go to a rock climbing gym and there are plenty of women, whom I presume are much weaker than me (I can do about 25 pull ups) easily outclimb me.

      • Adam says:

        They’re squad-level events, so in practice what people do is the first person gets to the wall, helps the second person get to the top of the wall, and that second person stays on top of the wall and helps everyone else get over. Your individually strongest climber should be the first person helping the second get up because he’ll be the last to go over and will have to get to the top himself. So it’s not really testing individual climbing ability. It’s testing teamwork. You only need one person who can even climb the wall alone.

        • Who wouldn't want to be Anonymous says:

          The primary differentiator on the wall, apparently, is whether they can throw their own bag over the wall, or if the assistant on top of the wall has to haul it over the wall before they can haul the Marine over. That is an upper body strength question.

    • Adam says:

      This is spot on. In fact, my own military experience is that you get worse at the physical tasks with experience, because you accrue injuries with age. I was pretty good at climbing walls, but some kid straight out of infantry AIT was almost always better.

      The thing that stuck out to me was related to what you just said above: real integrated units would have hardly any females at all. Testing a unit with a 1:3 female:male ratio isn’t going to give a realistic profile of what an actual unit would look like because they’re including women who would never qualify for or likely even apply to that job in the first place. For instance, military police and rocket-based artillery units have allowed females for decades, but you might see 1 in 20, not 1 in 4.

    • Cauê says:

      Weirdest part of the linked article:

      Men in the provisional infantry platoon who had not attended the infantry course were more accurate marksmen than women who had, hitting 44% of targets with the M4 rifle versus 28% among women trained at the infantry school.

      • Adam says:

        I didn’t notice that the first time and now I’m wondering what the heck kind of test they were given. Neither of those is even close to a passing score on a standard 40-target popup range.

        • keranih says:

          As I saw in a different article (which I now can not find, and it might have been on the radio) – it was a test of move-fire-move-fire, which is a) more realistic than shooting whilst standing in a foxhole (do they even call them that anymore?) and b) more dependent on cardiovascular fitness.

      • John Schilling says:

        I would wager that a significant fraction of male(*) USMC recruits entered the service having already put more rounds through a rifle than they will in basic training and infantry school, and the most popular centerfire rifle in 21st century America uses the same basic design as the marines’ M4 carbine. Female recruits are less likely to have been full members of American Gun Culture prior to enlistment.

        (*) for various definitions of “male”; I was genuinely surprised to discover that transgender individuals are probably overrepresented in the US armed forces in spite of the nominal prohibition. Not sure if that applies to the Marines specifically.

        • Who wouldn't want to be Anonymous says:

          You say that, but it is worth noting that the Federal Assault Weapon Ban was a thing for a significant portion of most recruit’s lives, and certain large population states have implemented variations of it themselves since the Federal ban expired (setting aside how those certain states may or may not be underrepresented among recruits).

          • Nornagest says:

            I live in one of those states, and it’s not hard to buy an AR-pattern rifle here — you just need to live with obnoxious restrictions on accessories and magazine capacity and/or mag release mechanisms. It’s kind of a joke.

            On the other hand, city-dwelling youth in NY and CA are a lot less likely to fully participate in the gun culture that John Schilling’s talking about; there’s a sort of gun culture in parts of the inner cities, but it’s not the kind that involves a lot of rifles or long-range shooting or really much shooting at all, if you’re counting by rounds of ammunition fired. But on the third, they’re probably underrepresented in the military anyway.

          • Adam says:

            As pure anecdote, almost every infantryman, scout, and tanker I personally knew had been hunting since they were a kid and still did it. AR-15 ownership wasn’t exactly universal, but pretty common. I don’t even think marksmanship was the biggest advantage the experience gave, either. One of the hardest things to get used to for a lot of people is just spending weeks at a time outdoors barely sleeping and being cold and wet or hot and dry all the time. Country boys are already used to that.

        • Sastan says:

          Former infantry NCO here, and once a shooting instructor for my unit.

          Lots of guys come in with shooting experience. We hate them. They have internalized terrible habits. We have to break them and retrain them, and it’s never as good as some of the blank slates. There are exceptions, I was taught by my uncle, who had been on the Marine long range rifle team, and did it by the numbers. I knew a kid from the UP who was a lifelong shooter who was good. All the other best shots had one thing in common, they’d never held a rifle before the Army.

          I’ve taught several female friends to shoot. By and large they’re better than the men I shoot with all the time, and on a tiny fraction of the experience. Unless you are taught and trained correctly, just shooting for fun will make you a slightly better shot, and ensure you will never be a good one.

          • keranih says:

            Have a friend who was in back in the 90’s, got out, went back in some years after 9/11.

            Apparently, back in the day, they shot with ‘plain’ M-16s, and didn’t shoot with helmets or flack vests at the range. After my friend got out, he did rec shooting much the same way.

            In the ‘new’ Army, helmets and vests were the rule, as were mechanical popup targets, and the standard rifle is really different.

            My friend went from ‘expert’ to ‘struggling to marginally pass’. Still bitter about it, which leads me to think there was something more to it than just rosy glasses on ‘the good old days’.

      • ivvenalis says:

        It was probably a stress-shoot (fire-move-fire-move etc).

    • Who wouldn't want to be Anonymous says:

      The other thing that Scott fails to realize that in the military, unless you have a job that is actually useful in Kansas (and let’s be honest, the infantry gives you zero useful skills), you are going to spend at least 50% of your time when you’re not deployed (or doing a predeployment trainup) cutting grass and doing KP. About 45% of the remainder is training that everyone gets (which, aside from periodic weapon qualifications and PT, is universally a waste of time). That leaves a tiny sliver of time to actually learn anything useful. There is basically no experience difference in the combat arms between a recent IET grad and someone who has been in a couple years and never deployed.

      If you’ve been in more than a couple of years–especially if you’ve deployed!–you’re scrapping the bottom of the barrel if you’re not looking at a promotion.

  51. Robert Liguori says:

    Re: the combat study: I’m actually a bit suspicious of the way the numbers were presented. We were told that the integrated units experienced lower performance in two-thirds of the tested-for categories, but we didn’t get a big table of Test, Results For Group A, Results For Control, which makes me wonder if the results were mostly pretty close but had a few bad cases that were emphasized.

    That being said, do the marines still have gender-segregated physical fitness requirements? Because if they do, then we can’t come back with equivalent groups because we have two different filters on two different subgroups. I have a pretty strong set of priors that tell me that there are overwhelmingly more men with the desire, physical ability, and mental capacity to join the military than women, and for every explosive-effort or high-impact sport I can come up with off the top of my head, we segregate gender competitions, because male athletes score higher and do better then their female counterparts.

    Am I missing something here? I don’t suppose that we can adjust for the experience confounder by looking at how fully-integrated armies like Israel’s score on similar courses?

    • Sastan says:

      The marines do have lower PT standards for females.

      And Israel does not integrate their front-line infantry anymore. They scrapped that years ago. Women do serve in armor, artillery and CAS. Fun tidbit, Israeli military policy prefers female instructors specifically to shame male recruits into trying harder.

      There are a number of countries that allow women into all roles, but the numbers that actually make it there are so tiny as to make study difficult if not impossible. France integrated in the mid-80s, last I checked nine women have made it into the infantry, none into their marines. Canada went about the same time, women have made it into armor and artillery, none into Infantry. Denmark integrated long ago, can’t find data on infantry, but they’ve been allowed into SOF for decades, no one has made it in yet. Norway is the only country I am aware of that has identical standards for men and women, and there are some in their infantry, but I can’t find how many.

      • keranih says:

        Fun tidbit, Israeli military policy prefers female instructors specifically to shame male recruits into trying harder.

        I forgot to comment on this earlier – Also anecdotal, but supposedly the US units training Latin American military paratroopers prefer to have female jumpmasters – and the smaller & prettier the better.

        When the stick (jumping group) of brand new jumpers, about to jump out of a perfectly good aircraft (particularly for the second time, because the first time you didn’t know how bad it was going to be) stands up, shoulder to shoulder and knocking knees with how scared they are (*) and the cute little gal stands up and bellows “FOLLOW ME!” as she jumps out the aircraft, it’s apparently dang near impossible for Latino machismo (**) to admit they’re too scared to do the same thing. (***)

        Dunno if they did any double-blind studies, but, as previously reported in the literature proper randomized paratrooper studies are in short supply.

        (*) Because Montesquieu was right.

        (**) American men, apparently, aren’t so sexist, and are occasionally willing to admit that a woman is more ‘crazy-brave’ than they.

        (***) A ‘jump refusal’ is a failure to obey a lawful order, and subjects the person refusing to jump to punishment. To be fair, all modern (and historical) paratroopers are volunteers.

    • Emily says:

      The point of the study is to figure out the effect of integrating combat occupations in the actual, existing Marine Corps. Not a Marine Corps in which men and women have the same physical fitness test. I am sure there are other research questions we could ask if men and women had the same physical fitness test, but they are less relevant to the current policy question that is being asked of this service.

      Edit: also, I’m having a “can’t tell if Scott is being sarcastic” problem with “since the military has only recently started accepting women, all the women were new but many of the men were experienced.” Obviously the military has not only recently started accepting women. The issue is that more of the men had gone to infantry school. But they did some of the comparisons between men and women who had not gone to infantry school, and also infantry school is not really going to dramatically affect lean body mass.

      • Hey, I’m just objecting to Scott’s objection. Scott himself isn’t questioning the potential policy impacts of the study, just what we can and can’t infer from it about the relative performance of integrated vs. all-male units in combat when we adjust for confounders like combat experience.

        Me, I don’t think that the results are being confounded by combat experience. But more than that, I think that this is more or less a replicate-common-knowledge study; I don’t think that it’s controversial that certain tasks are highly-correlated both with success in combat situations (traversing rough terrain quickly while carrying heavy loads, for example) and which have a much higher percentage of men than women who are capable of doing them.

        Of course, not being an expert myself, I have no idea about the relative impact of the tested skills in combat performance. But again, if we’re assuming that physical condition and carry weight affects combat readiness, that we have a cutoff point below which you’re not allowed to soldier, and that you can get in well under this cutoff point if you have a particular characteristic, then it’s blatantly obvious that on a statistical level, barring really interesting clustering effects on that characteristic, people who share that characteristic will be on average lower performers.

        Actually, I’m curious. Why do we have differing physical standards for men and women in the armed forces? Isn’t that an unequal-impact thing? Has anyone tried to sue under antidiscrimination laws around it?

        And wouldn’t it work better just to have a standard of physical ability for intended front-line combatants, hold to it firmly and rigorously, and another standard for different work (loading artillery rounds, I am told, is very physically intensive), another for mechanical work, and so forth?

        • alexp says:

          I think the rationale is that since women don’t serve in frontline combat units (infantry, armor, artillery), the physical fitness standards aren’t so much about combat effectiveness as they are about just making sure the military has baseline levels of physical fitness.

          So as long a female supply clerk is more fit than 70% of the American female population, it’s fine if she can’t carry 80 lbs on a 15 mile hike.

          • Who wouldn't want to be Anonymous says:

            I’ve always been under the impression that the physical fitness standards are a “give a shit” test. They are low enough that (medical waiver not withstanding) everyone can meet them with a little bit of effort, but high enough that if you don’t put in that effort you’ll probably fail. The standard is there to give the service a reason to [suspend favorable personnel action/bar to reenlist/administratively separate] any service members that are unwilling to put in even a minimum effort.

            As such, age and gender normed tests (and medical waivers) make sense. It doesn’t make any sense if you think of them as a functional capacity test.

            In fact, AFAIK, the functional capacity determination is made by a doctor.

          • Sastan says:

            The rationale is that if women had to do male standards, we wouldn’t have any women at all in the military.

            And in combat arms, the standards are FAR higher than the broader military standards. A 180 will get you a pass in a clerk brigade, but in the Infantry, anything less than a 280 gets you remedial PT.

        • Emily says:

          Echoing alexp, I think baseline levels of physical fitness is the best answer to this. In the Air Force, I believe it’s actually normed to health outcomes in some way. In the others, I’m not sure it’s actually normed to anything, but I’d argue that what they’re trying to get at generally is general physical fitness.

          We will indeed be moving toward occupation-specific, gender-neutral (as opposed to gender-normed) physical fitness tests.

          That is actually different from who is a front-line combatant, though. Which came up regarding the Combat Exclusion Policy, which was lifted a couple of years ago.

        • Adam says:

          It would make a ton of sense to change the testing to be occupation-specific, but it’s one of those coordination problems in a large bureaucracy where there’s a really simple solution to a really obvious problem, everyone knows it, and we still don’t do it. Our tests are generally biased toward people with short arms and torsos, and toward endurance over strength, which probably makes a lot of sense for straight-leg infantry where the hardest thing they do most often is walk really far day after day carrying their lives on their backs. For a mechanic or an artilleryman, it’s not at clear why we shouldn’t prefer a weightlifter to a 10K runner.

          Edit: I’ll add that we even do have things that tall people are better at, like the wall climb, and tests that favor strength, like dragging around and carrying a 200-pound dummy, but those are informal and mostly for fun or for specific schools and they have no bearing on the average person’s career the way the actual physical fitness test does.

        • ivvenalis says:

          “Why do we have differing physical standards for men and women in the armed forces? Isn’t that an unequal-impact thing?”

          Because if we didn’t, the military would probably be 1% female (or, conversely, we might as well not have standards for men). As an upper bound. At least as long as the requirements involve upper-body strength (fun fact: the sit-up requirement for men and women in the Army is the same). See

          “Has anyone tried to sue under antidiscrimination laws around it?”
          Haha, funny.

          • keranih says:

            Situp standards being equal across genders reflect, imo, far less an effect of absolute abdominal strength & flexibility than imbalance of upper body weight. (If women had to shift an equal percentage of their body weight with their abs, or if men had heavier hips to hold their butts down, the standards would not be the same.)

            If, if, if.

          • Sastan says:

            I have this great story. I went to Air Assault in the pause between the Afghan and Iraq invasions. All the other combat brigades were gearing up to deploy, so hundreds of AA slots were freed up for non-combat personnel. In the training company there was a platoon of men (around 100) and two of women.

            To get there you have to meet a long list of requirements, including a very easy, but still higher than military minimum PT test. A 240 I think? It’s been almost fifteen years….. Anyhoo. The only thing different about the PT test you get at school is they grade very closely. If you did your previous one right, and trained right, you should have no problem. 187 women failed that first morning. And failed at their far lower standard. No man failed it, including the guys with the flu, and one who had broken an ankle. Around 94 guys graduated, and not one girl. Three had made it to the last day, all they had to do was a very easy (to an infantryman) ten mile road march, medium packs, three hours. Easysause. The women fell out on mile two.

            I have said it many, many times. I am all for equality, so let’s fucking have some. Abolish the lower physical standards for females in the military. Hold everyone to the same rigorous standard. And the first woman that finishes her Bayonet, I will walk my broke-dick happy ass to Benning and pin her cord on myself. But until that day, stupid civilians can shut their mouths about “women can do the same jobs”. And lives are on the line.

          • Adam says:

            I went through the first air assault course when they first started offering it at Fort Hood, and they’d done a trial run to train the cadre with a National Guard unit in Austin the month before. 90% plus failed on the first day. To be fair to the guard, though, it was whatever year it was a few years back where it was triple digits for 74 consecutive days. People were passing out from heat exhaustion at 9 in the morning.

          • Sastan says:

            Lol! Nasty Girls!

            Only triple digits? Drink water, drive on!

          • Emily says:

            If you abolish the lower physical standards for women, you will have a hell of a time getting nurses.
            I’m not kidding about this. And they’re already paying out a lot in terms of scholarships and signing bonuses.
            Doesn’t matter for the Marine Corps. They don’t have nurses. But it matters for the other services.
            I think that as the military does their big personnel reform, they should rethink what positions need to be uniformed military. Maybe a lot of the female-heavy positions don’t. But until that happens, you probably don’t want to make women meet the male physical standards. Because, nurses.

          • Pku says:

            Why not just reduce physical standards for nurses then?

          • ivvenalis says:

            Nurses are already categorically exempt from a boatload of personnel regulations, by not being part of the Army Competitive Category (i.e., simplifying somewhat, they’re not actually “real” officers). This has been the case for a very long time, going back to the earliest we started putting them in a uniform. Not requiring them to meet certain PT standards wouldn’t be any different from the way they’re already not required to meet a bunch of other standards currently.

            The fact that making nurses pass a hypothetical non-gendered PT test would get them all fired has no more relevance than the fact that making helicopter pilots pass the Abrams Master Gunner final exam would get them all fired.

          • Because if we didn’t, the military would probably be 1% female (or, conversely, we might as well not have standards for men).

            I guess I’m not seeing the problem with the latter option. If a physical standard of N is the minimum necessary to contribute to the army for a given role, what does it matter if the person who meets that standard are men or women?

            And I’m not getting the joke with regards to the antidiscrimination thing. I know that the military operates under the UMCJ. Does that preclude civil suits? Are there just an extreme dearth of potential complainants? Or is one of those “Everyone knows, no one cares” scenarios?

          • Science says:

            Given that the entire point is to provide make-work jobs, I don’t see what the big deal is with lower physical standards. I suppose it makes it a little harder for macho men to lie to themselves about how they are heroes “protecting our country” rather than collecting a strange, and sometimes destructive, sort of welfare. But that seems like a small price to pay for opening up the program to half the population.

          • notes says:

            Robert – discrimination isn’t illegal when the government does it through one of the least restrictive means narrowly tailored to advance a compelling government interest.

            National defense is a relatively easy category to establish as a compelling government interest: the military is not commonly understood as a make-work welfare program.

            After that, there’s still an argument about tailoring — still quite straightforward, while you’re checking physical fitness for a physically demanding job. Some potential issues for those not supposed to be seeing combat, but there’s plenty of historical evidence of ‘non-combat’ units seeing plenty of combat.

            Which just leaves the question of ‘is there a less restrictive way to serve the compelling interest as well?’ And that’s a fairly high bar for a plaintiff to clear: any decrease in combat effectiveness (assuming that’s the compelling interest chosen) ends the lawsuit.

          • Emily says:

            Absolutely, lower physical standards for nurses. But it’s not just nurses that are disproportionately female and don’t need to be intensely physically fit.

            So that’s basically what the military is going to do – keep the gender-normed physical fitness test, add gender-neutral occupational tests.

          • Adam says:

            Most medical personnel in the military are already not uniformed, but at least some of them have to be because we need to deploy them and, although we can deploy contractors and civilians, you have to pay them way more and they have to volunteer for it.

            Although the nurses now already don’t compete with everyone else, they had to go to LDAC with us and do at least the bare basics of infantry training (though even there, they only needed to pass and didn’t get ranked). I kind of felt bad for this 4’10”, 90 lb woman in one of the other companies who tore both ACLs just from carrying around a ruck for two weeks.

          • keranih says:

            @ Science

            Given that the entire point is to provide make-work jobs, I don’t see what the big deal is with lower physical standards.

            Factualizes an opinion not shared by a majority of the audience.

            I suppose it makes it a little harder for macho men to lie to themselves about how they are heroes “protecting our country” rather than collecting a strange, and sometimes destructive, sort of welfare.

            Not a statement that will be helpful in convincing military personnel to buy into your proposal. Consider restating, or dropping entirely.

            But that seems like a small price to pay for opening up the program to half the population.

            Assumes facts not in evidence: that (transient) emotional damage to a (non-significant) fraction of male military personnel would be the only negative effect of this change.

          • Science says:


            I don’t have any interest in convincing (existing) military personnel to buy into my proposal. They get the same vote as anyone else. If some find women intolerable they can quit. Given 20 year cliff vesting getting people to quit before that is seriously +EV.

            As for facts not in evidence, given general equality norms, the burden is not on those who seek equality but those who would seek an exception.

            I see how people love to throw around probabilities on this site, so I’ll put it like this: what is the change in probability over the next 50 years that the United States will be invaded and conquered if we put in place lax physical standards for members of the military versus not doing so?

            In terms of likely outcomes, at worst our next ill conceived neo-colonialist adventure will have more military causalities. But if that causes the war to be shorter it may well be a utilitarian positive.

          • FacelessCraven says:

            @Science – “In terms of likely outcomes, at worst our next ill conceived neo-colonialist adventure will have more military causalities. But if that causes the war to be shorter it may well be a utilitarian positive.”

            That is probably the worst outcome for us, you are correct. The knock-on effects for all the countries that have built their policies around the stable global order we maintain might be significantly worse, but hey, not our problem right? I’m sure Japan, South Korea etc won’t have any problems coming to terms with their new Chinese overlords. I’m sure Russia will likewise be happy to step in as Globocop for their neck of the woods as well. Heck, that sort of reallignment might even reverse the leftward push Europe exerts on global politics. Win-win, right?

            I am a little concerned about our ability to maintain global economic dominance in the absence of global military dominance in the long-term, but eh, worth a shot.

            In short, and the debate over fitness criteria notwithstanding, your manifest contempt for the military does not appear to me to be a rationally sustainable worldview, but it seems like the sort of error that rapidly self-corrects. Rock on, sir.

          • Who wouldn't want to be Anonymous says:


            You’re not making a very good case. Those existing military personnel you have no interest in convincing are pretty important to convince. Not only because they get an metric fuckton more say in military policy than you do regardless of how many votes they get, what with the carte blanc they get to write policy about compelling government interests. They also have an incredibly strong moral responsibility to minimize the casualties required to achieve whatever mission they are given–whether or not you get to feel proud of yourself for derisively calling it “neo-colonialist adventure.”

          • keranih says:

            Regarding lower physical fitness standards for women – one might want to talk to current medical people (civilians, I mean, and particularly those in elder care/nursing homes.) With the on-going surge in BMI, the relative lower physical strength of women is becoming something of an issue in lowering healthcare costs.

            If a woman (or two) can not safely move a patient, but must call a male college/orderly (and wait for him to arrive) it slows the actual delivery of health care, and sharply increases cost of staffing. As a practical matter, the physical strength of people counts, no matter what job they’re doing.

            The question – the balancing act – will be how important is strength, and what can replace it, and how much, just as we ask the same question re: intelligence and empathy.

          • Emily says:

            Sure, upper body strength may be important in nurses.
            But no one would design a test for nurses which required doing 40 push-ups to pass and 75 to get the maximum number of points. That would be a very bad way of getting at whether they could do the necessary tasks. And a good test for getting at whether they can do the necessary tasks would not be a good test for infantry.

          • HlynkaCG says:

            I just want to thank keranih, FacelessCraven, and Who wouldn’t want to be Anonymous for stating the case far more eloquently and charitably than I would have.

    • keranih says:

      A scattering of thoughts, culled from a variety of readings and conversations with military professionals over 30+ years:

      Any military unit (like any branch of a large corp) will, at any time, have a mix of more experienced and brand new people. From what I have read about this study, the units were balanced on new people, and matched for gender, and the ratio was designed to match the current USMC gender ratio (which is currently the smallest of all the services.) As the study is essentially US govt intellectual property, I expect it to be fully & freely released. (They might be running into HIPAA/IRB problems regarding the size of the units, limiting the data that can be released. <–this is a guess on my part.)

      2012 US military demographics.

      There are not as many things that count as much, in military terms, as do speed and lethality. If the mixed gender units failed on those terms – and I note that they put the mixed units against multiple male-only units, so that one could not say that the mixed units went up against one stellar male-only unit – it’s likely not possible to convince military leaders that mixed units are good at all. And it might not matter what the spread was, either. Either the company makes money, or it doesn’t. Either the unit wins the fight, or it doesn’t.

      All the services have different physical performance standards, and these were determined (IIRC) off a baseline back in the 70’s on a bell curve of civilians off the street – something like (but not exactly, don’t quote me) mean+ 1 standard deviation = minimum pass, mean + 3 SD = max/excell. All of them have weight standards which have a lot more to do with aesthetics than health or performance. The US Air Force has a waist measurement that *is* specifically calibrated to health/heart attack(/metabolic disease?).

      The Marine Corps standards are interesting in three ways – running emphasis, pullup emphasis, and “Every Marine is a rifleman.” Women who can pass the running standards struggle with pullups/flexed arm hang. And there are few ‘support’ roles in the Marines (ie, no Marine medical personnel.)

      Again, by my reading, there was a serious push back in the day for gender-equal physical standards (and haircuts.) Often, these were made in good faith by people who thought that equal standards (as had been allegedly used during racial desegregation) was the fastest way to integrate women into the military. For various reasons – both cultural and practical – that was not the chosen option.

      OTOH – many senior military leaders were leery of female volunteers in the 1980’s, afraid of being swamped with “leftist feminist lesbians”. What they got, culture-wise, were the sisters and daughters of men who were already in the military. And “getting pregnant” is still a major hazard of being young and female, in the military or out of it – despite the universal healthcare provided all service members.

      The comment about “we need lighter/different equipment for the women infantry” is, by historical standards, baloney. When lighter versions of things have been invented, the military gives the infantry more things to carry. (IE, a heavy weapon to every four people, instead of every 12.) More stuff (assuming it’s the right stuff, like weapons, ammo, explosives, radios, food, cold weather gear, water) means fewer dead soldiers. People are *expensive*.)

      A statement I have seen in various places and forms over the last few years – an ideal squad (12 people) is going to have at least one huge guy and one little limber guy – but you can’t have 12 little limber guys or nobody can carry the heavy weapon (‘hump the hog’???)

      As said above – no, Israel doesn’t have integrated infantry anymore. Aside from that, the higher social-status units are either all male or supporting all male units (ie, the parachute repair and packing teams (usually all female) who are assigned to all male paratrooper units are higher-status than their sisters supporting mixed gender units.)

      My opinion is that “men fight, women only fight to defend the home, if at all” is one of those things which work very well for humans, and so is a highly conserved cultural trait. This doesn’t mean that we have to make it a legal rule to only do things this way, but it does mean (imo) that we should not expect unlimited goodness from doing it another way.

      Not personally crazy about that division of gender roles, but it’s not like I was given the option to be born a dolphin or an elephant.

      • Emily says:

        Whoah. I’ve never heard that story for norming. That’s really interesting. Do you have a paper on that?

        • keranih says:

          @ Emily


          [Edit] – Oh, the 1970’s pt standards, right.

          No recollection of the original study, but this rather exhaustive review probably has the details in the notes someplace. (If my memory failed me and I reported wrongly, please to let me know.)

      • Sastan says:

        Your read is pretty good, but one caveat from experience.

        When I joined, I was very skinny. So skinny I had to put on twelve pounds to be admitted. The first thing my first unit did when I arrived? Put me in Weapons Squad and give me the 240 tripod. It seemed cruel at the time, my pack weight was well over my body weight. But combat is cruel, and every member has to be able to do every job, all the time. There is no room for weak links. So you suck it up and lift weights like a mofo, and eat eight times a day. You can’t just stick a woman with a male team and think “well, she’ll be their little guy”. The demands are identical.

    • bean says:

      Israel has adopted a unique solution to the gender-integration problem, but not one we can copy. They’re not fully-integrated, although they do have one integrated infantry battalion, described by them thusly:
      “The IDF Karakal is an elite infantry battalion established in 2000 and it is the first Infantry battalion to incorporate women combat soldiers. Karakal is entrusted with a vital mission for its nation: securing Israeli‘s borders against terrorists and border infiltrators.”
      Now, I happen to have a copy of The Joint Dictionary of Military Terms (not really) and when I look up elite infantry, I find as an example the 75th Ranger Regiment. Which spends its time training for raids and airfield siezures and the like. The Israelis have similar units. This is not one. When I look up their mission, I find that they’re border guards. This battalion seems an attempt to keep the Israeli feminist lobby happy. If it was such a success, the IDF would probably have adopted it for other units.
      Of course, maybe the IDF is hidebound and reactionary. This is a common charge leveled at militaries. There are two reasons to doubt this. First, the IDF is a notably progressive military by nature. This is not usually a good thing, and leads to poor staff work and other hindrances to performance. Second, they actually are under pressure, which puts a cap on how reactionary they can be.

  52. david says:

    This week in neat methodologies: evidence in bias at Olympic games. Winners from last year gain a greater premium in subjective contests (eg figure skating with judges) than objective contests (eg the hundred meter dash), suggesting that judges are biased upward by an athlete’s previous reputation.

    The paper is weirder than that – it argues that judges are biased upward by by athlete’s country’s previous reputation, i.e., country X benefits from country X doing well in the past, even if it is sending an entirely different team of athletes.

    A plausible hypothesis, but ill served by the paper (which avoids panelling out whether countries are, in fact, sending the same athletes). This exacerbates the confounder of whether the sports that demand subjective judging are intrinsically more subject to smoother career trajectories (and therefore sustained performance) than sports which are not (after all, objective sports may favour extreme physical characteristics that fade exceptionally rapidly with age, or are perhaps not amenable to the greater investments of national resources that would come with bringing home an Olympic medal).

  53. TMK says:

    Eh, the best comments about left-right were from Anonymous and Eli, who, basically, stated that the spectrum got left on the social issues, and right on the economical issues.

    That is why both statements are true.

  54. The “Harvard cannot afford journal subscriptions” meme often fails to acknowledge that, if you were a rich university like, say, Harvard, you’d probably want, on purely selfish grounds, a system where prices are fixed for all universities because, as it stands, publishers are able to get more money out of you than they get from, say, Greek universities (even on a “per-researcher” basis). As it so happens, the open access system does charge the same per paper to Harvard and to Greek universities. However, the 2000 USD-something that it costs to publish a paper in PLoS are irrelevant for Harvard, but a significant cost in Greece.

    Now, it may be the case that it is socially beneficial to move to open access as a whole, but Harvard et al are not disinterested parties.


    Many of these “institution X cannot afford Y” are part of an ongoing negotiation between X and Y, they should be taken as such.

  55. DavidS says:

    Harold Lee thing is interesting! Though more on atheism than immigration. When he talks about children slipping from their parents values, my immediate thought was ‘but why is this seen as bad – maybe their values are more appropriate in the new situation’. And of course this applies massively when you change foundations (e.g. by becoming an atheist society) but expect the super-structures to remain the same.

    • Earnest_Peer says:

      The connection of decadent rulers and third generation Chinese immigrants to atheist values decay sounds spurious to me. The suggested process is “new group outperform the norm and regress as they stop being new”, but he doesn’t establish atheists as outperforming anyone, so they would stay at the norm rather than regress to it.

      Also: Visibly first generation Chinese immigrants might outshine their descendants, but is that borne out by the numbers? I thought that both in the US as well as Asia, the pattern goes “Chinese immigrate to new country, outperform everybody and continue doing so forever”.

      • LeeEsq says:

        I work with a lot of Chinese immigrants as part of my job. A lot of them aren’t Tiger Parents pushing their kids to do really well in school. Kids of working class Chinese immigrants seem to be just as apathetic about school as a normal American kid. The Tiger Parent thing is really from Chinese or Asian immigrants that are working and earning at a middle class income in the United States.

    • Scott Alexander says:

      Some people will consider this a copout, but it’s bad insofar as, if we are the parents, we don’t want our children deviating from our values.

      (some people will say they’re okay with that, but I think this is only insofar as one of our values is “letting our chidlren find their own way within limits”. If they started really deviating from a value we hold sacred – like thinking murdering the weak is okay – then we would be upset).

      It’s also a problem insofar as “our values” are not just “things that we want” but our ability to pursue the things we want. Presumably even decadent Manchus understood that being a strong and powerful ruler was better than being decadent, they just didn’t have the drive for it.

      • Weevilbits says:

        I’m not sure if this has been explicitly pointed out, but I think this gets at a basic difference between the socially liberal and socially conservative understanding of what values are and how and why they develop. Call it a decadence frame versus and adaptive frame.

        Under the decadence frame, there is a clear objectively best set of values that are often enumerated in terms of religion or culture but on a meta level are basically something close to the “zombie apocalypse” survivalist values you mentioned previously. As societies get wealthier and more powerful, the need for those values erodes and people grow decadent. You can frame this in purely moral or spiritual terms, or frame it in purely practical ones (their ability to pursue the things they want is reduced). The social conservative project then is to forestall this process. Obviously here I’m not saying anything new, I’m in fact basically summarizing the article that started the discussion.

        Often when people (liberal or conservative) try to describe the alternative frame, they seem to just go back to describing the decadence frame while sort of tacking on “who cares, we’re rich enough to afford it and people *like* decadence” on to the end. This is sort of the whole idea behind the “self-expression” ethic that supposedly dominates rich societies.

        The alternative to the alternative is to say that societies develop the values that help them succeed in their current setting. If third generation immigrants are studying less and partying more, maybe that’s just because partying is more fun than studying or maybe it’s because as you transition from the working class to the middle class to the upper middle class and true elite social networks become relatively more important than pure academic performance. Maybe being first ranked in your class is a suboptimal strategy compared to being twentieth ranked but becoming best friends with someone whose dad will kick in venture capital for your startup.

        • Ghatanathoah says:

          s societies get wealthier and more powerful, the need for those values erodes and people grow decadent. You can frame this in purely moral or spiritual terms, or frame it in purely practical ones (their ability to pursue the things they want is reduced). The social conservative project then is to forestall this process.

          Wouldn’t a better solution just be to tell everyone, “Look, here are some values that are useful day-to-day life, here’s another set we need for emergencies, but are otherwise useless. Let’s keep both of them handy.”

          It seems like a lot of survivalist fiction deals with this a lot. You have people who are thrown into survival scenarios and have to act differently than normal. Maybe I consume way more fiction than normal people, but to me the idea that we need to act differently in a survival scenario than in normal 1st world country life seems like a cliche to me. It’s like “power corrupts,” “don’t be intolerant,”

          Also, it seems like a lot of social conservative values are pretty useless, even in survival scenarios. What is killing gay people supposed to accomplish exactly? Wouldn’t gay people make useful hunters, farmers, and soldiers in a zombie apocalypse? If you really need them to reproduce you don’t need to force them to live a straight their whole lives. You just need to make them put up with straight sex until they’ve conceived a few kids, and then let them resume being gay. Maybe have gay couples and lesbian couples have special quad marriages.

          The distrust of foreigners value seems similar. It seems to me like xenophobic societies tend to be full of conflict, either internal or external depending on how homogenous they are, while pluralistic societies tend to be peaceful. I wonder if in this case social conservatism is one of those “social cancers” that Scott talks about in “Living By the Sword.” Being distrustful of foreigners gives short term advantage to groups that expand forcefully, but ultimately causes them to lose out due to all the conflict it causes.

          • FacelessCraven says:

            @Ghatanathoah “Wouldn’t a better solution just be to tell everyone, “Look, here are some values that are useful day-to-day life, here’s another set we need for emergencies, but are otherwise useless. Let’s keep both of them handy.””

            Well, for starters, when you lock the alternate power set away for “emergencies”, all you’ve actually done is shift the battlefield to a fight over what exactly constitutes an “emergency”.

            “Also, it seems like a lot of social conservative values are pretty useless, even in survival scenarios. What is killing gay people supposed to accomplish exactly?”

            I am pretty sure that killing gay people is not a normative example of a “social conservative value.”

          • jaimeastorga2000 says:

            What is killing gay people supposed to accomplish exactly? Wouldn’t gay people make useful hunters, farmers, and soldiers in a zombie apocalypse?

            “Killing gay people”? Way to fail the ideological Turing test. Try reading “On Homosexuality and Uranus” and “Roofs and Closets” to understand what actual traditional attitudes towards homosexuality were like. As long as you did your duty by getting married and having children, society was prepared to overlook a certain amount of private vice and indiscretion.

            If you really need them to reproduce you don’t need to force them to live a straight their whole lives. You just need to make them put up with straight sex until they’ve conceived a few kids, and then let them resume being gay. Maybe have gay couples and lesbian couples have special quad marriages.

            The distrust of foreigners value seems similar. It seems to me like xenophobic societies tend to be full of conflict, either internal or external depending on how homogenous they are, while pluralistic societies tend to be peaceful.

            Sounds plausible. And I am sure communism sounded plausible to the Russians and the Chinese, too. To quote Harry Potter and the Methods of Rationality, “It is only by harsh experience that we learn which principles take priority over which other principles; as mere words they all sound equally persuasive.” The conservative argument is that our ancestors have already determined which values and institutions are truly useful to civilization through generation after generation of bitter experience, and that we would be fools to throw away the wisdom they have passed down to us.

          • AlexanderRM says:

            An even better option than teaching two sets of values to alternate between would be to actually teach a form of consequentialist value system (in our case something like Utilitarianism, but traditional societies probably would have wound up with “what’s best for the tribe” instead) and explain to people which values are just situational heuristics.

            I think the problem with both these strategies is that nobody is actually planning out which values we ought to have, and that it’s hard to pass on nuance to an entire society in a political context. There’s a memetic selection effect, much like the way genes for “sugar tastes good unless you a ton of it” don’t do well in an environment where you never encounter too much sugar.

            (Of course, that doesn’t mean we shouldn’t try to add in said nuance, I’m just saying why we don’t have that nuance to begin with and also why it’s not emerging with liberal values. That might be helpful in some way to figure out how to get that nuance to pass along, actually, but I’m not sure how- obviously a unified consequentialist system would help a lot by reducing the amount of information that needs to be conveyed, and being a generally principle meaning it would see use fairly often in politics for some application or other.)

      • Randy M says:

        (I’m going to use up my allotment of unhelpful sarcasm here and basically illustrate your point while trying to slip in my own assumptions.)

        “like thinking murdering the weak is okay ”

        “Murder”? C’mon Dad, you’re totally privileging the hypothesis there. Your geneation considered it murder, but Grandpa’s generation was against reproductive choice and euthanasia, as well. You just don’t support reducing environmental and social tension because you are so reactionary.”

        • AlexanderRM says:

          Personally I’d like to say that if my children or grandchildren were able to come up with a good, universalizable justification (like utilitarian, rawlsean or the like) for murder along some lines, then I’d at least listen to that and take it seriously, on exactly the same grounds as I do when conservatives come up with such justifications (ex. war, capital punishment, police shootings, letting poor people die from lack of money).
          A justification *regardless of circumstances*, I can’t picture except with a corresponding acceptance that it’s OK for others to try to murder them, which I’d probably still disagree with.

      • DavidS says:

        Fair. I think what I was thinking was more that ‘being tough and good at fighting’ or ‘clawing your way up the socioeconomic ladder’ look from the outside like things that are obviously helpful practical values for newcomers, but their benefits for later generations will be less. I guess I’m sort of assuming that the nomads/tiger mom’s values aren’t terminal values.

    • At a possibly irrelevant tangent … . Ibn Khaldun pointed out the pattern long before the Manchus. It was a driver of his cyclical model of history.

      • AlexanderRM says:

        General on this subject, but this seems like as good a place as any to post it- I’d like to point out that both the Mongol Yuan Dynasty *and* the Manchu Qing were brought down by internal revolts of native Chinese, not nomadic invasion. In fact, while I haven’t checked over every early Chinese dynasty in detail, the only example I know of where a nomad dynasty was conquered by a second nomad invasion was the Jin Dynasty, and they 1. had only ruled for 100 years by that point and 2. hadn’t finished conquering China (just the northern half while the Southern Song held the rest), so they probably hadn’t descended fully into decadence yet.

        More generally, from a wide variety of historical examples I know of, I’d like to propose an alternate model of empires rising and falling, although I don’t have a catchy name for (the “chance model”?). Every dynasty or state over the course of several centuries will eventually encounter problems such as bad rulers (or a few in a row), external threats, or various other factors, and if they’re either bad enough or combine with multiple bad factors at once, the empire will fall. Thus, over a long enough period, by chance alone every empire will eventually fall.

        I don’t think the cyclical model is entirely unfounded, but this definitely explains history at least as well at that does, and certainly a really hard, inevitable cyclical model (“empires last about X number of years” or whatever) doesn’t match with history.
        There’s certainly something to the general principle of “Group has some advantage, group uses that advantage to expand, group loses that advantage “, but I think that’s actually more common with the advantage spreading to the conquered territory than it is for the conquerors to lose the advantage- mainly in various forms of technology, most famously European colonialism, along with things like the Muslim religion.

    • Ghatanathoah says:

      The thing that bugged me about this is that Christianity does not actually provide any good reasons for why people should care about these virtues that atheist children will supposedly lose. It seems to me that the reasons atheists usually give to follow these virtues are no worse than the ones Christians give, and often better.

      I think the problem is that people with a certain authoritarian mindset often make some sort of huge and unwarranted leap of logic that something being commanded by God is somehow a better reason to do things than one’s own conscience and experience. I think they sort of fetishize obedience to authority, and have trouble comprehending the idea that a person could rely on their own conscience and experience to be good without having an external authority telling them what to do. I think it’s not a coincidence at all that these same people are often the ones who bend over backwards to defend police brutality, get furious at people who burn flags, and otherwise venerate authority figures to a ridiculous degree. Often they try to justify their response by referring to Hayekian rule of law, but as soon as one of their venerated authority figures breaks the law they bend over backwards to defend them.

      I personally have always had the exact opposite intuitions as these people have. Even when I was at my most religious I thought that morality can’t come from God because then it wouldn’t be objective morality, it would just be some dude’s preferences. The fact that that dude had supernatural powers was irrelevant, his powers didn’t make his viewpoint any less subjective. I couldn’t understand why people thought Euthyphro’s dilemma was a dilemma at all. It was obvious to me that the answer was that the gods love it because it is good.

      But then, I’ve noticed at other times that my sense of metamorals is different from a lot of people. For instance, I consider motivational externalist accounts of morality to be obviously true, and motivational internalism to be obviously insane. Other people seem to have such strong intuitions in favor of internalism that they consider refuting internalism to be the same as refuting morality.

      • houseboatonstyx says:

        @ Ghatanathoah
        Even when I was at my most religious I thought that morality can’t come from God because then it wouldn’t be objective morality, it would just be some dude’s preferences. The fact that that dude had supernatural powers was irrelevant, his powers didn’t make his viewpoint any less subjective.

        Bingo. I have seen a serious literate Christian theologian post exactly the same thing, under the heading “God’s Subjective Values”. CS Lewis too finds it an important point, so apparently it’s canon.

        • The way I like to put the argument is that, if you have no independent basis for judgements of good and evil, how do you distinguish between God and the Devil? What is your basis for considering a particular very powerful supernatural entity good, hence wanting to do what he tells you (for any reason other than fear of punishment or hope of reward)?

          This is, of course, a very old issue. A major Islamic controversy more than a thousand years ago was whether humans had any basis for moral judgement outside of revelation—mutazilite vs Ash’arite.

          • FacelessCraven says:

            At the severe risk of exposing my own ignorance, wouldn’t the Good entity be the one whose values have consenance with our own, and with the observable order of the world we live in? I realize that creates problems with establishing what values the God or Devil hold, hence revelation, and the debates get hairy from there…

            But in an effort at a clearer picture, we all broadly agree that Elua and Moloch are not real Gods, yes? We assign values and draw distinctions between them entirely via reason, not brute revelation. And yet there is no real disagreement that Moloch is the enemy to be destroyed and Elua is the God to be honored and worshipped and served, yes? A people may worship Molochj blindly or out of despair, but surely no society who could trivially escape Moloch would choose to stay with him out of affection. Wouldn’t that seem to indicate that on some level, God and the Devil might be distinguishable in a rational manner?

          • Randy M says:

            If the God who made the universe is the same as the one giving you the rules, unless that God is a sadist, he rules will help you in some way on net, at least to the extent that you want out of life what the designer of the universe engineered it to provide (I understand some bugs may have been introduced at some point, or at least some features are poorly understood).
            Kind of like listening to the tech support who tells you you need to stop downloading free stuff from questionable sites no matter how appealing.

            If you are not happy with some of the rules or results of following the rules, you may label the maker and/or revealer (to the extent you believe in one) evil, but at the same time, consider the posibility that you do not have the foresight or understanding as He.
            Or said another way, we don’t have to necessarily presuppose that God = Good, but we should have strong priors that way.

          • keranih says:

            if you have no independent basis for judgements of good and evil, how do you distinguish between God and the Devil?

            CS Lewis goes into this rather clearly (to my read) in Mere Christianity. His writings are worth checking out.

          • houseboatonstyx says:

            @ keranih
            CS Lewis goes into this rather clearly (to my read) in Mere Christianity. His writings are worth checking out.

            Better in his serious apologetics, especially in _Reflections on the Psalms_. Istr, but I may have wishfully edited, his “There arose terrible theologians who taught [the fiat doctrine]”. Concerning which he had some quite strong words.

          • Samuel Skinner says:

            “CS Lewis goes into this rather clearly (to my read) in Mere Christianity. His writings are worth checking out.”

            According to wiki
            -He bases his case on a moral law, a “rule about right and wrong” commonly known to all human beings, citing the example of Nazism; both Christians and atheists believed that Hitler’s actions were morally wrong.

            Wiki might be summarizing wrong, but the example is poor- Hitler and a substantial portion of Nazis and Germans thought they were doing the right thing. As for universal, I’m pretty sure there are individuals who do not know any moral law, but I don’t know the clinical definition.

            “Better in his serious apologetics, especially in _Reflections on the Psalms_. Istr,”

            Okay, online pdf

            Which section?

          • Matt C says:

            > And yet there is no real disagreement that Moloch is the enemy to be destroyed and Elua is the God to be honored and worshipped and served, yes?

            I don’t think Moloch can be destroyed as long as there are independent agents who sometimes wish to compete with one another. “It” is just a sometimes consequence of competition, and can’t be destroyed without destroying independent agency altogether.

            I find the Elua idea horrifying. A totalitarian god who locks all of humanity in a box and keeps them at the status quo of 2075 forever.

          • FacelessCraven says:

            @Matt C – Perhaps “destroyed” was too strong a term. Opposed? Shunned?

            As described in “Meditations on Moloch”, I generally see Moloch as the representation of defection, conflict, hatred and strife, and Eula as the representation of cooperation, harmony, love and peace. I don’t think questions about any specific implementation really change that much. Ignoring the question of FAI, don’t we all feel that Eula-like values are good and should be promoted if practical, and moloch-like values are evil and should be opposed if practical?

          • Matt C says:

            I think Scott is more specific about Moloch. “The opposite of a trap is a garden. The only way to avoid having all human values gradually ground down by optimization-competition is to install a Gardener over the entire universe who optimizes for human values.”

            In this sense, Moloch is arms-race type competition, or to some posters here, any competition at all viewed from the perspective of the loser.

            Sure, it can be good to stop arms races before they start. Sometimes this can be negotiated. Stepping on potential competition and claiming you’re doing it for the good of all is something else, though.

            > I generally see Moloch as the representation of defection, conflict, hatred and strife

            Well, it is nicer when we can all get along, but sometimes we can’t. Whether peace is better than strife, or not, depends on the specifics.

  56. suntzuanime says:

    I guess I don’t know what I was expecting with a link to Vox, but a lot of the questions on that political bias quiz seem like the sorts of things that there isn’t a known neutral answer to and claims that there is such an answer are themselves indicative of political bias. Like if you know how to definitively prove an answer to the question “did the Obama administration’s 2009 stimulus package reduce or increase unemployment?” you can probably go ahead and collect your fake economics Nobel Prize.

    “The average US citizen would be better off if a larger number of highly educated foreign workers were legally allowed to immigrate to the US each year.” jesus fucking christ if you can even rigorously define “better off” you’ve made a major philosophical breakthrough. This is what really bugs me about Vox dot com, the baked-in epistemic arrogance. They think they can link to a survey of a few dozen selected “experts” and treat the majority response as truth.

    (This is not my political bias talking; my own best guess agrees with them in both cases, just not with the level of certainty they’re willing to use.)

    • Allan53 says:

      Also seemed intensely US-centric, I ended up guessing the vast majority blind because I’m not from the US and as such I don’t know most of those figures.

      Plus, they were, at least, incredibly arguable. Like the “scientists agree global warming is due to human action” one. The 97% consensus idea has been, at least, shown to be arguable and it’s not impossible the science is biased from the get-go .

      So someone could take a “we don’t know yet” position entirely reasonably, without any kind of bias.

      • Cerebral Paul Z. says:

        I didn’t have any problem with that particular question, since the proposition given was so anodyne that there surely is a 97% consensus (at least) in its favor. The arguability only comes in if you take that 97% figure out of the “humans have warmed the climate at least a little” motte and into the “Earth will turn into Venus later this century unless we adopt the entire Greenpeace agenda *right away*” bailey.

        • Sigivald says:

          Absolutely true!

          Problem being I have little faith in how Vox will interpret the question, as motte vs. bailey.

      • “scientists agree global warming is due to human action”

        What does that even mean? I doubt anyone in the field claims that it is 100% due to human action, since we know global temperatures have changed in the distant past for other reasons. The 97% figure from Cook et. al. 2013 was for humans as a cause—one example of a paper that qualified used the terms “contribute to.”

        • Adam says:

          I don’t really feel like taking the quiz again, but I believe he’s misquoting it and it only asks about human action contributing, not being the sole cause.

          • Allan53 says:

            You’re right, I apologise. The actual question is “Have average global temperatures risen due to human activities?”

            So, yes, I misquoted it. But the thrust of my point still stands, albeit kind of weakened; if the consensus of scientists is arguable, then how am I, Mr Regular Person, supposed to know? I would think in that case “We don’t know” is really one of the better answers than the definitive “yes” or “no”.

      • xq says:

        The BBC link does not at all support the claim that “the science is biased from the get-go.” It’s making the opposite claim.

    • Scott Alexander says:

      I thought under the circumstances they did a good job focusing on things where there was a decent level of academic agreement, and at least they didn’t seem biased to either side.

      • Adam says:

        My only complaint right now would be the effect of extended unemployment benefits question. The papers they linked to studied interstate differences as a correlate with how long each state extended unemployment, but the way the question was worded sounded like they were asking whether extending unemployment benefits would correlate with the rate within the state subsequently rising, not correlate with having a higher rate than states that did not extend benefits. On the second question, of course the answer almost has to be no, because benefits are extended in response to some exogenous shock causing an extremely high rate which is going to go down in the near term no matter what you do because of regression to the mean.

        I did get 0% bias, though, so thanks to them for flattering me.

      • MichaelT says:

        No, the stimulus question was ridiculous. I don’t understand how most economists are reasonably sure that the stimulus decreased unemployment when actual unemployment exceeded the economic models’ prediction of what it would be without stimulus.

    • dndnrsn says:

      It seemed odd to me that the quiz equates not knowing the answer with being biased. I’d complain about it being US-centric, but it’s not like I lack for exposure to US media, and I’d probably not know the answer to the same questions regardless of country.

      • Adam says:

        That isn’t what it’s testing, though. I only got 11 out of 18 correct, but got a 0% bias score. It’s testing whether there is a pattern to your wrong answers such that you tend to be wrong in a way that agrees with what you stated your politics are at the beginning of the quiz. You can get nothing correct at all but still be unbiased. If you’re wrong, but every wrong answer just so happens to align perfectly with a particular party platform, that’s not likely to be an accident.

        • dndnrsn says:

          Point taken.

          I got more correct and less biased than most, but whenever I didn’t know the answer I just picked “don’t know” or whatever seemed like the split-the-difference option.

          If I understand it correctly, someone who got them all right wouldn’t be assessed to be biased, though, regardless of the actual reason they chose the answers they did?

  57. suntzuanime says:

    Most instances of “I can’t afford X” are “I don’t want to pay for X”; consider the unspoken last half of the sentence to be “I can’t afford X (without sacrificing some priority which is more important to me on the margin)”. Perhaps it would be easier to think of it as “I can’t afford X (under the default shared assumptions of this conversation)”. If I were prepared to move heaven and earth I could afford my own apartment, but I’m not, so I live with roommates because I can’t afford my own place.

    In particular it sounds like what Harvard doesn’t want to do is spend more money on the library. So the people running the library are complaining of being unable to afford the closed journals, because they represent the institutional priorities of Harvard that form the basis of the conversation. This is one of the ways organizations simulate actual thought; the budget allocations give the organization a sort of synthetic set of priorities.

  58. JE says:

    Appearantly I have zero political bias… More likely, I’m european and so my political bias doesn’t map neatly onto either side in american politics.

    • MichaelT says:

      No, it probably means you lean left, especially on economics. I don’t understand how most economists are reasonably sure that the stimulus decreased unemployment when actual unemployment exceeded the economic models’ prediction of what it would be without stimulus.

      • Scott Alexander says:

        Eh, maybe so. I was impressed that they (correctly IMHO) said that there was no evidence banning guns changed the crime rate, even when other parts of Vox had published poorly-thought-out arguments that it did, and figured that meant they were more honest than usual and I could trust them even on things I didn’t know about.

        • Izaak Weiss says:

          I’m confused about that. I mean, isn’t the violent crime rate much lower in countries where owning a gun is basically impossible? And don’t we see drops in violent crime in those countries at the same time the gun laws were introduced? That seems like pretty important evidence.

          • This just in: it’s possible to commit acts of violence even without a gun.

            If you only count homicides, then gun control might(?) actually help, since violence without firearms is less likely to be lethal. I don’t know what the stats are.

          • Nornagest says:

            Stats on this are fiendishly difficult to interpret, because it’s basically impossible to make it sufficiently hard to get guns on anything less than a country-wide scale and that introduces all sorts of cultural confounders. You can ban gun sales or institute safety regulations on a city or province scale, and that gives us better data on those sorts of interventions, but you can’t ban guns.

            About the closest thing we’ve got to a natural experiment is the UK, and that has much lower murder rates than the US (the difference is less extreme if you apply various demographic controls, but it’s still there) but much higher rates of assault and a number of other violent crimes. The incentives check out, so I kinda think that might be representative, but it should definitely be taken with a grain of salt.

            We have much stronger evidence that making guns marginally harder to get doesn’t do much to the marginal crime rate, but that doesn’t tell us much about the overall shape of the graph.

          • Orphan Wilde says:

            It’s impossible to say even whether or not the UK has a lower murder rate than the US, because of radical differences in the way the two countries count homicides. The US categorizes suspicious deaths and “justified homicide” as homicide; the UK categorizes some types of “justified homicide” as not-homicide, and usually, if I understand their court system correctly, doesn’t even declare a death a homicide either until an arrest has been made, or until a conviction has been reached.

            It’s like infant mortality rates. The countries are counting different things.

          • John Schilling says:


            The violent crime is much lower in countries where owning a gun is basically impossible (e.g. Japan), and it is much higher in countries where owning a gun is basically impossible (e.g. Mexico), and it is much lower in cases where owning a gun is basically mandatory (e.g. Switzerland), and it is much higher in cases where owning a gun is basically mandatory (e.g. Yemen). Other variables dominate, including the ones that mindkill almost everyone who tries to discuss the subject.

            In none of the cases I am familiar with, and none of the cases generally cited for the superiority of gun control, was there actually a significant drop in the violent crime rate at the time the major gun control laws were enacted.

            That seems like pretty important evidence

            If you have to start by asking “isn’t it the case that…”, then what you have is not evidence.

          • Cauê says:

            Scott’s classic post on it:

            (my favorite part is the observation that the US homicide rate without guns is higher than the UK’s total, guns or not)

          • Mary says:

            ” And don’t we see drops in violent crime in those countries at the same time the gun laws were introduced?”

            The very opposite of the truth. We see increases in violent crime when gun laws are introduced. After all, in the 19th century, British crime was a fraction of today’s with no gun control law at all.

          • Mary writes:

            ” After all, in the 19th century, British crime was a fraction of today’s with no gun control law at all.”

            That struck me as a little surprising, so I went looking for data, and found them at:


            Using their figure for total violent crime, there were about four thousand offenses a year at the end of the 19th century, about 800,000 a year at the beginning of the 21st. Almost 100,000 of those are “harassment,” which is zero in the early years, presumably because there was no such crime, but that still leaves an increase by a factor of about 175 over a period when population less than doubled.

            Looking at individual categories, murder, attempted murder, and “more serious wounding or other acts endangering life” all increased by more than population, the last category enormously more.

            The reason I was surprised is that this is the opposite of the very long term trend, at least in murder.

          • John Schilling says:

            The reason I was surprised is that this [very high UK violent crime rate] is the opposite of the very long term trend, at least in murder.

            My interpretation of this is that the UK has for two or three generations held to a tacit agreement between police and criminals that, so long as the latter refrain from actually shooting people, the former will write off their lesser crimes as “boys will be boys”, “yob culture”, or otherwise part of the natural order of things and not worth more than a slap on the wrist. Criminals who do go around shooting people, OTOH, get to be the focus of all the police attention that isn’t being devoted to all the lesser crimes, including all the guns that the police still have left.

            This worked well enough until the inevitable, exponential growth of not-actually-shooting-people crime started to reach intolerable levels. Including smart-ass criminals saying essentially, “so you’re OK with us stabbing people to death, so long as we don’t shoot them, right”, and the police comically backpedalling on that one and calling for a nation free of pointy knives.

          • Nornagest says:

            There was even a Clash song that touched on that dynamic:

            You know it means no mercy
            They caught him with a gun
            No need for the Black Maria
            Goodbye to the Brixton sun

            …though punk rock sociology should probably be taken with a grain of salt, too.

          • Sastan says:

            Others have covered this, but to answer your two questions:

            1: Sometimes. Sometimes not. Some countries have massive gun ownership and almost no violent crime (Finland, Switzerland), some have tons of it (Yemen). Some countries have almost no guns, and very little crime (Denmark) and some have tons (Jamaica, Mexico).

            2: No. You don’t. Gun crime is up in the UK since the bans. The US Assault Weapon ban of the ’90s had no effect whatsoever (likely because those weapons, despite a few outliers, are almost never used in crimes). Australia has seen a decrease, I think. You most certainly cannot show a strong causal link between Gun Ban—>Less Violent Crime.

          • Phil says:

            I’m suspicious of comparisons between C19th crime rates and C21st ones. Between cultural differences (the past is another country and all that) and data recording changes, without digging very deeply into the figures I wouldn’t put a lot of weight on them.

            Murder numbers? Sure – it’s hard to hide a body after all & people tend to have relatives & neighbors who notice their absence. Lessor forms of violence? A lot of the time that doesn’t get recorded at all even today, so the recorded level has as much to do with cultural factors as it does the actual underlying level of violence.

            Look at the rape figures in the dataset linked to by David for the strongest example of this: does anyone really believe that the level of rape in the C19th was 50 times less than today?

          • chaosmage says:

            Scott’s classic post on gun control is excellent as usual. I smirked at its (entire) answer to the argument that citizens bearing arms can’t fight tyrannical governments because they have jets and tanks:

            > I’ll just answer this one by mentioning that Bashar al-Assad also has fighter jets and tanks. Ask how that’s been working out for him.

            A couple of years of horrendous civil war later, with Assad still in power, it does appear jets and tanks are useful. If maybe not as useful as friendly relations with UN security council members.

          • Mark says:

            >. I mean, isn’t the violent crime rate much lower in countries where owning a gun is basically impossible?

            They’re talking only about the impact of concealed carry laws, which you can look at independently of laws regulating ownership.

            When you compare between US states and cities, regions where concealed carry is legal tend to have significantly lower crime rates than regions that don’t, as any number of pro-gun people on the internet will tell you at length.

            The tricky part is disentangling that from the fact that these regions also tend to be whiter and less urban, so I tend to agree with the test-maker that you can’t make a firm conclusion either way.

          • To un-dangle the comparison, the UK homicde rate is about 1/3rd of the US, but the affray/assault rate is higher.

          • It would probably be very revealing if someone could distinguish between rural-style and urban-style gun ownership. (The Swiss are another matter again).

            And while we are on the subject, crime rates should be looked at in terms of population *density*.

        • Jon Gunnarsson says:

          As far as I understand, the makers of the quiz are not affiliated with Vox and Vox merely wrote an article about the quiz.

      • XerxesPraelor says:

        The models being wrong is the whole reason why the collapse happened, so you can’t trust them.

      • Desertopa says:

        Measuring zero bias doesn’t require you to get a large proportion of answers right, it just requires that your wrong answers be uncorrelated.

  59. BD Sixsmith says:

    Very nice small study with neat methodology finds that more intelligent people are more honest.

    And better liars.

    • Desertopa says:

      In practical terms, not just the artificial terms constructed for the experiment, the two are strongly correlated. People who demonstrate honesty when they have an incentive to lie build up credibility and make it easier to sell lies in the future, while people who lie with little provocation use up the trust necessary to sell even plausible lies.

      In my experience working with kids and teenagers, some of whom are stunningly dishonest, the less intelligent among them are less likely to recognize trust as a resource that can be cultivated or spent. The most dishonest among them don’t seem to understand that they might ever have an incentive to tell the truth when having people believe a falsehood would be in their interests.

  60. I agreed with a good deal that Oliver Lee wrote, but I was left with one question. Does he feel that his own academic work, research not teaching, is work worth doing or only a ritual, make believe, needed to fit into his academic role?

    • Sniffnoy says:

      Yes, I noticed that too — almost everything he says is about teaching (or sometimes outreach, a more public form of such) rather than research. He concludes he’s “doing nothing meaningful with his life” because people don’t care about the teaching. So I’d like to echo your question: What about the research? Why does he make so little mention of that, and does he really consider it worthless?

      • John Schilling says:

        He’s an academic historian, and I’m having trouble finding anything he’s done on Google Scholar. Possibly curating that museum exhibit about professional wrestling was the high point of his research career, or at least the part that he felt might be interesting to the not-academic-historian audience.

  61. You link to a discussion of the way in which values erode over the generations. One version of this that has occurred to me involves the welfare state. Start with a society where there is relatively little redistribution. Getting a job and working hard is the obviously right thing to do. One result is that people who fail at it lose status and feel a loss of self-respect.

    Now introduce levels of redistribution high enough so that, for a significant number of people, it isn’t clear whether the additional income by working is worth the loss of leisure. People still work, because they know that to be unemployed is to be a failure, in their own eyes and those of others. Over time, that attitude erodes. Why work hard at a boring job when you could be spending your time vacationing in low cost parts of southern Europe?

    • Nita says:

      people who fail at it lose status and feel a loss of self-respect

      … not to mention hunger, cold and the unpleasant itchiness from going months without a shower 🙂

    • TheAncientGeek says:

      Then you end up with the peopke who don’t want to work .not worling, and the
      people who do, working. Which is fairly desirable from a utilitarian peperspective, bearing in mind that modern economies dont want or need full employment.

      • Ever An Anon says:

        Even assuming for the sake of argument that we can in fact support a large (let’s say low double-digit) parasite class without adversely affecting the economy, that is not a sustainable situation in the long term.

        Firstly, because the size of that group can only increase. Either because they have more children who follow in their footsteps, because more immigrants (legal or otherwise) will show up to claim the free benefits, or because productive people on the margin see the futility of working when they can get a similar deal by joining the parasites. Unless you believe that a small productive enclave can support an unlimited number of parasites the system must eventually collapse.

        More importantly, why would the productive people continue to support them? Either they have no choice, because of a threat of state violence or revolution otherwise, and you have just reinvented the slave society or they can withhold the money that would be redistributed and the redistribution stops.

        I’m not an Objectivist or something: communities should take care of unfortunate people. In fact it’s essential for a functonal society. But a welfare state that large cannot possibly survive and it just might take its nation down with it.

        • meyerkev248 says:

          FWIW, we have:

          * A population of about 320 Million
          * About 140 Million workers, of whom 115 Million are full-time.
          * About 60 Million people on Social Security/Medicare aka the Elder State.
          * About 14 Million people on either SSDI or SSI.
          * 103 Million are receiving federal welfare benefits. This doesn’t count Social Security or Veterans Benefits, which brings the number up to 153 Million.

          In other words, any “We can’t handle low-digits on welfare without a death spiral” needs to accept that we’re already there, and thus needs to argue that we’re already well up the death spiral.

          Even just using SSI/SSDI over total workers gets you 10%.

          • brad says:

            I’m curious as to what “federal welfare benefits include.” Medicaid has 71 million enrollees. So you’d need to find another 32 million that don’t qualify for medicaid but qualify for something else. I can’t think of what program that would be.

          • Adam says:

            Pell Grants and Stafford Loans? Federal extensions to unemployment insurance? Also, WIC goes up to 185% of Federal poverty level. Medicaid expansion only goes up to 133% and not every state accepted the expansion.

          • meyerkev248 says:

            It doesn’t really affect the argument (ie: It’s trivial to get to low single digits), but I’m curious myself.

            “109,631,000 Americans lived in households that received benefits from one or more federally funded “means-tested programs” — also known as welfare — as of the fourth quarter of 2012, according to data released Tuesday by the Census Bureau.”


            Looking at Q4 2012…

            They’re saying 83 Million on Medicaid, 22 Million on WIC, 20 Million on SSI, 13 Million on Section 8 and variants, and 51 Million on Food stamps. I’m presuming that there’s some obvious overlap there.


          • Adam says:

            It could also be people eligible for Medicaid but not using it, in addition to the WIC users not eligible for Medicaid. Notably, if you’ve received SSI your entire life, you can still keep getting it even when you’ve reached Medicare eligibility age and don’t need Medicaid any more, though it doesn’t seem like that would cover very many people. You have to be pretty damn disabled to receive SSI your entire life. Also, Section 8 is tied to the local median income, so you can be eligible for that while being way above 133% of federal poverty line if you live in a very expensive city.

          • brad says:

            WIC probably accounts for some of the non-medicaid people, but I think you automatically qualify for medicaid if you qualify for SSI. Food stamps seems to have a similar scaling to medicaid, but in states that didn’t expand medicaid you have singles that qualify for SNAP but not medicaid, so that’s another sliver. Never bothered figuring out what the rules are for section 8 given that as a practical matter it isn’t open to new applicants. Maybe there’s something there. Still hard to see how you get 32 million out of these. (the medicaid number is from here BTW:

            Pell and Stafford make a lot of sense, I didn’t think of those at all. With a two kids in college you can be quite well off and still have a child eligible for subsidized Stafford loans.

            You can be on medicaid and medicare at the same time. In addition to older people, another way to get in that situation is after two years on SSDI you are eligible for Medicare regardless of age.

            When you have both, medicare is primary and medicaid pays co-pays and covers the gaps. It’s a rather nice healthcare set up — medicare means that doctors will actually see you, and medicaid means you are out of pocket almost nothing.

          • Whatever Happened to Anonymous says:

            Wait, where are the school aged children in that number?

        • lmm says:

          In a country with state-provided education there’s no reason that the children of those who don’t work would necessarily not work.

          A small productive enclave can probably support close to unlimited others. As long as productivity grows faster than population it should be fine, no?

          Of course a welfare system can only survive if it has popular support (or at least support among the people’s trusted representatives). But that’s true of any government policy. Productive people would be willing to pay taxes to support the rest of the population for much the same reasons they do now. Or they won’t, and we won’t adopt that system.

          • keranih says:

            In a country with state-provided education there’s no reason that the children of those who don’t work would necessarily not work.

            Postulate: That “employability” and/or “desire to find gainful employment” is based, in our current situation, more on cultural/home-taught behaviors than it is on level of certification of formal education, at least in the lowest quartile.

          • anon says:

            “In a country with state-provided education there’s no reason that the children of those who don’t work would necessarily not work.”

            The reason is genetics.

        • Anthony says:

          Even assuming for the sake of argument that we can in fact support a large (let’s say low double-digit) parasite class without adversely affecting the economy, that is not a sustainable situation in the long term.

          Oh, having only a quarter to a third of the population working is plenty sustainable, so long as the system doesn’t provide incentives which will lead to the proportion of people working to decline over time. The social system of the 50s, where women were only supposed to work in marginal or reserved jobs, and weren’t supposed to work for a big chunk of their adulthood, was sustainable. The system gave a strong incentive to adult men to work – having a job and decent prospects was a prerequisite for getting married, and discouraged others (women, teenagers) from competing with adult men, by making it not just acceptable but encouraged for those others to not work.

          • “The social system of the 50s, where women were only supposed to work in marginal or reserved jobs”

            Women were expected to work in the job that has been dominated by women for a very long time, and historically absorbed a large fraction of the labor force—household production, in particular but not exclusively producing and rearing children. “Not working as an employee for pay”≠”not working”

          • Who wouldn't want to be Anonymous says:

            Also keep in mind that, for example, up until maybe WWII in the UK, the second largest employment category (after agriculture) was domestic servants.

            Household production used to be an enormous labor sink. It employed for pay servants and the woman of the house.

            Increases in productivity has made domestic work less labor intensive and is generally be done by one or two people working part time now. Freeing them to participate in other sectors of the labor market with the remainder.

        • TheAncientGeek says:


          The sysstem needs a certain number of working age people to not be working , no more and no less. If the number becomes too high. It can be adjusted by raising wages or cutting welfare.

          Its in the interests, up to a point, . of currently working people to support the welfare state, because they might get downsized themselves. That’s more like insurance than slavery.

          You say you are not an Objectivist, but it is typical of them to tacitly assume that society is divided between people with jobs for life, and people who never work.

      • Jaskologist says:

        The idea that most people want to work, and particularly that they also want to support a bunch of layabouts, strikes me as absurd. I gain no utils from having welfare recipients and old people take money that I would rather spend on my children. I’d be better off if they got a job and contributed to society (even if it was a small contribution) instead of draining from it.

        I’m a pretty productive citizen, and I don’t even want to work, I’m just willing to do so in order to get other things that I do want. Society is wise to make sure that I produce in order to get what I want.

        • anon says:

          The idea is that there are more job-capable people than jobs that need doing, because industrialization and robots and whatnot. This may or may not have caused society to invent busywork, aka jobs that aren’t useful but keep people that need employment, employed. And from the working people, some want a job for money only (group A), while others have other reasons – enjoyment, structure in life, status, whatever (group B).

          If group A get welfare, group B gain because they now have a much higher chance of getting the job they want (less competition) and maybe they also get a lot of status since they’re suddenly members of that super charitable group that keeps society afloat – which is extra nice of they want to work because its a high-status thing. And maybe some people in group A produce art with their newfound free time that group B appreciates, who knows.

          The obvious questions are, have we reached the point where the jobs that need doing are few enough that group B occupy them all (if not, when will we?) And, would such a system continue giving birth to enough people of group B to maintain itself, or will a culture with 90% slackers start deriding the voluntary workers instead of respecting them, leading to less and less voluntary workers and collapse?

          • Jon Gunnarsson says:

            There is no such thing as a job that needs to be done unconditionally. Rather, a job needs to be done if you want to produce a particular thing that people want. Since human wants seem to be virutally without limit, there is virtually no limit to how many people can be usefully employed. First Worlders today, despite living in luxury unfathomable to the ancients, still desire a great many things—bigger houses, tastier foods, better schools, more diverse entertainment, faster computers, and on and on it goes.

            As long as these desires exist and can be supplied by human labour, there will never be a shortage of work to be done. Maybe these conditions will cease to apply at some point in the future, but we are not remotely close to that point yet.

          • “The idea is that there are more job-capable people than jobs that need doing”

            I don’t know what “jobs that need doing” means.

            Of possible relevance … . The estimate by economic historians is that real per capita income in the developed world is currently twenty to thirty times what it was for most of the world through most of history. For a more recent datum, from Mao’s death to 2010, real per capita income in China went up twenty-fold—and China is still, by our standards, a poor country.

            By either of those measures, we already do many times more jobs than “need doing.” But by the measure of “are there goods and services people would like that they don’t have,” we are nowhere close to doing all the jobs that need doing.

          • Randy M says:

            “As long as these desires exist and can be supplied by human labour”
            I feel like you are eliding a lot here by assuming human labor = human labor.
            I suspect there are a great many people in the world who cannot meaningfully contribute to tastier foods, better schools, more diverse entertainment, or faster computers.
            And the number may well grow when we compare the contributions of others against the best new automated processes for any of those things.

            This is the one argument that tempts me towards socialism, or at least the UBI talked about around here.

          • Jon Gunnarsson says:

            @ Randy:
            To take one example, people could get tastier food by hiring a cook. The reason very few people today do this is not that microwave ovens, frozen dinners, and take out have made personal cooks obsolete, but rather that labour today is very expensive. Becoming a competent cook requires some training, but it’s well within the means of most people.

        • TheAncientGeek says:


          If most people don’t want a welfare state, why dont they v vote against it?

          • John Schilling says:

            Because we don’t live in a direct democracy where people can vote for or against specific policies or institutions like the welfare state, and because no electoral system incentivizes politicians to perfectly align their platforms with the desires of the mean or median voter.

            Whether it is “most people” or just “many people” who don’t want a welfare state, there are no serious, electable candidates who are promising to eliminate the welfare state.

            Even if it is “most people” who don’t want a welfare state, it is a Bad Plan for a serious candidate to promise to eliminate the welfare state. Because each of those voters want nine other things as well, but they don’t agree on those nine. Since you can’t offer them all everything they want, the winning strategy is to not offer any of them anything so offensive as to make them forget that you agree with them on the other nine issues. Instead, offer up watered-down inoffensive versions of most of the things they want, which in this case means restricting benefits and cracking down on fraud.

          • TheAncientGeek says:

            Proves much too much. You could say that any number of things. Also, people do have the opportunity to vote for welfare trimming parties, and if you do that consistently enough, welfare goes to zero…but people dont do that consistently, reveling their preferences.

          • John Schilling says:

            People do vote for welfare trimming parties. But that doesn’t consistently require that welfare go to zero unless welfare is the only policy dispute at hand (or that everything else perfectly correlate , and that everybody’s welfare utility function be monotonic as it approaches the zero-welfare limit. Neither of these are the case.

      • “Then you end up with the peopke who don’t want to work .not worling, and the
        people who do, working. Which is fairly desirable from a utilitarian peperspective, bearing in mind that modern economies dont want or need full employment.”

        There is a sense in which modern economies do not need full employment—if per capita real income was cut in half it would still be much higher than in most past societies. I’m not sure if that’s what you mean,

        From a utilitarian standpoint, the right rule is that you choose labor over leisure as long as the marginal utility of what your labor produces is greater than the marginal disutilility to you of working (aka utility of leisure forgone). A market system with zero redistribution doesn’t do a perfect job of that because marginal utility of income varies among people, but wouldn’t you expect it to come closer than a system in which I get the utility of leisure and someone else pays all of the cost of my consuming without producing?

        • You have already given the first third of the answer: in modern economies, the income level is so much above the comfortable susbsistence level that paying a portion of it in tax is not a major loss of utility. (People who object usually object for political reasons, rather than being pauperised).

          The second third is that subsidising welfare is not a nett loss for workers, since they are insuring themselves against future sickness and unemployment.

          The remainder is that you are not going to have a highly efficient and flexible labour market or organic reasons. are not going to have brain surgeons working a few hours a week, even if they ant to, you are not going to have a huge demand for blacksmiths even if it is a popular choice.

      • Tracy W says:

        It seems quite plausible to me that if you have to work then you generally have to do some getting along with people to get paid. If you don’t have to work you lack that carrot for sociability, compromise, etc. And all that society is left with is punishment if your anti-sociability gets too out of control (oh, and whatever socialisation was managed by Sesame Street and the like.)

    • Anonymous says:

      Why work hard at a boring job when you could be spending your time vacationing in low cost parts of southern Europe?

      In an economy where jobs are scarce, this is arguably a good thing. People who can live low-cost do so, while those with families to support face less competition for jobs.

    • grendelkhan says:

      Why work hard at a boring job when you could be spending your time vacationing in low cost parts of southern Europe?

      You have a very surprising model of what poverty looks like in the first world. I’ve seen claims that welfare recipients live high on the hog, but never this extravagantly.

  62. g says:

    The “three girlfriends” article is in a tabloid newspaper (hence kinda dubious from the outset) and doesn’t even claim that in Dongguon “the average man has three girlfriends”. It says that some men have two or three, and it finds exactly one example of a person with three girlfriends, who says it’s kinda pathetic to have only one, and that’s it.

    I had a quick look for actual sex ratio information and unfortunately the only thing I found was from another even less reputable UK tabloid. But, for what it’s worth, it claimed the male:female ratio was 89:100. Unless that’s very, very, very wrong it cannot possibly be true that the average man in Dongguon has three, or even two, girlfriends.

    • Tracy W says:

      Surely the plausibility of the average man having multiple girlfriends depends on the number of boyfriends each woman has?
      (My favoured solution to the double-sexual-standard problem in Elizabethan English history my class discussed was that a small number of women were too busy to fill out surveys.)

      • g says:

        D’oh, of course I was wrong to say “cannot possibly be true” and you could be right if there are lots of multiway poly relationships. But that’s not the sort of situation the article was describing; the narrative was “there’s a big excess of women over men, so the men have lots of girlfriends”.

    • Hari Seldon says:

      Without a rigidly enforced norm of monogamy and marital-sex-only, a significant percentage of men never or only rarely have a mate. Even when there are more males than females, there are going to be a few guys at the top with multiple girlfriends. I suspect that polygamy is the default state for homo sapiens.

      • TheMule says:

        Just so!

        • houseboatonstyx says:

          Not everyone. William James* had it right:

          Hogamus, higamous
          Man is polygamous
          Higamus, hogamous
          Woman monogamous.

          * Or somebody. See – which is very interesting but could use better layout.

          • jaimeastorga2000 says:

            Females are not monogamous. They are hypergamous. Hypergamy is sometimes known as “serial monogamy”, but there is still a huge difference between that and the traditional definition of monogamy.

          • DrBeat says:

            “Female hypergamy” is a bullshit concept and has no more explanatory power than “not hypergamy”. The only way to justify it is with perceptive bias, where women ditching their partners for richer ones is proof but men ditching women for younger and prettier ones can’t possibly be part of the same pattern.

          • jaimeastorga2000 says:

            The only way to justify it is with perceptive bias, where women ditching their partners for richer ones is proof but men ditching women for younger and prettier ones can’t possibly be part of the same pattern.

            Men only ditch their current women for younger and prettier ones when serial monogamy is socially enforced. If it at all possible, they prefer to add the younger and prettier girl to their lives as a second wife or as a mistress instead. In contrast, women lose all interest in their current men if a higher quality alternative is suddenly willing to commit to them.

          • Nita says:

            @ jaimeastorga2000

            they prefer to add the younger and prettier girl to their lives as a second wife or as a mistress

            Sure, they don’t mind the “old” wife still doing their laundry, cooking their meals and raising their kids. I’m sure even the most hypergamous ladies don’t object to their “low quality” men still buying them stuff and giving them compliments, either.

          • jaimeastorga2000 says:

            Sure, they don’t mind the “old” wife still doing their laundry, cooking their meals and raising their kids.

            Men find a variety inherently attractive in a way that females don’t. See the Coolidge effect.

          • Nita says:

            1. The Coolidge effect is about mating with the new female despite seeming “exhaustion” with the original one.

            2. If this effect in male rats means that “men find a variety inherently attractive”, then the same effect in female hamsters means that women find variety inherently attractive.

        • Hari Seldon says:

          You could call it a just so story, but I think that phrase gets overused. Especially, when we have such good evidence for this being true. No, we don’t have epoch spanning direct observational studies to back this up, but we don’t have that for anything in biology.

          What we do know is that the female contribution to our modern genome is about 2:1 when compared to the male contribution. Yes, there are confounding factors such as early violent male death. But with such a stark contrast, it is pretty hard to conclude anything other than the fittest males (however nature defines that) mate with multiple women and a significant percentage of males are left without a mate.

          Any cursory glance at a modern college campus will tell you that model still fits reasonably well. The further we get from socially enforced monogamy, the more it seems to hold true.

          There is a tendency to call something like that a just so story because it seems to give credence to icky red pill ideas. But the conclusion is hard to avoid. What society should do with that conclusion is, I guess, up for debate.

          Being on the… uhhh… homelier… end of the male spectrum, I am in favor of socially enforced monogamy. But if I were 6’4″ and beautiful I can see how I might feel differently.

      • Jaskologist says:

        The evolutionary best case for men is to impregnate as many women as they can, then skip town. Best case for women is to get impregnated by those men, while keeping a rich man around to pay for the bastard.

        Polygamy is a stable defect/defect in the societal mating game, where top lets the bottom go hang. Monogamy (life-long, not the serial kind we pretend at now) is cooperate/cooperate. We don’t get everything we want (particularly those at the top), but overall most people get a better outcome, and the outcome for those at the top still isn’t that bad. (Societal bonuses: equality for women becomes practical, married men are much more productive, less criminal).

        I sometimes suspect liberal sensibilities evolved for these kinds of situations. Mating really is a zero-sum game, and to build a just society the guys at the top really do need to take a hit for the team. The trouble comes from applying this thinking in the economic realm, where we can grow the size of the pie, just as it is the opposite mistake to apply free-market thinking to the mating game, where increasing the pool of mates just increases the pool of people needing mates at a rate of 1:1.

        • Randy M says:

          “Best case for women is to get impregnated by those men, while keeping a rich man around to pay for the bastard.”

          So, I was going to object to it still (in modern environment) being the best case for women to be impregnated by *those* men (cads), given that those are probably no longer the traits she wants in her offspring (in comparison).
          But, if those are the traits that turn women on and encourage their own propagation, maybe it is still in her “interest”, dyscivic as it may be.

          • Nornagest says:

            Yeah, women (and men) choose people for a lot of reasons, but conscious deliberation about what traits they want in their kids isn’t usually one of them. We could reasonably assume our unconscious heuristics for mate selection evolved to take that into account, but we can’t assume that the traits they cue off of are well calibrated for the present environment.

          • keranih says:

            She will have more grandkids if her kids are cads, than if her kids are gentlemen.

            That’s a win, in evolutionary terms.

          • drethelin says:

            The prevalence of dads, and even just a hundred years ago dads with a dozen kids, implies that they’re not at as much of a sexually selective disadvantage vs cads as one might think.

            There’s also another issue: Context-sensitive behavior. There could easily be people who have it in them to be either dad or cad depending on how their life turns out. Other animals with less complex relationships, like Sealions, are able to switch from sneaky to dominant roles depending on their position in the heap.

          • Jaskologist says:

            Looking back at this, I really mixed up two things. Monogamy is to keep the rich guys from monopolizing all the women, who are generally willing to share a man if he’s got enough going on. Stigmatizing sex before marriage is what keeps the cads in check.

            They’re hard to disentangle, because they work in tandem and rely on each other. Stigmatizing sex before marriage pushes people from short-term to long-term mating strategies, which are far more beneficial for society. Keeping the elites from hogging all the women makes it possible for the maximum number of men to pursue those strategies.

            Once you allow sex outside of marriage, incentives shift towards short-term mating, monogamy degrades to serial monogamy, to serial monogamy + one-night stands but those don’t count, to polygamy.

          • Nita says:

            Once you allow sex outside of marriage, incentives shift towards short-term mating

            You need to specify what you mean by “mating”. Romantic relationships? Sex? Pairing up for reproduction? Because I’m seeing different behaviors in those three types of “mating”.

  63. You mention a test for political attitudes with the question “Do studies show capital punishment decreases crime?” Is this a test where it matters whether your answer is correct, or only what political views it correlates with? I ask because that is not a question to which there is a clear answer—statisticians have been arguing both sides of it for decades.

    • BD Sixsmith says:

      I thought this test was bogus but when it turned out that I was in the least biased 20% of respondents I learned that it was a flawless masterpiece.

      • Protagoras says:

        I came out totally unbiased, but I am a little concerned that I knew it was a test for bias. A lot of the answers I just got right, but while some of the questions were trivial, a few of them were fairly obscure and I just had to try to make a plausible guess. Did I guess in a more conservative direction because I was unconsciously trying to do well on the test? Given that people do that sort of thing all the time, it seems depressingly likely.

        • Saint_Fiasco says:

          Isn’t that what you should do all the time, not just on tests?

          If you don’t know the answer to a controversial question, and you are aware that you strongly desire the answer to be something, you should believe the opposite.

          Not all the time, but on the margin, most people should believe the opposite of their bias more of the time.

          • Orphan Wilde says:

            Bias doesn’t imply you got the answer wrong more often, but rather, that you tended to answer in certain ways more often. It’s entirely possible to be biased in a way that arrives at the truth more often than not (indeed, that’s more or less how heuristics work).

            Or, alternatively, reversed stupidity isn’t intelligence.

        • Cauê says:

          I suspect the same as Protagoras, from a different direction. They say I’m 2,78% biased, and I don’t think they say which way, but I was “too liberal” on the ones I got wrong.

          (also, this thing was basically focused on Dem X Rep talking points, and my score is “higher than 99% of test takers”? Step up, USians!)

      • merzbot says:

        I came out in the least biased 2%, but I basically guessed for half the questions. The main thing I learned from that quiz was that I don’t know shit about politics and don’t have any business having opinions about most political issues.

    • Nathan says:

      Yeah I didn’t like the bias test at all. For example I got a lot of the “According to a consensus of economists” type questions “wrong”. Apparently disagreeing with Keynesian models makes you objectively wrong. Would much prefer it if it stuck to much more objective criteria like “What % of the US budget is spent on food stamps?”.

      Alternatively, it could just ask “What is the prevailing academic opinion of X?”

      • Nita says:

        “According to a consensus of economists”
        “What is the prevailing academic opinion of X?”

        What’s the difference?

        • Lupis42 says:

          Because the “According to a consensus of economists” wasn’t stated until after the fact.

          One is an explicitly factual question – i.e. “What do a bunch of economists say about this?” The other entirely on the users model preference, which makes it more of an explicit “what is your bias” question.

          • Urstoff says:

            And there was no “we don’t know” answer for that question either, which I would think is the proper answer.

        • Nathan says:

          There isn’t one. The point is the question was “Does X do Y?” I say no as this is my view. I’m told nope, the majority of economists disagree with you. Like, I knew I held a minority opinion before.

          It’s like being asked “who would have made a better President, Obama or Romney?” and then being labelled as biased if you said Romney because actually Obama won the election.

          (Not the best analogy since Romney would have genuinely been terrible but you know what I mean)

    • Lupis42 says:

      The answer they considered “correct” was “we don’t know yet”.

      My big beef is with the stimulus question, as Nathan mentioned. While flawed, it’s still much better than most tests of it’s type.

    • ADifferentAnonymous says:

      On a few questions, including that one, the available answers are ‘yes’, ‘no’, and ‘we don’t know yet’.

    • Acheman says:

      I think they specifically asked if capital punishment decreases the homicide rate. I answered ‘no’, because I believe capital punishment *is* homicide. I suppose it’s possible that this is a trolley problem where you murder one criminal and save two other people, but that’s a pretty high bar to clear, evidence-wise.

      • Mary says:

        Closer to five or six.

        You have to actually USE it mind you.

        Would you also object to imprisonment as kidnapping and fines as robbery?

      • “I suppose it’s possible that this is a trolley problem where you murder one criminal and save two other people, but that’s a pretty high bar to clear, evidence-wise.”

        My memory of the result from Ehrlich’s statistical analysis, which I think more or less set off the serious statistical arguments, was that each execution deterred about ten murders.

      • Pku says:

        Much as I hate to get into object-level debates, this was part of my beef with the question too.
        On second thought, I considered that executions are rare enough that even if they only cause a mild deterrence in homicide, it’s still enough to win out.
        On third thought, even if they do, they’re so ridiculously expensive for the justice system that even if you strongly believe it deters crime, a utilitarian approach still implies it’s a terrible idea.

        • Adam says:

          Well, there’s the additional problem that if you accept the innocence project stuff and all the recent news about the FBI just inventing forensic science without any actual science involved to get convictions, that even an effective deterrent doesn’t work if you execute the wrong person.

          • Mary says:

            Why not?

            This was C.S. Lewis’s objection to the theory that we should drop this crude retribution stuff for deterrence (and rehabiliation). The point of deterrence is to deter, not to punish the guilty; therefore it is more important that he look guilty than be guilty.

          • Jaskologist says:

            Sure it does, as long as the rest of population believes you executed the right person.

            (This is one of the classical problems with utilitarianism. Show trials are nearly as full of utils as real ones if they convince the plebs. The real evil is innocence projects which break the illusion.)

          • Pku says:

            Why wouldn’t an effective deterrent work if you execute the wrong person (so long as it’s not widely known that that’s what you’re doing)? It worked in Proven Guilty. And by worked, I mean let an evil wizard spy into the white council.
            (But more seriously, this does bring up the question of whether having a death penalty that never gets used and lawyers don’t even try would be a good idea – you don’t waste money on expensive trials and appeals and it might still scare people. OTOH, having a penalty that never gets pursued might make the justice system look less sincere and thus less scary, which would backfire).

          • Adam says:

            You’re bringing up fiction. If I murdered someone and you executed someone else for it, I feel like that would encourage me to go ahead and do it again.

          • Randy M says:

            Sure, it wouldn’t deter you, but I think deterrence is usually argued on the basis of others looking at the effects of being convicted of a crime. So long as there is at least a credible belief in the (however tenuous) connection between guilt and conviction, it should have an effect on premeditated crimes.

          • Adam says:

            I probably estimate my ability to get away with murder higher than most other would-be murderers. I was actually under the impression that most murders weren’t even all that planned anyway and tend to be kind of spur-of-the-moment emotional crime of passion things. Well, murders that aren’t just straight up gangland executions and shoot-outs where the participants probably don’t expect they’re going to live past 25 anyway.

          • Nornagest says:

            I remember being surprised at just how low the threshold for premeditation is set: I might have some of the details wrong, but I seem to recall that e.g. going into the next room to get a kitchen knife that you then use to stab somebody would get you Murder 1 in a lot of jurisdictions.

          • Pku says:

            And I’ve always been confused that premeditation makes the punishment different. I mean, it makes sense from the perspective of “some people are evil, and the purpose of the justice system is to find and punish them”, but doesn’t really make sense elsewise (from the deterrence perspective, I’d just think “how do I make it look spur-of-the-moment?”).
            (Also, yeah, the fiction thing was a joke (though the original book is actually pretty good in its discussion of the situation)). I pretty much agree with your original point, though it also has the implication that the innocence project might be one of those places where we should practice the virtue of silence – which seems wrong to me, since I believe that abolishing the death penalty would do enough good (in various ways) that the benefit gained by criticizing it outweighs what we lose by breaking the virtue of silence here.

          • Adam says:

            I honestly don’t feel like the absolute number of unjust deaths either prevented by killing murderers, or caused by executing people who were not actually murderers, is sizable enough as a social problem that the utilitarian value of something like the innocence project is more than negligible anyway, whether or not it’s a positive or negative value. If it matters at all, it matters on principle.

          • keranih says:

            From a rational standpoint, does a principle matter if it does not have a practical effect?

            (srs question)

          • Adam says:

            Not to me, but I phrased it as a conditional so as not to imply that anyone who cares about principle is necessarily not rational, I mean, excepting something like Scott’s crusade to care about principles specifically because they produce rationality.

          • Mary says:

            (from the deterrence perspective, I’d just think “how do I make it look spur-of-the-moment?”).

            How is that any different from your thinking, “how do I make it look accidental?” to get off entirely?

            Do you think that people who kill when obviously they have had no time to think ARE as guilty as those who plotted it?

          • Nita says:

            Someone who killed because they rationally decided that it would benefit them might kill again if another murder would benefit them.

            Someone who killed because they were really, really angry in that particular moment might kill again if they experienced another fit of anger.

            The purpose of the justice system is to make life safer, not to satisfy our emotional itch to hurt those who trigger our sense of outrage.

          • John Schilling says:

            The purpose of our emotional itch to hurt those who trigger our sense of outrage, is also to make life safer. You’ll probably get better results if you arrange for all of these things to work in parallel rather than setting them against one another.

          • Doctor Mist says:


            The purpose of the justice system is to make life safer, not to satisfy our emotional itch to hurt those who trigger our sense of outrage.

            I’ve heard it both ways.

          • Mary says:

            “The purpose of the justice system is to make life safer, ”

            No it isn’t. If it were, it would be used against a lot more things than it is.

      • lmm says:

        So you use different definitions of words from your interlocutors to make claims that you know are misleading? Sounds pretty biased to me.

    • Earthly Knight says:

      The question I was least happy with was the one which asked whether the US has a higher GDP per capita than “nearly all” or “most” European states. I know the number is all but 2-6 depending on what source is used, how many principalities, duchies, and popedoms are counted, and vicissitudes in the price of oil. But there is still the further question: is this “nearly all” or merely “most”?

      • Cerebral Paul Z. says:

        The correct answer also depends on whether you do the comparisons using exchange-rate parity or purchasing-power parity– something the question doesn’t tell you.

      • Jon Gunnarsson says:

        I agree that the difference between “nearly all” and “most” is ambiguous, but the rest of the question is entirely clear. The question was asking about EU members, not European countries, so the Vatican, Liechtenstein, Monaco, and the like are not included.

        • Earthly Knight says:

          It seems unlikely that a question which hinges on whether I know that Liechtenstein is not in the EU is going to be much use in sounding out my political bias.

          • Jon Gunnarsson says:

            But if, in the absence of such knowledge, you consistently make assumptions about the world that favour your preferred political ideology, that’s evidence of bias, which is what this test is designed to detect. Given the brevity of the test, it obviously can’t do this remotely perfectly. With just 16 question, a pattern picked up by the test may well be pure chance, but then most people wouldn’t be willing to take a much longer test.

          • Earthly Knight says:

            The problem is not the brevity of the test, it’s that the question is a bad one. The following seven European countries, out of ~50 total, have a GDP per capita that is by some measure higher than the US’s: Luxembourg, Liechtenstein, San Marino, Norway, Switzerland, Monaco, The Vatican. Of these, only Luxembourg is one of the 28 European Union member countries. Consequently, a person who has an accurate sense of the comparative wealth of the US and Europe, but does not keep a fastidious chart of which microstates are EU members and which aren’t, will reasonably conclude that “most” is the appropriate answer. By restricting the question to the EU, the quiz actually rewards people with an anti-welfare-state political bias, because their estimates of Europe’s wealth will, by chance, project better onto the EU’s unrepresentative sample.

          • suntzuanime says:

            On the other hand, the microstates are, well, micro, so someone who has an accurate sense of the comparative wealth of the US and Europe but does not pay attention to microstates because they have only a minimal effect on the overall wealth of Europe would get the right answer. Basically, yeah, it was a bad question.

  64. You explain greater variability in men as due to their having only one X chromosome, which only makes sense for genes on the X chromosome. There is a more general evolutionary explanation. Being male is, reproductively speaking, a high risk profession, with the potential for greater reproductive success at the high end, zero at the low, due to the fact that wombs are the scare reproductive input. A design that sometimes produces very high quality organisms at the risk of sometimes producing very low quality ones—think of it as a sports care in comparison to a sedan—has a larger payoff if the return to being a success is very high.

    • Nita says:

      Even the most plausible just-so story needs some sort of mechanism to work.

      • Vaniver says:

        Agreed that knowing the mechanism is nice, but there’s an important point about probabilistic reasoning here: conjunctions cannot be more likely than their components, but people often consider them more likely.

        To make it explicit:

        A) Males have higher variance on a particular trait.

        B) Higher variance is caused by Mendelian genes on the X chromosome.

        The conjunction A & B seems more reasonable to the person on the street because it supplies a reason–instead of just being a weird claim, it’s a weird claim with evidence to support it. But if you see B as a claim in its own right that requires its own burdensome evidence to prove, it could very well be incorrect or incomplete, and thus the combination A & B must be not more likely than A by itself.

        There’s also a point about reasoning around evolution: if there exists selection pressure, then if a mechanism exists it will be found. But if there exists a mechanism, whether or not it will be found depends on the selection pressure. Why is the Y chromosome atrophied and the X chromosome large? If that had been sufficiently unfavorable to reproduction, something else would have happened.

    • Douglas Knight says:

      You could test the sex chromosome version of the theory (which I think is stupid) by looking at birds, which have the opposite sex determination system. Do they also have greater male variance? I don’t know.

      • vjl says:

        There is a common misconception that any sex differences need to come directly from the sex chromosome. I can see how this makes sense at first blush, since sons and daughters draw from the same set of autosomal alleles, but in practice this is a non-issue. The only thing it necessitates is a simple signal on the XY that says “It’s a boy”. From there, any imaginable sex difference can be derived through if-then switches in the autosomes.

        • Ptoliporthos says:

          So, yes, Sry acts as a switch in the embryonic gonad to turn on a set of genetic programs that make a testis and testosterone and shut off the pathways for making an ovary and estrogen, and that obviously has profound consequences. But Sry is on (and just barely on, maybe only a few copies per cell) for all of about a day in one cell type.

          On the other hand, there are hundreds of other genes on the X chromosome that differ in their expression levels and tissue expression patterns between males and females.

          There are also about a dozen of them that have similar, but different proteins encoded by the X and Y. These dozen are mostly regulators of transcription and translation, and are ubiquitously expressed, at high levels, from the 2-cell stage till the day you die.

          Any of these other genes could have potent switch-like behavior that governs pathways for male or female phenotypes in a broad array of tissues and cell types, both before and after Sry is expressed.

          It would not be prudent to ignore everything on the sex chromosomes except Sry.

    • Randy M says:

      I took the explanation to be that negative recessive genes on the X had a sort of spill over effect on other traits from other chromosomes. Our the lack there of, as sometimes having one bad, one good allele still brings some detriment, iirc.
      Say the man inherits a slightly off regulatory enzyme for a hormone, this could have an effect on other organ systems not other wise on a sex chromosome.
      Similar to the Mutational load theory for IQ.

      • Randy M says:

        That said I think our default assumption would be the means would not be the same. I mean, we know that, for example, sex hormones effect the brains, men and women perform differently on different cognative tasks, and brains of men and women scan differently with MRI (or something, correct me if I’m wrong). Given that there are factors differentiating brains, which generate intelligence, it seems rather serendipitous that the means would happen to come out the same, although certainly not impossible that they are the same to within statistical noise.

        It’s like saying, “We have two cars here, one designed for highway driving, one for city driving, from different manufacturers. But since they both run on gasoline, we’ll assume they get the same mpg.”

        • “I think our default assumption would be the means would not be the same.”

          I’m assuming you are talking about IQ. My understanding is that the reason the means for males and females are the same is that the tests were designed to give that result. Men do better on some sorts of questions, women on others, so the relative means depend on how many of each sort you put in the text.

          Am I mistaken?

          • Adam says:

            I think you’re right about that. It’s sort of the difference between testing owls and bats for vision and hearing, which would give tremendously different scores, and testing them for ability to fly in the dark, which they’re both really good at.

          • Randy M says:

            Scott brought up the subject of variability and similar means in the context of IQ, so that is what I was thinking; it hadn’t occurred to me that the test was normalized in such a way as to guaruntee it, but I suppose such a thing is quite possible.

            However, I do take exception with Adam’s analogy, which implies that any two groups which have comparitive advantage for a facet of a task will do equally well at the task. If Jack is a great runner but can’t swim, and Joe is a great swimmer but a poor runner, they aren’t going to have the same performance on the Ironman triathalon.

            Though, of course, in the case of IQ tests, as you two are saying, it is possible to design a test in such as way, it would of course then be disingenuous to use the results to prove much about similarities and differences between the groups.

          • Adam says:

            I really didn’t mean to imply that any two ways of accomplishing a similar task would lead to equal performance. I just used that as a singular example of a case where it does (and it might not even be accurate – I have no idea if it’s actually the case that bats and owls are equally good at flying at night).

      • Mark Z. says:

        “Mutational load” is bullshit. 100% of our DNA got there by mutation, so it’s ridiculous to talk about who has “more” mutations.

        The proposed explanation is that recessive genes on the X (let’s say one is the gene for being really awesome at physics, and the other is Lesch-Nyhan syndrome) are all fully expressed, leading to many rare traits being more common in men. This is a known mechanism in many genetic diseases so it’s plausible it could apply to natural variations in intelligence or, say, athletic performance. That’s the male variability hypothesis.

        Does it help to explain it as “women get to roll 2d6 for their attributes, and men get to roll 1d6 and double it”?

        • Randy M says:

          I can’t tell if you are agreeing or disagreeing. All of that was understood; do you think this could explain a broad range of traits despite only affecting 1/26th of the genome?
          That was the point I was trying to discuss.

          ““Mutational load” is bullshit. 100% of our DNA got there by mutation, so it’s ridiculous to talk about who has “more” mutations.”

          Also, I find this giberish. Yeah, sure, all the DNA is mutation from ancestral proto-cells, but that doesn’t dispute the fact that most mutations that arise are negative, and certainly most potential mutations.

          • Randy M says:

            The question is, are we under the same selective pressure as we were when the genes in question were first selected for, from whatever un-mutated precursor we had then? I think obviously not, ironically due to those same genes allowing us to band together against our environment.

    • It might be possible to test how much the variability is from having one X chromosome and how much is from selective pressure by looking at male and female variability in a species with a ZW sex-determination system (wherein the sex chromosomes for males in ZZ and for females is ZW).

  65. You mention a purported legal rule in China that makes it in the interest of a driver who accidentally runs over someone to make sure he is dead. The equivalent was true of traditional Anglo-American common law. If you tortiously injured someone, you owed him damages for the injury. If you killed him, his claim died with him. That rule was modified during the 19th century by state statutes giving his family a claim for the injury to them imposed by killing him, but it’s still the case that the injury to him is, in theory, not part of what you owe.The alternative rule is referred to as hedonic damages—damages for his loss of a lifetime worth of utility.

  66. Dirdle says:

    Latest big meta-analysis finds “little evidence” for Roy Baumeister’s ego depletion hypothesis, describes itself as “strongly challenging” the idea of willpower as a limited resource.

    Well, there goes that excuse. What is the thing that depletes when you do something unpleasant, then? I mean, the feeling of not wanting to make further effort after having made some is a real thing; it must be coming from somewhere. Some kind of evolved trick to avoid expending actual resources? It seems like “there’s unlimited willpower but also secret neural circuitry to create a convincing illusion of limited willpower” is pure epicycles.

    • Ptoliporthos says:

      It’s called laziness, and as any programmer can tell you, it’s a virtue.

      • Dirdle says:

        Agreed that that’s the thing we’re talking about, but I don’t think it really suffices for an explanation. Unless I missed the LHC finding a fundamental Fuk’s Givon or something =P.

    • Exactly. I consider claims that such and such a study challenges the idea of limited willpower much like reading a claim that “this study strongly challenges the idea that food is needed to satisfy hunger.”

    • 27chaos says:

      It’s not pure epicycles. They can do experiments where first they “drain” someone’s willpower and then they change the situation, by giving them a new incentive, by altering the problem, by asking them to pray while working, and are able to get their willpower to persist beyond where it was.

      Willpower is less like a pool of limited resources and more like a rubberband with variable elasticity.

      • Dirdle says:

        To be clear, the comment about epicycles was with regards to my own guess that maybe the brain has an unlimited supply of ‘willpower’ but informs you that it’s exhausted to get you to be lazier, save calories, whatever. But that didn’t seem different to the ‘limited supply’ case. Thank-you for your clarification of what they meant originally, though, it’s much appreciated.

    • fire ant says:

      Wellll, _maybe_ the researchers just didn’t have enough willpower to reproduce the results…

      (Is it okay to make bad jokes about that? :D)

  67. Markus Ramikin says:

    “Mitochondria have always been creepy. But now it’s starting to look like they talk to each other. Let’s hope they’re not chatting about how eukaryotes suck and it’s time to take over.”

    You’re not hoping they are just figuring out how to communicate the will of the Force to us?

  68. Richard Gadsden says:

    I thought there was plenty of evidence that bisexuals of “both” sexes* had more mental health issues in general than monosexuals of the same sex.

    It’s not therefore all that surprising to find that the same is true of a specific mental health disorder.

    * The data collection I’ve seen is strictly binary by sex, so we don’t know whether this is also true of non-binary bi/omni/pan sexuals.

    • Pku says:

      Just out of general curiosity, are there enough non-binary bi/omni/pan sexuals to skew statistical results? AFAIK (which isn’t very far) they’re fairly rare.

  69. lmm says:

    I feel bad for even thinking this, but the bisexual women/eating disorder result would be predicted by the naïve/evil model that both correlate with attention-seeking. I hope that’s not it.

    • Anonymous says:

      Personally I tend to associate bisexuality (male or female) with mental instability in general. Not sure if it’s true.

      • Richard Gadsden says:

        I think it is, but the question of causality seems to be rather tricky.

        From what I’ve seen, there are several candidate explanations:

        Some genetic, foetal-environment, or early-childhood environment factor causes both bisexuality and mental health issues.

        WEIRD societies are oppressive to bisexual people and the oppression causes mental health issues. (biphobia among LG communities is noted as a particular problem).

        Amongst bisexuals, lying about being bisexual and pretending to be monosexual is associated with the lack of mental health issues, so the samples are skewed.

    • My admittedly limited understanding of eating disorders is that in general they are related to basing one’s self-esteem on conformity to a particular body ideal, and feeling dissatisfied with one’s appearance insofar as one does not meet this ideal. I think this is quite different from attention seeking, particularly since there are many more ways to seek attention from others that do not involve dysregulated eating. As to bisexuality, I don’t know that much about the topic, but I find it implausible that people work out their sexual orientation based on attention seeking. I have heard that bisexual people actually endure a certain amount of flak from within the gay community from people who think that people are supposed to ‘pick a side’ for some reason, so as an attention seeking strategy it seems like a pretty crummy one.

      • suntzuanime says:

        If they’re enduring a certain amount of flak, it sounds like they’re getting a certain amount of attention? Seems fine to me.

      • Tracy W says:

        There’s a treatment for anorexia, the

        Maudsley Approach

        , which seems to be effective for suffferers under age 18 which mostly consists of feeding the kids back up again, and while the approach is agnostic on what causes anorexia there’s an associated hypothesis that the mental problems of anorexia are caused by weight loss (starting say from a brief diet or an illness or something), and in small number of people that causes crazy thinking.

        • Nita says:

          According to Wikipedia, the main difference between the Maudsley approach and other forms of therapy is involving the parents and teaching them how to persistently encourage eating without alienating the kid.

          It does require a lot of attention. On the other hand, attention is a basic human need, just like food or water.

      • Cadie says:

        Bisexual women have two sets of “beauty standards” to work with – pressure to be appealing to men, AND pressure to be appealing to other women. Maybe that confusion affects the eating disorder rate; they have the same pressures that heterosexual women do, plus a few more.

        • Who wouldn't want to be Anonymous says:

          I was under the impression that women are much, much more critical of other women’s appearance than men are.

        • houseboatonstyx says:

          Bi-sexual women have more partners available: het men, bi men, bi women, and cool lesbians. No wonder we don’t have time to eat.

    • Pete says:

      Anecdote: My wife is both bisexual and has/had an eating disorder. She’s also bipolar. I wouldn’t say she’s particularly attention seeking though. She was certainly accused of that by her parents growing up (which lead them to not giving her the support she probably needed).

    • anonymous says:

      I commented about the bisexuality issue a bit below in a comment that begins as “Regarding bisexuality and eating disorders”, what do you think about the idea that the problem is simply that a certain type of person will not define herself as bisexual even though she is, because she doesn’t want to signal infidelity to her monogamous partner.

  70. JK says:

    Also, I am suspicious because you never get a real correlation of 0.9 with anything; even measurement error should be worse than that!

    While I’m skeptical about Piffer’s methodology, too, the correlations he reports are ecological correlations between group means, which can be very high even when the individual-level correlation is modest. Within-group variation is completely ignored when ecological correlations are used.

    • Scott Alexander says:

      Okay, but:

      1. Global IQ data is notoriously unreliable. I wouldn’t even expect a correlation of 0.9 between global IQ data from one source and global IQ data from another source.

      2. Genetic differences explain only 50-80% of within-country IQ variation. I’m not sure how directly comparable the two paradigms are, but I would expect between-country IQ variation to have a lot more confounders.

      • Ever An Anon says:

        If the other 20-50% is noise, wouldn’t we expect it to smooth out on a national level?

        I haven’t read the paper yet and I don’t have a background in psychometrics so I can’t form an opinion on the paper itself, that just seems like a weird objection.

      • JK says:

        Look at the graph of parental income and SAT scores here. The linear correlation between income brackets and average SAT scores within them is very high (close to 1). However, at the individual level, the correlation is probably 0.3 or less. If there’s a monotonous linear relation, no matter how modest, at the level of individuals, the correlation at the level of group means will approach 1.

        Global IQ data are actually reasonably reliable. For example, the ecological correlation between Richard Lynn’s IQ data and PISA-type student tests is >0.80.

        I would also expect genetic and environmental influences on IQ to be positively correlated at the global level (the better the genes, the better the environment). This would tend to make the correlation between IQ genes and IQ larger.

      • Ahilan Nagendram says:

        Genetic differences explain only 50-80% of within-country IQ variation.

        Not even, it’s more like 50-80% heritability for smaller groups, like higher SES whites in U.S. for example, not national populations. This differs by country, we have data for U.K., Sweden, and U.S. to go by when it comes to twin studies and 50-80% heritability is there when there are no SES confounds (i.e. a stable and enriching environment, little bad luck etc.) and such.

        In Scottish samples lower SES cohorts seem to have much more stable an environment than U.S. cohorts, might be a racial confound and genomic as well as phenomic differences.

    • Richard Gadsden says:

      Everything at Ex Urbe is really good. The series on Machiavelli (SPQF) is amazing in particular, but even the post recommending gelaterias in Florence had me reading to the end unable to stop until I ran out of words to read.

      The author revealed herself about a year ago. The blog was pseudonymous until she got tenure, which seems like a smart move. She’s a professional historian (specialising in the renaissance recovery of ancient thought) and also an SF author, with her first novel coming out next year, so that might be a useful recommendation to some people.

        • Richard Gadsden says:

          Some people are just excessively talented.

          A Renaissance woman, indeed.

        • Luke Somers says:

          And Cosplayer, and DM. /personal experience

          Also, I was one of the alpha readers for the SF book. One of the less helpful ones, I’m afraid.

      • Douglas Knight says:

        She doesn’t have tenure. She was very vague about the reveal. My best guess is that publishing a book legitimized her, largely based on this post. But if it were that simple, I think she would have said so.

        Why do you say it is a smart move? Which move, exactly, do you mean? The strategy of having lots of public low-brow hobbies, but hiding the topical blog seems very strange to me.

        • Richard Gadsden says:

          Hobbies were harmless – do what you like in your free time.

          Blog was writing pop-history, which is dangerous, until you have a strong enough publication record. If you write pop-history, the fear is that you’re a pop-historian not a real academic one. So you do your pop-history on the quiet until you’re established enough that no-one can pretend you’re just pop-hist.

          Look at the sneering at John Norwich

  71. nico says:

    Corbyn intends to withdraw from Nato and opposes the renewal of the Trident nuclear deterrent.
    He is in favour of unilateral nuclear disarmament and has called for a “radically different international policy” based on “political and not military solutions”.

    I’m an American, and both my parents have worked for NATO at points, so I’m obviously going to be biased.

    But this sounds absolutely crazy. Like, this seems on par with some hypothetical Republican saying we should cut diplomatic ties with Canada. It may not immediately lead to disaster, but what could you possibly hope to gain? Someone help me out here.

    • Nita says:

      He believes that Russia’s recent misbehavior is (at least from Putin’s point of view) a defensive response to the expansion of NATO. The reasoning goes: expanding NATO escalates the sort of international tension that can lead to war, so we should do the opposite to deescalate.

      Also, he’s a pacifist, obviously.

      • FJ says:

        I’m not quite sure how Corbyn thinks the conversation in the Kremlin would go.

        DATELINE: MOSCOW, 2020
        “President Putin, the Labour government has won the election and is withdrawing from NATO!”
        “Excellent! Remind me again, this is the Polish election?”
        “Nyet, Mr. President.”
        “One of the Baltic states, then?”
        “Sadly, no. They are more passionate about NATO membership than ever.”
        “… America? Does America have a Labour party?”
        “No, Com… I mean, Mr. President. The United Kingdom. Jeremy Corbyn won the election. You know, from Islington North.”
        “… Oh. So a nation that does not border Russia, and does not border any state that borders Russia, has withdrawn from NATO. Is that remotely relevant to Russia’s security?”
        “Well, I guess it means they won’t try to bounce when we order F GoB -> Nwy.”

      • brad says:

        He has a point. Keeping the military alliance in place after you win the war, and even expanding it to your former enemy’s break-off states, is not exactly a way to guarantee reconciliation. In fact it sounds a bit like Versailles.

    • lmm says:

      Britain went to war in Afghanistan and Iraq against the wishes of much of the population. Parliament approved the war on the basis of statements by then-PM Tony Blair which were false, which senior people in Blair’s administration knew were likely to be false, and which much of the press and public thought to be false, as they ultimately proved to be. There was a widespread perception that Blair did this because the US wanted him to, and didn’t care what the British people or the British public wanted.

      No-one is suggesting breaking off diplomatic ties, but NATO is a military organisation; one popular hope is simply that we would feel less obliged to fight America’s wars for them.

      • Ptoliporthos says:

        Wait, what was false about the rationale for Afghanistan?

        • Sastan says:

          There was bad intel on Iraq, which caused the neutrinos to time travel and make the Taliban not harbor Al Qaeda.

        • Richard Gadsden says:

          That the objective was to capture Osama bin Laden, when it clearly wasn’t because (a) the US did sod all about it until Obama was elected (b) then they killed him, not captured him and (c) having done so they didn’t pull out.

          Now, this depends a bit on what you think the rationale was, and whether retrospective evidence is valid (personally, I supported the invasion of Afghanistan and don’t think the rationale was dishonest) but it seems reasonable enough.

          • LeeEsq says:

            The invasion of Afghanistan was made necessary by the realities of politics. Politicians generally want to stay in office rather than lose office. Presidents do not want to face impeachment. There was no way that any President of any ideology would be able to get away with saying is that “this horrible terrorist attack can be traced back to the government of this country but we are going to do nothing to retaliate because that would only make things worse”.

          • John Schilling says:

            “Capture Osama bin Laden” was the Geopolitics for Dummies version.

            The more complicated and realistic version is that we needed Al Qaeda to not have a secure base of operations and financial support provided by a hostile nation-state, whether or not Osama bin Laden was the guy at the top of the Al Qaeda chain of command. Al Qaeda as a bunch of guys hiding in caves dreaming of past and future glories, OTOH, was relatively harmless even when bin Laden was still alive.

            To the extent that we preferred him dead or in prison, attacking Afghanistan was necessary for that as well, but it wasn’t a sure thing and in fact he managed to escape. Why we didn’t turn around and immediately invade Pakistan, is left as an exercise for the student.

        • Deiseach says:

          I think it was more the perception that the US couldn’t get a coalition of allies to support it in going into Afghanistan, so it was leaning heavily on Britain, and that Blair and his government were perceived as dragging Britain into Afghanistan as a favour to the Yanks (being their lapdog) and not for any pressing reasons of British interest.

          Add in the furore over “sexing up” the documents alleging WMD etc. in Iraq, and you had the general public not much impressed with their Prime Minister and his veracity or lack of a backbone to stand up to the Americans.

          Why we didn’t turn around and immediately invade Pakistan, is left as an exercise for the student.

          Because Pakistan was (ostensibly) an ally and invading an ally and blowing up its citizens is not considered good form?

          I think if the USA had gone into Afghanistan first, and not gotten bogged down in Iraq and what seemed like a personal vendetta by Bush against Saddam (whether or not it was), much of the European opinion (which was very sympathetic and pro-American after 9/11) would have remained on the side of the US and would have seen that as necessary. But giving the impression that your country didn’t know what it was doing or what it wanted, and that it was not only relying on dodgy intelligence but deliberately manufacturing it to sell to the public, lost a great deal of support and conviction from the rest of the world.

          • Chris H says:

            Don’t forget, Pakistan at that point (and now) had nukes. In most international policy circles, invading a country with nuclear weapons is generally seen as a Bad Idea.

          • gattsuru says:

            … much of the European opinion (which was very sympathetic and pro-American after 9/11) would have remained on the side of the US and would have seen that as necessary.

            This seems… uh, contrary to my experiences and to external data.

            I mean, maybe for like a week after the attacks? There weren’t giant celebrations about 9/11 attacks, true. But even early 2002, polls on Americans were already under their 2000 numbers, while as of August 2002 the Iraq War was mostly a twinkle in Rumsfield’s eye and yet another sporadic increase in no-fly zone bombings. And on foot, reflexive anti-Americanism was present even well before you reached France or southern or eastern Germany.

          • Brn says:

            The US (and the UK) did go into Afghanistan first.

      • Randy M says:

        You mean with them, right? Good or bad, it’s hard to argue US blood and treasure weren’t spent.

    • Pku says:

      Bringing up trident reminds me of the Yes Prime Minister scenes about it

      • Pete says:

        Always good to see YPM linked. It’s fascinating how much of what was covered in that show is still relevant to British politics today (including the Trident issue).

  72. J says:

    Request: anything you can tell us about diet. After reading a particularly unenlightening comment thread on another site, I came back here and re-read your “Perfect Health Diet” review, but remained muddled. It sounds like there’s been a big shift in the past few years away from “fat is evil”, but it also sounds like there are plenty of things that are considered settled among those in the know. So I guess part of what I want to know is the epistemic status; is it all just completely noisy and overwhelmed by genetics, subverted by Big Nuts and Vegetables, or what? Regardless of whether wine and coffee are good or bad this week, am I at least safe with a super conservative plant based diet, and should I care about whether they’re cooked in oil?

    My personal interest is in longevity and cardiovascular health. I lost a bunch of weight to improve those things, so weight loss isn’t a problem for me right now, although after losing all that weight I was annoyed to read the claim in the other thread that BMI doesn’t actually improve those things. And of course I was proud of myself for getting off of meat and making lots of homemade bread, but apparently refined carbs are the new saturated bat.

    • Berna says:

      I’m very interested in this too, except I want to lose quite a lot of weight, as well. (Already got rid of 10+ kg and no longer obese, want to lose like 15 more.)

    • Nicholas says:

      The answer I received from a dietitian goes: There’s a certain ratio of macro nutrients you need to consume based on your recommended calories per day, and the “bad macro” of the week will be the one we most recently determined Americans eat too much of. All you need to worry about is hitting your ratio, not what the source of the ratio is. Micro nutrients basically only matter if you have a medical condition or family history. Everything else, unless a doctor has told you specifically otherwise, is personal taste and marketing.

      • Deiseach says:


        I’m trying to lose weight and yes, Number One Enemy is sugar (the single most deadly element in the entire universe, worse than radiation, poison and having Donald Trump wanting to make you Mrs Trump No. 4).

        In second place is refined carbohydrates. Now, if you hand-grind in a stone quern your own whole-grain, whole-meal, any grain but wheat, flour for baking and have only one slice a week with no butter or anything at all on top of it, you might be all right, but you’re tempting fate.

        Eat nuts – but not too much because the oils in nuts make you fat! Eat vegetables – but no fattening (tasty) dressings! Raw broccoli and cauliflower is the staff of life!


        • kernly says:

          with no butter or anything at all on top of it

          Wait, no. This is the exact opposite of what I understand. Adding fat to a meal helps *reduce* its glycemic load. If you’re worried about fast blood sugar rises, fat is literally the best way to get calories. I even read somewhere that a big reason past diets weren’t unhealthy, despite involving large portions of bread or equivalents, was that they also involved copious amounts of fat.

          • Deiseach says:


            Can you tell I’m finding it hard to cut down on butter and spuds? 🙂

          • Anonymous says:

            I am not a doctor but AFAIK fat + sugar is a great way to become insulin resistant. I don’t know if having them separately is a better way though.

    • Randy M says:

      Stay away from teams fats and anything deep fried. Pretty universal agreement there. Also, I think most would agree some fish and eggs well do you good if you are primarily vegetarian. Meat (red, processed, or otherwise) is still the battle ground of dueling studies, as are saturated fats, carbs, sugar, and salt. I have opinions about those, but will just point out I think the field is split with reasonable opinions on both sides.

    • Loquat says:

      Basically everyone (who hasn’t been paid large amounts of money by a soda company) agrees that high sugar consumption is bad for you, and that green leafy vegetables are good for you.

      There seems to be a fair bit of consensus that fats can be good for you, particularly olive, fish, and avocado, though beyond that you run into arguments between people who say (pastured) animal fats are good and seed oils are bad and people who say the opposite. Red meat you can group in with animal fat – some people say it’s healthy, generally if pastured but not if factory-farmed, while others say it’s bad.

      As for carbs, there does seem to be general agreement that if you’re at all at risk for diabetes you should avoid anything with a high glycemic load. Making bread with a long fermentation process supposedly can help reduce that, and of course eating your bread with fat and/or protein helps slow the insulin response as well.

      • How much of the pastured vs factory fed issue is actually about how good for you the food is and how much about treatment of the animals?

        • Whatever Happened to Anonymous says:

          And the much more important question: Does it have a noticeable effect on the taste?

          • Lesser Bull says:

            Yes. Grass-fed meat is usually richer in flavor but harder to cook properly because its leaner. So more upside, but also easier to screw up.

        • Richard says:

          I guess that would depend on what is actually good for you, but if we assume that fat is not desirable, the fat content of factory farmed animals is ~10x higher than that of pastured. This due both to diet and exercise levels.

        • keranih says:

          I’m going to repeat my call for a definition of “factory farmed” as used in the comments here at SSC.

          Animals on fresh fodder (ie, actively grazing green grass and brush) have a better omega fatty acid ratio in their meat, milk, and eggs. This is actually fairly well proven, and is a decent reason to eat meat, milk, and eggs from animals on pasture at the time of milking, laying, or slaughter. (Assumes pastured meats, etc, are equally economical with other sources of omega acids.)

          Meat, etc, from any animal in local winter will not have this improved omega fatty acid ratio. Additionally, an animal on grass which is supplemented with grains, silage, ect, will also have improved omega fatty acid ratio.

          As noted above, because purely pastured animals have a lower energy balance than those supplemented or confinement fed, the animals are skinnier and have less fat all over, esp less marbling (ie, within the muscle tissues) fat. Marbling potential is genetically determined, while actual marbling degree is limited by feed intake. (Excess fat outside the muscle tissue is typically trimmed during butchering to industry standards, regardless of the animal husbandry, and in any case contributes less to cooked taste and texture.)

          Additionally, because of the lower energy balance, pastured animals are stunted in growth compared to their supplemented cousins, and so take longer to reach the same weight. As a result, portion sizes for meats of pastured animals are smaller, or else the animals are older (and hence the meat is tougher) at time of slaughter.

          There is a great deal of interest in developing cattle/hog/sheep breeds and management styles that allow an animal to be brought to slaughter size on pasture, while allowing the meat to taste like confined/supplemented products. Because consumers the world round show a marked overall preference for the younger, fattier, “better tasting” confinement raised meat. (There are those of us who do really like tough, flavorful meat, but we are not a majority.)

    • onyomi says:

      I stand more on the high-carb end of the diet debate spectrum, which does still exist, despite the recent trend for Atkins, Paleo, etc. I recommend Dr. McDougall and Michael Greger on youtube. That said, I think total veganism like they recommend is probably a bridge too far, since we need B12, etc. The ideal diet, I’d bet, would be mostly fresh fruit and vegetables with a little seafood thrown in for healthy oils, etc. The paleo people do have to contend, however, with the fact that most of our ancestors have been eating grain for 30,000 years or more:

      And also that it’s very difficult and expensive to have a low-fat, high fruit and veggie diet without grain (though not impossible, especially if you include tubers).

      But as for consensus, I think there’s broad consensus that highly processed food is not as good as fresher food: many people think a high carb diet of rice, beans, etc. is good, but no one thinks a lot of sugar is good. And I think there is also broad consensus that, regardless of what you think of meat, fresh and minimally processed fruits and vegetables are good for you.

      Michael Pollan once summarized:

      “Eat food.* Not too much. Mostly plants.”

      *which he defines as “real food” i.e. not highly processed food; aka the stuff you find on the edges of the supermarket, rather than the middle.

      • Nornagest says:

        We don’t need to go back 30,000 years for hardcore paleo to run into problems; we barely need to go back 30. Any comprehensive theory of nutrition needs to contend with the fact that the so-called diseases of civilization were a lot rarer before the latter 20th century, even in populations that were not calorie-constrained. Not an era that abstained from bread, by any means!

        That said, I’m basically on board with weaker versions of the paleo hypothesis, with the important caveat that the forager diets we know anything about were hugely varied.

    • Richard says:

      I suspect that diet needs to be tailored to activity level and possibly something to do with genes. I run a high everything diet and spend at least two hours per day doing ‘strenuous’ physical activity and score about 25 years ‘younger’ than my real age on health tests. My main problem is gaining enough weight in winter to last me through the bicycle season…

      I probably wouldn’t recommend such a diet to a ‘normal’ person if such a thing exists.

  73. Vanzetti says:

    Street dogs of Moscow link is not working.

  74. Ted Sanders says:

    Just because paying for a surrogate mother costs $100k+, that doesn’t mean the surrogate mother earns $100k+. I think most of that money goes toward things like medication, doctors, lawyers, agents, travel, etc.

    • Matt says:

      The mother can likely work a second (primary?!) job for most of the gestation. Half the money going to the surrogate is equivalent to a 5K/month raise.

  75. Deiseach says:

    Re: Davide Piffer, when I saw he had Richard Dawkins* following him I was inclined to go “Pfft, yeah, this guy must be rubbish” but then I saw he’d retweeted a photo of Mamadou Sakho and Lucas Leiva so he’s obviously a fellow masochist Liverpool fan and now I’m torn between condemnation and sympathy 🙂

    Re: Olympic bias, I had been led to believe that things like figure skating and gymnastics were notorious for politically-motivated scoring, such that commentators would frequently forecast along the lines of “And the Russian judge will give this X and the French judge will give this Y” and be pretty much spot-on?

    Re: Chinese hit-and-runs (and hit ’em again) – isn’t this an updating of the old saw going around forever that “In China, if you save someone’s life, you are responsible for them ever afterwards”, and so likely to be an urban legend rather than factual, or at least based only on one incidence if it ever happened? I never saw an attribution for that, either; it was simply one of those “Everyone knows” things (and my current favourite new example of the genus is a dippy quote along the lines of “Muslim scholar discovered the world was round 500 years before Galileo”, getting in everything from the ever-popular White People Suck and Stole All Their Discoveries from POC, complete lack of understanding of the actual discoveries or philosophies involved, no apparent knowledge of history or chronology, and invocation of totemic name e.g. Galileo).

    Re: street dogs of Moscow – these intelligent animals sound like the proverbial “dogs in the street” 🙂

    Re: surrogacy, part of me is grimly pleased that now we have equal-opportunity exploitation (turn about is fair play, after the thriving surrogacy business using Third World women to gestate the offspring of well-off Westerners) but I’m curious why the Chinese picked America. Is there some kind of folk idea that the embryo would acquire characteristics of the surrogate, so that racism/ambition (pick your favourite) means they don’t want their children picking up traits from Indian, etc . mothers? Legal ramifications, that the child (if it ever wants to emigrate to the U.S.) would have some rights from an American citizen surrogate giving birth in the U.S.? Simple ostentatious consumption?

    *Yes I have a rule of thumb that when I see Dawkins popping up somewhere, if it’s not to attack something, I tend to judge it as rubbish. Yes I’m irrationally biased but the man annoys the ever-living crap out of me.

    • lmm says:

      I think that’s the wrong way around to view Dawkins. When he’s attacking something he’s alternately hyperbolic and ineffectual. When he’s talking about the joy and wonder of biology he can be brilliant.

      • Deiseach says:

        When Dawkins is talking about something he knows about, he’s probably tolerable. But when he opens his yap about stuff he hasn’t a bull’s notion about, he makes me want to punch the smirk off his chops.

        I’m probably absurdly over-sensitive about his ‘joke’ about Northern Ireland, but imagine an upper-middle class white professional American opining that the reason for the problems of modern day indigenous peoples in the United States was all due to their forebears clinging to their belief in spirits, then brushing that off as “It was only a joke!” and see how you’d feel 🙂

        • Vaniver says:

          I’m probably absurdly over-sensitive about his ‘joke’ about Northern Ireland, but imagine an upper-middle class white professional American opining that the reason for the problems of modern day indigenous peoples in the United States was all due to their forebears clinging to their belief in spirits, then brushing that off as “It was only a joke!” and see how you’d feel 🙂

          As discussed on MR, vodou, a religion in Haiti, is a joke in the US and is both a cause and symptom of Haiti’s problems.

        • Mark Z. says:

          Dawkins needs to hire a student to follow him around and, whenever Dawkins starts to talk about anything that’s not biology, clamp an ether-soaked towel over his face.

          The student gets to attend lots of cool conferences and lectures and meet interesting scientists, and Dawkins avoids making a fool of himself.. Win-win.

        • Mary says:

          “Like a maniac shooting flaming arrows of death is one who deceives their neighbor and says, ‘I was only joking!'”

    • ivvenalis says:

      My guess is birthright citizenship. The article downplays this to the point of absurdity, but the United States’ ius soli rules are anomalously lax and birth tourism is a well documented phenomenon, particularly on the West Coast. If you want to use a First World surrogate, why not take the free American Citizenship upgrade?

      • Deiseach says:

        Having read the article, the Chinese agencies involved claim that U.S. citizenship doesn’t come into it: the parents are wealthy, many already have Canadian or American citizenship anyway, and why would they wait 21 years for the kid to grow up and sponsor them for green cards?

        I do wonder if it’s all part of dodging the legal ban on surrogacy in China; if Chinese courts are likely to rule the surrogate as the ‘real’ mother, then presumably they’d be more favourable to “American citizen” children than children borne by Indians, etc. (the idea that Americans are probably not likely to try to immigrate to China, unlike other Third World countries? Or that if they tried to forcibly repatriate “American” children to the USA it would create an international incident, so they’re more likely to let the Chinese parents keep the child?) Or maybe I’m wildly off course. I am curious how the surrogate child from abroad gets around the “one child policy” – if it’s considered an adoption, is the loophole that you can adopt as many as you like but you can’t physically beget/bear more than one?

        As well, I’d like to note that any aspiring surrogate mothers won’t make $100,000+, that’s the range for the total fees and costs involved. Swiping this off an American surrogacy site, you’ll more likely make around $30,000 for your nine months’ work:

        Gestational Carrier Compensation and Other Expenses $29,600.00:

        Carrier’s base fee (9 payments starting with heartbeat ultrasound) $25,000.00
        Carrier’s IVF transfer payments (for completion of each embryo transfer procedure) $500.00
        Carrier’s maternity clothing allowance (paid at 3 months gestational) $500.00
        Carrier’s monthly allowance (in lieu of itemized costs) ($200/month est. for 12 months) $2,400.00
        Carrier’s post-birth bed rest (following a vaginal delivery) $1,200.

        As always, the lawyers make the most money 🙂

        • ivvenalis says:

          Actually, he didn’t say that citizenship wasn’t a concern. He said that a “green card” (work visa) for the parents wasn’t a concern.

          “Having read the article, the Chinese agencies involved claim that U.S. citizenship doesn’t come into it: the parents are wealthy, many already have Canadian or [Australian] citizenship anyway, and why would they wait 21 years for the kid to grow up and sponsor them for green cards?”

          Then why don’t they go to Canada or Australia? (Maybe they do? Wouldn’t be the first bogus trend article.)

          By the way, “investment immigration” (EB-5) in the United States requires an investment of between $500k-$1m (more for more desirable locations) and employment of at least 10 non-immediate family members for at least two years. ( And that’s for a visa, not citizenship, which requires a further investment of time. That’s a lot more than $100,000, not even including the fact that it would require them to move to the United States, which they may not want to do.

          Anyway, all he has to back up his assertion is “well this won’t pay off for years, obviously no one thinks that far ahead!” Is that a true statement, or a false statement? I say it’s a false statement. I still contribute to a retirement fund I don’t expect to be able to draw on for decades, myself.

          I’m sure the Chinese government doesn’t like dual citizenship. Neither does the American government, actually. The funny thing is though, the United States Government never actually checked whether this particular American was a foreign citizen until they were trying to decide whether I should be entrusted with state secrets, although they did ask me about it when I was commissioned into their armed forces. I believe that if I had been, I would have actually had to get a verified statement from the foreign government’s embassy acknowledging that I was not a citizen.

          Given that, do you think the process of two Chinese people getting Chinese citizenship for their Chinese son in China actually requires the United States Government (who is the sole arbiter of who is and is not actually a United States citizen) to formally revoke the citizenship of an infant? My guess is that the answer is no. Sure, later in life the kid might have to renounce one (or more! “Many” of these Chinese people have Australian and/or Canadian passports! Even though that’s not allowed!) of his citizenships to varying degrees of effectiveness in order to do this or that. But it’s better to have the option, no?

          And sure, American citizenship (not a green card, I find it hard to believe this isn’t intentional misdirection) probably, usually doesn’t matter when applying to high school in China. But it might help when applying to a college or a job in the United States. I hear it also helps if you’re working overseas and the US government decides to evacuate its expatriate citizens back to the homeland. But of course that’s all in the future, not right now, the sole factor in all decisions, so obviously this doesn’t matter.

          • Hanfeizi says:

            “I’m sure the Chinese government doesn’t like dual citizenship.”

            Legally speaking, it is prohibited. Like many matters Chinese, however, there are ways around it- if a Chinese national gains citizenship elsewhere, there’s no reciprocity agreement requiring that they report it, so whatever the de jure status, many Chinese nationals have de facto dual citizenship (in the US, Singapore, Canada, Australia, etc…)

          • Richard Gadsden says:

            Lots of countries prohibit dual citizenship.

            In practice that generally means “if you acquire another citizenship and we get it formally called to our attention, then we will remove your citizenship of our country”. But if you acquire another citizenship and you just don’t tell them…

            The US requires you to renounce other nations when you become a US citizen. That means you say you do in the oath of loyalty, not that you must produce documentation proving that you have abandoned your other nationalities. Since the other nations generally regard that oath as meaningless, lots of people are dual citizens.

          • Nornagest says:

            There are also a lot of people with dual citizenship in the US who have never taken that oath. Usually this happens because of interactions between jus soli and jus sanguinis citizenship rights.

          • AlphaGamma says:

            Nornagest: Exactly. I am a citizen of the United States (jus soli and jus sanguinis) and of the United Kingdom (jus sanguinis only). The situation is reversed for my siblings, who were born after my parents moved from the US to the UK.

    • Ever An Anon says:

      It’s definitely about citizenship.

      In general the Chinese are very practical about this sort of thing. Gaming the system by marrying Americans (preferentially Chinese Americans) isn’t seen as slimy or disreputable the way we’d view it. Having kids in the US and then moving back to China or another Sinophone country is also something you see, particularly with Taiwanese who can travel to and from the US more easily.

      The big tradeoff is that the PRC doesn’t like dual-citizenships. If you’re an American citizen it might be more of a hassle actually living in China. Although then again money would probably help smoothing that over.

      • Hanfeizi says:

        I managed to stay in the country by registering a company with myself as proprietor, then applying for work visas as a CEO. That kept me in the country for six years, though not without some hassles. Now that I’m married to a Chinese with a Shanghai hukou, I could return to China and get permanent residency quite easily, but it’s not quite as easy for people who aren’t Chinese and aren’t married to Chinese nationals to get that (though there is a residency program for foreign nationals with advanced degrees who have worked in China for seven or more consecutive years).

        • Ever An Anon says:

          That is a hilariously cool trick. Can you go into more detail as to how that worked? Was it just as simple as registering a personal corporation or was there a lot of finagling with e.g. paying tax on the money you paid yourself as CEO?

          With tagging onto your wife’s hukou, you mention that it wouldn’t be easy for a non-Chinese to get permanent residency under those circumstances. As someone who might be in that position soon, do you have a guess on what effect being a white American married to a Shanghai girl would be? No plans to move, I like the city too much, but kind of curious.

          • Kevin P says:

            Hanfeizi is wrong here. While the law as written is fairly permissive in terms of permanent residence requirements, in practice it’s almost impossible – last I checked they issued about 1000 green cards per year across the whole country. There are basically two additional unwritten questions on the application forms – how many million is your business worth, and what position is your father in the communist party.

            [EDIT: I just checked the statistics. 4,752 foreigners were granted permanent residence from 2004 to 2014, so less than 500 per year. In contrast Beijing alone had 200,000 foreigners as of 2014.]

            Otherwise the best you can get is a family (Q) visa, but this doesn’t allow you to work legally. Most foreigners married to Chinese are on regular work visas.

    • AnonymousCoward says:

      Am I the only one not on the Dawkins-isn’t-that-great train? Because I can’t help but see him as on the right side of pretty much every conflict so far, his area of expertise or not.

      It seems like it’s becoming a social requirement to roll your eyes at him before complimenting him, and I wonder how many people are doing it because they know he’s unpopular, even though they secretly agree with pretty much everything he’s ever said like I do.

      • Whatever Happened to Anonymous says:

        While I agree that Dawkings gets perhaps more unfairly maligned than necessary, it’s hardly strange for a Catholic to not be very fond of him.

        Personally, I just like posting fedora memes, of which Dawkins is just an unfortunate casualty.

      • Adam says:

        I don’t think it’s hard to see why a religious person wouldn’t like him, but I don’t follow Twitter and my first exposure to him was as a biology student (and biology work was my only exposure to him for a very long time), so yeah, to me he’s a pretty awesome guy. I can only imagine how many great scientists and writers of the past would have been absolute dicks to everyone if they’d had Twitter.

      • lvlln says:

        I felt that his Dear Muslima letter was a poor argument, but otherwise, “right side of pretty much every conflict so far” does seem to describe his history in Internet arguments fairly well in my view.

        • Zorgon says:

          He was on the right side of that one too.

          (“Right” in this context meaning “least filled with ambitious ideologues.”)

      • Urstoff says:

        I don’t follow his social commentary much anymore because I got over my interest in atheism years ago (now I’m just an apathetic atheist). However, The Selfish Gene and The Extended Phenotype are still important and fantastic books that deserve to be read.

      • dndnrsn says:

        One thing that really struck me about “The God Delusion” was that, for all his talk about how bad it is that children are indoctrinated with religion, his soft spot for Anglicanism (maybe high church Anglicanism specifically?) and distaste for Catholicism are exactly what you would expect of someone of his age brought up Anglican (or, high church Anglican, which is usually associated with being fairly well off).

        So, he’s thrown off all the shackles of religion … except for the anti-Catholicism, which in an English context at least is often associated with national prejudice.

      • Deiseach says:

        Oh, I disliked Dawkins way before it was cool or popular to do so 🙂

        It wasn’t his atheism as such (if you’ve got any degree of functioning vision and any exposure to mass media, not to mention the Internet, you’ll see more than enough bashing of religion to do you), if you’re Irish you get used to a certain section of the neighbouring island being sniffy in that particular English way about Catholicism (Philip Pullman is something the same: “Religion is horrible. Well, except Anglicanism. Anglicanism is lovely and fluffy. Of course, that’s because it’s not really religion. BUT CATHOLICISM IS DEMONIC AND THEY SHOULD ALL BE STAMPED OUT!”) Even the bit about having the Pope arrested as a war criminal or something when he visited Britain only made me roll my eyes.

        No, it was politics and history, not religion as such, that got my blood boiling. It was the smug, arrogant, all-too-familiar “Oh we Brits had nothing to do with what went on there” attitude about the North and about Ireland that drove me up the wall. It’s the same idea that “Oh, those crazy Irish killing each other, who can tell why? We English came in with clean hands as neutral referees to keep the peace” that totally ignores the history and politics that bind our two islands.

        He’s very keen to make us Catholics all personally responsible for the Spanish Inquisition and the Crusades. But he shows no awareness of, nor interest in, how his own social class and nationality and religious background (yes, even the lovely fluffy Anglicans) had a part to play in keeping the pot stirred and the violence going in Ireland over the centuries, playing one section off against the other for advantage in ‘the mother of parliaments’.

        A little more realisation that playing the Orange card had something to do with why, even today, Catholics take holidays and leave the North during the Marching Season and a little less “haw haw ain’t it a thigh-slapper” about “Protestant vs Catholic if only they were all atheists there would never have been trouble” on his part would be nice, but I’m not holding my breath waiting for any Oxbridge don and member of the English Establishment to doff the mantle of arrogant assumption of inborn superiority over the mere natives anytime soon.

      • John Schilling says:

        Am I the only one not on the Dawkins-isn’t-that-great train? Because I can’t help but see him as on the right side of pretty much every conflict so far

        Being on the right side of every conflict is not a sufficient condition for greatness. And some people, you don’t want on your side.

        • Samuel Skinner says:

          Why? Even though Hitler is a horrible person, I’m fine with him being on the “smoking harms people” side.

        • chaosmage says:

          Then what’s a sufficient condition for greatness, according to you?

          I hold that Dawkins contributed more to the scientific body of knowledge than most accomplished scientists (especially by inventing memetics and by defending gene-centric evolution against other interpretations of Darwinism) and he’s a borderline brilliant writer who can make previously highly technical evolutionary theory seem self-evident and obvious the second time you read it. That’s an exceptionally rare combination of skills and it would be enough for greatness in my book.

          • John Schilling says:

            Inventing memetics might qualify one for scientific greatness if memetics were a science. And maybe it will grow up to be one someday, but right now it’s at about the level of astrology or alchemy.

            Dawkins is supposed to know what a science looks like, and he’s had plenty of time that he could have devoted to developing memetics but instead chose to piss away on e.g. pissing off atheists to no good end. He’s a cheerleader who gets people to shout, “Yay Science!”, without actually doing any science. Maybe he’s a great cheerleader, but that doesn’t impress me much.

          • chaosmage says:

            You’re flat out wrong on both points.

            Memetics may be a proto-science, because as always happens with social science topics, any idiot can feel they have something to contribute and will generally not be weeded out by academic selection. Still, the evolutionary perspective on idea propagation has quickly developed lots of applications in cultural studies, marketing, history of ideas and other fields, and Dawkins did that. If you don’t think that matters I can only conclude you don’t work in any of these fields. The comparison with astrology and alchemy (both obviously different in being stagnant non-academic fields) I find hard to parse as anything but a crude insult.

            It is ludicrous to say Dawkins hasn’t been actually doing any science. The guy’s been writing evolution papers since the sixties, and stopped only when he was well over 60 years old, like most professors do. The Extended Phenotype alone was a seminal contribution, and it built on the meme metaphor in a rigorous way. Being a cheerleader for science was his actual job (1995 to 2008) and he continued to do science anyway. And after he’d done science for four decades he got into The God Delusion and what came after as basically a retirement hobby. Do you claim to know what you’re talking about here?

            And you did not answer the question. Who impresses you then, if Dawkins doesn’t? What’s a sufficient condition for greatness?

  76. J says:

    Jonathan Haidt’s article about a paper about the transition from “honor” to “dignity” to “victimhood” societal norms:

    • Nita says:

      the paradox: places that make the most progress toward equality and diversity can expect to have the “lowest bar” for what counts as an offense against equality and inclusivity

      Uh, where’s the paradox? How else could things possibly work?

      I’m a big fan of dignity culture, but unfortunately it has no solution to long-term, low-grade conflicts based on the denial of the dignity of one of the parties — e.g., bullying. Falling back on the honor culture solution (violence) does work in some cases, but not all of them.

      The authors write:

      People are to avoid insulting others, too, whether intentionally or not, and in general an ethic of self-restraint prevails.[..] People might even be expected to tolerate serious but accidental personal injuries.

      So, the dignity norm is: try your best not to insult anyone, even if you don’t understand why they find it insulting — in return, gracefully tolerate mistakes made in good faith. What happens when some people end up having to tolerate a lot more mistakes than others?

      And what happens when other people loudly declare that they will no longer practice self-restraint or apologize for accidental insults?

      • Deiseach says:

        I agree that’s not paradoxical; as you achieve the big targets in promoting equality and reducing discrimination, your bar gets set lower and lower (because it never stops: achieve gay rights, and the push for trans rights gets stepped up because queer activism has to keep moving forward; same for feminism and anything else you like).

        So over the years it moves from “Look, all we want is to not be arrested for consensual sexual practices in private” to “We demand punitive compensation for refusing to provide flowers/cake/photographic services for our wedding!” and “Don’t use offensive language when talking about other humans” to “You can’t use ‘gy*psy’, that’s an ethnic slur! So we’re going to refer to the giant mecha suit of armour as something other than Gypsy Danger, even though that’s based on the phonetic alphabet and not meant to refer to the Roma/Romany/Romani/Rroma/however the flavour of the month spelling goes now”.

        No longer casually referring in the middle of a novel to black people as [redacted] is a good thing and I approve. Getting your knickers in a twist about “gypsy” is a bit precious if you’re not a Traveller, Roma, or person of similar ethnic/cultural background.

        • roystgnr says:

          The phonetic GD would be Golf Delta.

          The screenwriter says that Gipsy came from the WWII-era De Havilland engine of the same name.

          • Deiseach says:

            I stand corrected and thank you for that, but I still think it’s gross over-sensitivity. I think currently there is a lot more discrimination towards Roma/Travellers/gypsies of a worse kind than using out-moded terms.

            I agree calling someone a “gyppo” is a slur; talking about a giant suit of alien-fighting mechanised armour that is intentionally alluding to Second World War aircraft is taking the delicacy of feeling up a notch too high.

            What are the suggestions for other languages to drop terms like zigane, Zigeuner, etc? Is “Bohemian” also an offensive term?

            If people of the ethnic background find such terms offensive because of usage in certain cultural contexts, that’s fine. But very young white Western European-ancestry folk earnestly looking for offence where none was intended – the term is problematic? Thank you for educating me, but ease off a little on “You must NEVER EVER say this and if you do YOU ARE A HORRID RACIST”.

        • Nita says:

          Apparently, the mecha suit is named after de Havilland Gipsy, which, in turn, was named after you know who. The author was asked about this, and responded with an apology for any offense inadvertently caused, which was held up as an example of excellent conduct. Victimhood culture or dignity culture?

        • Saal says:

          Not trying to be offensive here, but…what’s the big deal with calling a Roma person Gypsy? As I understand it, “Gypsy” comes from a misunderstanding wherein European folk thought they were originally descended from Egyptian emigres, yes? So it shouldn’t be any more offensive than referring to a Native American as an “American Indian” or just Indian, which seems more inaccurate than rude.

          Anecdotally, My grandmother is the child of Gypsies who emigrated to the US and settled down, and she felt that explaining the misconception and leaving it at that was enough.

          • Jaskologist says:

            Complaining about microaggressions and cultural appropriation pretty quickly morphs into complaining about the use of any term at all associated with a given group. That’s why one of the centerpieces of the “victim culture” discussions is a girl biting a guy’s head off for using the word “futbol.”

            (Sadly, this also removes the primary motivating force for learning a new language: to better hit on girls.)

          • “As I understand it, “Gypsy” comes from a misunderstanding wherein European folk thought they were originally descended from Egyptian emigres, yes?”

            As best I can tell, it comes from the claim of some of the earliest Roma to show up in western Europe that they were descendants of people from Little Egypt in Greece.

          • keranih says:

            but…what’s the big deal with calling a Roma person Gypsy?

            Practically, for all people: The tone of voice when you say it, and the tone of voice of the last person you heard say it.

            (IE, don’t expect human people who have been called “upper-class rich” in a sneering voice to not tense a little when you use the phrase in an otherwise neutral sentence. We are creatures of habit, and we do identify patterns, even when the patterns are not there.)

            For activists: How much utility are they going to get out of making a fuss?

          • Viliam says:

            I heard a version — not sure how reliable — that Roma first moved from India to Egypt, stayed there for a generation or two, and then moved again to Europe.

            Their Indian origins were rediscovered centuries later, by comparing their language to a language of some group in India.

          • gattsuru says:

            Not trying to be offensive here, but…what’s the big deal with calling a Roma person Gypsy?

            Think of it akin to calling a gay person a f******. There’s very well a certain overlap between gay men and two sticks being held together, and even some gay men that use the term in certain contexts… but it’s still a word you don’t use in mixed company.

            Largely that’s because language is a social construct where underneath the ‘real’ definition of a word there’s an amorphous blob of semi-shared understanding and common knowledge, and for a lot of places that actual Roma live the giant amorphous blob of shared understanding is just a giant pile of negative loading. Even if you yourself aren’t intending to use the negative loading — even if you’re not in a culture where that loading exists, or aware that the negative loading exists anywhere — people you encounter will, especially on an increasingly globalized online world.

            ((This is separate from the question of whether avoiding terms is a /good idea/, just that there’s a reason behind it.))

        • HlynkaCG says:


          If it never stops it stands to reason that there can be no such thing as gross over-sensitivity.

          Today’s accidental insults will be tomorrow’s hate-crime.

      • Taradino C. says:

        How else could things possibly work?

        Well, here’s one possibility: as discrimination became less frequent and severe, people could spend less time and energy fighting it. They could take offense in proportion to the size of the problem, instead of maintaining a constant level of offense while the problem shrinks.

        • Nita says:

          Well, they do seem to take much less drastic action — compare tweets and administrative complaints to Black Panthers or Nation of Islam.

        • Jaskologist says:

          Shouldn’t we be comparing those to Black Lives Matter instead?

        • Mary says:

          They could, in fact, get lives.

          One would think the POINT of getting rid of discrimination was to let people get a life.

          • HlynkaCG says:

            What if “the struggle” is their life?

          • Mary says:

            Then they are exactly the sorts of people whom “Get a life” was invented for.

            The point of struggle to obtain peace. Struggling for struggling’s sake is evil, and avoids being as bad as war for war’s sake only in doing less harm.

          • sweeneyrod says:

            Usually the only way something can be less bad than something else is by doing less harm.

          • HlynkaCG says:


            I agree with you, I think that you need to consider the fact that struggle carries a great deal of social cachet people will loath to give that up and that same cachet will draw others to take up the cause.

            As I said up-thread, today’s accidental insults will be tomorrow’s hate-crimes.

          • keranih says:

            @ HlynkaCG


            *ahem* Responding for a friend. 🙂

          • Mary says:

            ” I think that you need to consider the fact that struggle carries a great deal of social cachet people will loath to give that up ”

            And what consideration does such contemptible reasoning deserve? I have considered it, indeed, more than it deserves.

      • gbdub says:

        “So, the dignity norm is: try your best not to insult anyone, even if you don’t understand why they find it insulting — in return, gracefully tolerate mistakes made in good faith.”

        The “in return” part seems to be the major current failure mode. Heck, the whole point of “microaggressions” is that they are still aggressive and intolerable even if no harm was intended.

        • Nita says:

          From a certain point of view, “microaggression lists” can be seen as a good-faith effort to inform individuals about potentially insulting phrases — sort of like those “Business Etiquette Tips For International Travel” published in books or online.

          In that light, the loud outrage about these lists looks like defecting from the dignity norm.

          • houseboatonstyx says:

            @ Nita
            From a certain point of view, “microaggression lists” can be seen as a good-faith effort to inform individuals about potentially insulting phrases — sort of like those “Business Etiquette Tips For International Travel” published in books or online.

            Those certainly have a much better presentation. Labeling certain phrases as ‘micro-AGGRESSIONS’ is already begging the question, and attributing a problematic motive to the speaker. This does not indicate good faith.

          • Nita says:

            In a world where being polite and kind is seen as uncool, and there’s no material compensation for it (unlike in business), how would you persuade people that it’s important?

            Yes, “microaggression” is a suboptimal term. It’s based more on the effect on the receiver in a certain social context than on the intent on the speaker. Whoever dragged it out of obscurity for this purpose should feel bad about it.

            But, suppose that you were tired of being asked “but where are you really from?” or “are you angry because you’re on your period?”, and you wanted to make it stop. What would you do?

          • keranih says:

            But, suppose that you were tired of being asked “but where are you really from?” or “are you angry because you’re on your period?”, and you wanted to make it stop. What would you do?

            Get better people as friends.

            Srsy. What kind of jerk keeps making “jokes” like that about people they know?

            If one is in a work position where one is dealing with the tourist public constantly, I’d come up with a quick reply along the lines of “Cairo. The one in Illinois, not Egypt. My mom’s folks were from the west side of Korea. Where are your folks from?”

            A politely-spoken question deserves a polite answer.

            The “are you on your period”, otoh – this is juvie “humor” and it’s generally trying to deflect attention from whatever earned ones ire.

            Responding directly to it permits the deflect. Better to ignore it, or develop a quick snap to the effect of “no – and you’re lucky. Because if you’d pulled such a stupid stunt on a PMSing woman, you’d have gotten killed. Now quit being an idiot and fix this problem.”

            what would you do?

            Assume good faith. Act and speak in charity. Be clear and polite, even under provocation. Focus on the larger picture and the big things. Assume everyone else around you is being similarly scuffed – or has been, or will be – by the human herd.

          • houseboatonstyx says:

            @ Nita


            — One
            Child: “Why do rap singers say n_____ when nobody else does?”
            Parent: “Why do you want to say it?”

            — Two
            A: “I don’t support SJWs.”
            B: “Really? Why are you opposed to justice?”

            — Three
            A: “ I don’t support PC language.”
            B: “But it’s nothing more than normal decency and politeness, and not hurting people’s feelings, you bigoted asshole.”

            In Two and Three, A is using words in their current central meaning. B retreats to a motte of triviality to shoot an ad hominem at A. In One, the parent skips the motte and goes straight for the ad hominem. It’s a destructive message: “A logical question will be answered with a confusing and shaming dismissal” — which can lead a child torn between doubting his parent’s sanity or zis own, and giving up all logic in despair.

            There are other strands in these fallacies that I’m too lazy to sort out, and would feel perhaps impolite doing it with yours.

          • houseboatonstyx says:

            @ keranih

            Right. Your method works very well against sub-text, too.

            “I saw you at the movies with X.”

            “It’s pretty good most of the way through, but the ending was silly, because blah blah. – See you later.””

            Istr Miss Manners using something like that, too.

    • Nita says:

      Also, “victimhood culture”? Perhaps the quest for a more “scientific” sociology would be aided by choosing less emotive terms. On the other hand, that wouldn’t help them catch the outrage train.

      • David Moss says:

        What alternative term would you suggest for a culture that is literally about claiming victimhood?

        • wysinwyg says:

          Is there a culture that is literally about claiming victimhood?

          I mean, I’m not arguing that there isn’t a culture that frequently and ostentatiously claims victimhood. I’m questioning whether that actually defines the culture. Not sure it does.

          • Taradino C. says:

            As defined, victimhood culture is a culture in which people frequently claim victimhood because doing so in order to get third party support is the primary legitimate way of resolving conflicts, and victims are seen as having the moral upper hand even if victimhood isn’t exactly something to aspire to. I think the term fits.

          • Urstoff says:

            It seems to me that a lot of the nonsense going on in college campuses is middle-upper class (mostly white and female) students trying to protect third parties who they see as victims.