More Links For June 2014

Nassim Taleb’s List of Aphorisms, Rules, and Heuristics. 135: “Studying neurobiology to understand humans is like studying ink to understand literature”.

I have spent enough time around Finns that I find the Finnish Problems meme pretty funny.

An angle of the global warming discussion I hadn’t heard before: US greenhouse gas emissions have fallen ten percent in the last ten years; world still doomed because China.

Related: Deforestation in the Amazon is down by 75% in the past twenty years. I’d love to read a very good explanation of how it was accomplished; it seems to involve some sort of interventions in supply chains but I’m not sure how or why those interventions were made.

A couple weeks ago I blogged about how monetary windfalls to the poor don’t last very long and definitely don’t help the next generation in the way the idea of “poverty traps” would imply. But I was short on randomized controlled trials in a modern setting. Well, one just came out: Human Capital Effects of Anti-Poverty Programs: Evidence From A Randomized Housing Voucher Lottery. Result: when families get free public housing, there’s no difference in outcomes between their children and children of families who didn’t.

Related: Four generations after Cornelius Vanderbilt made his hundred billion dollar fortune, his descendants are firmly middle class. But other very wealthy families seem to do better.

Someone in the genetics blogosphere asked an interesting question: when Pol Pot killed all the intellectuals in Cambodia, murdering a third of the population, did that permanently depress Cambodian IQ for genetic reasons? As far as I can tell, the answer is no one has really checked. (Edit: Jason Malloy looks deeper)

From our Department Of Cats And Dogs Snuggling: progressive crusader Ralph Nader and anti-tax crusader Grover Norquist are newfound friends and working on a project of left-right convergence.

A snide objection I’ve heard to the free market is that cable companies are capitalist and look how terrible they are. But one sign of how far cable is from an ideal market is the new poll finding over half of people would switch cable companies if they felt like they could.

Study: There is no silver bullet solution to fix climate change. I will come out and say it: I hate studies like this, because it’s inconceivable anyone would announce “there is a silver bullet solution to fix climate change”, so I haven’t learned anything new. But one thing did encourage me. Apparently the economic lure of carbon offsets is now so great that companies are carrying out legally ambiguous unauthorized experiments. I feel like once we reach the point where corporations are going behind government’s back to try and come up with novel lucrative climate change solutions, the fight against climate change is in damn good shape.

Is Casual Sex On The Rise In America? The answer appears to be “yes, kind of, but we’re still going to make you feel bad for believing that it is.”

Possibly the most metal band in the world: NǽnøĉÿbbŒrğ VbëřřĦōlökäävsŦ (Nanocyborg Uberholocaust) plays “ambient cosmic extreme funeral drone doom metal” and records their songs at the South Pole.

the impotence trials of 16th century France. At this point I think we just need to admit that everything about sex in historical times was horrible and oppressive and terrifying no matter what gender you were.

American Enterprise Institute reports on the Wal-Mart in a North Dakota oil field paying $20 an hour because workers are scarce. I don’t agree with the particular conclusions they draw, but it’s a useful reminder of the laws of economics and that the low wages of workers is neither a universal truth nor a morality play but a reflection of the contingent economic conditions prevailing in the relevant area.

Various levels of panic about one group getting 51% of the Bitcoin mining power and all the nefarious things they could do thereby. People reasonably point out that they’re not going to, because that would crash the price of Bitcoin and make all their hard work useless. The real story here seems to be that this is possible, which means if the government ever wants to destroy Bitcoin, instead of going through some weird cyberpunk inquisition all they have to do is get the NSA to build the appropriate number of supercomputers and take it over. I expected better from Satoshi Nakamoto.

There is a World Pun Championship, and it is absolutely terrible.

Skeptic community blogger wants to drink a homeopathic preparation of HIV positive blood to spread awareness of how homeopathic preparations don’t retain any of the active ingredient in them. Skeptic community throws a fit about reckless self-endangerment. So, uh, I guess that awareness is getting spread already. No word on whether he’s going to spread awareness of fact that you can’t actually get HIV by swallowing it.

A poll asks consumers whether, in the case of a sudden crisis, they would want their self-driving car to save their own life or maximize the total number of lives saved. For example, if it was about to hit an oncoming car and couldn’t get out of the way, should it veer off the road and down a cliff (100% chance of killing self, 0% chance of killing others) or hit the oncoming car (75% chance of killing self, 75% chance of killing others)? Respondents overwhelmingly in favor of cars programmed to protect their driver alone – which makes no sense, since presumably everyone’s car will have the same programming so this kills extra people for no reason. Immanuel Kant is not amused. (h/t Carl Shulman)

You know tilt-shift photography? The kind that can make anything look like a miniature? Apparently there’s an opposite process that makes things look huge, and it’s pretty scary.

Job Interviews Reward Narcissists adds to a growing body of evidence that job interviews are terrible and strictly inferior to judging candidates based on their accomplishments or test scores. I wonder if the social justice people can be directed to attack companies using job interviews? They’re the number one opportunity employers have to be influenced by irrelevant criteria (like attractiveness, race, class, et cetera) and there should be a straightforward profit-maximization case for getting rid of them.

From the Department of Help I Don’t Know What To Believe Anymore: Noah Smith writes:

Japan’s Shinzo Abe is the world’s best leader…Japan has escaped deflation. The stock market is up, growth is way up and even wages are finally starting to rise. In other words, unlike everyone else in the world, Abe listened to Milton Friedman, and the results are looking good.

Meanwhile, from Zero Hedge:

Japan is what a Keynesian dystopia looks like. Its entire economy is now hostage to a fiscal time bomb. Namely, government debt which already exceeds 240% of GDP and which is growing rapidly because even the recent traumatic increase in the sales tax from 5% to 8% does not come close to filling the fiscal gap. Moreover, even at today’s absurdly low and BOJ rigged bond rate of 0.6% nearly 25% of government revenue is absorbed by interest payments. Now comes the coup de grace. Japan’s savings rate has collapsed and its vaunted current account surplus is about ready to disappear. Japan launched upon the greatest experiment in Keynesian fiscal stimulus ever imagined. The catastrophic results speak for themselves and are a potent remainder that bad ideas can wreak immense damage once they are embraced by the machinery of the state.

I’m okay with people disagreeing on implications, but it’s annoying I can’t even get a clear picture on whether a country’s doing miraculously well or dystopianishly bad. I’m going to follow my usual heuristic of trusting the person who isn’t Zero Hedge.

A while back some reactionaries suggested that some supposedly oppressed groups are actually more privileged, because one sign of being privileged is that they get extra rights and respect and you get in trouble for insulting them, and it’s much less acceptable to say mean things about minority groups than it is about white men. I argued they were conflating a lot of different possible ways in which different groups can have more rights than each other. But I think a recent article by Robin Hanson offers a much stronger counterargument; he says that allowing something to be mocked is a form of countersignaling proving high levels of respect for that thing.

Carl Shulman, who wrote some excellent posts about parapsychology that inspired some posts here is back, with a suspicion that scientific fraud may be common enough to seriously affect the literature.

You know how everyone says the research shows men and women aren’t too different in terms of personality after all and there’s huge overlap? Now a new group of scientists crunches the data slightly differently and finds men’s and women’s personalities are hugely different, practically no overlap. Thus the circle is complete: both sides of the debate have a study to cite and can accuse the other of being “science denialists”. I hate this sort of thing, because it’s clear that I would need to study large amounts of very complicated math and develop strong opinions on statistical minutiae before I can have a reasoned opinion on this subject, plus after I do that whichever side I disagree with will still tell me I’m wrong and dismiss the other side’s paper based on the first nitpick they can find.

Marginal Revolution: “Economics assumes that people are rational, self-interested, lightning fast calculators. Obviously a bad assumption, as we are constantly told. Chimps, on the other hand, are rational, self-interested, lightning fast calculators.”

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

348 Responses to More Links For June 2014

  1. Andy says:

    The “impotence trials” link is broken and leads to a 404 page.

    Re: Reactionaries and mockery, wouldn’t that suggest that old-timey Southern minstrel shows mocking black people for being stupid and brutish (“blackface” tropes, etc) indicated a high level of respect for black people?
    Perhaps “acceptability of mockery” has an uncanny valley effect – things that are low-status can be mocked, and things that are high-status can be mocked, but things in the middle cannot.
    Or as Hanson says:

    I should note that in simple models counter-signaling there are three types, and the same signal is sent by the high and low type, which is a different signal from the mid type. So yes there are also low status groups today, like animals, which one is allowed to lampoon.

    So in the Reactionary idea that fathers on TV shows are often portrayed in a mocking way – Homer Simpson forex, either we’re denigrating fatherhood or gently mocking it, and it’s really hard to tell the difference.

    • Sniffnoy says:

      The “impotence trials” link is broken and leads to a 404 page.

      To be clear, it’s broken because of a stray <br> tag, not because the article has vanished. (Vanderbilt link broken for same reason.)

    • Anonymous says:

      Just a “valley effect”, yeah? No particular connection to the uncanny valley.

      • Andy says:

        I accidentally forgot to type “uncanny.”
        But this could be an effect of the San Fernando Valley…
        *silence*
        …I’ll see myself out.

    • Eli says:

      the Reactionary idea that fathers on TV shows are often portrayed in a mocking way – Homer Simpson forex

      That’s not really a distinctly political idea of any kind. What’s actually going on is that the mocking portrayals of fathers are parodies of previous admiring portrayals of fathers, and the parodies became funny and popular enough on their own that people forgot what was originally being mocked.

      Homer Simpson is a mockery of Leave It To Beaver, but nobody even remembers what that was any more.

    • nydwracu says:

      Crocodile humor: using mockery to further lower the status of things that are already low-status. Hanson accounts for this:

      I should note that in simple models counter-signaling there are three types, and the same signal is sent by the high and low type, which is a different signal from the mid type. So yes there are also low status groups today, like animals, which one is allowed to lampoon.

      So if you can mock something, it’s either high-status or low-status. I think it’s much more complicated than that: high-status mocking is different than low-status mocking. High-status mocking is about acknowledging flaws to signal that they don’t matter — you give your father a Father’s Day card about how he’s lazy and spends all his time drinking beer and fishing to signal that you wouldn’t care if he were lazy and spent all his time drinking beer and fishing, because his status to you is that high — whereas low-status mocking is about tearing into the object of mockery, making it seem as flawed as possible to provide as many rationalizations for dismissal/status-lowering as you can.

      White mockery is the latter type.

      Also, respect isn’t power — and some people would rather have power than respect. The Arthur Chus of the world don’t care whether they’re liked; they care whether they can force people to sing praise of them, even if the praise is faked.

      • Jake says:

        I think the high status and low status varieties of mockery have more to do with the feelings of the group being mocked rather than the ones doing the mocking – and thus whether the mockery is malicious or friendly is irrelevant. What matters is whether those being mocked have the power to do something about it, and whether they use that power. So we break down into:

        Ultra-Low status – Groups that are mocked and have so little power that they can’t do anything about it. At this point there aren’t really any racial groups in this category in the US anymore, but this was basically the situation of the black population for most of the country’s history – minstrel shows and such.

        Low status – Groups that have some political power, and use it to quell mockery. This includes to varying degrees pretty much every racial/religious/sexual sub-group in the US except white Christian men.

        High status – Groups with tons of power, but who counter-signal by not trying to prevent mockery, showing that they’re so far above such things that it doesn’t even matter.

        Also, unless I’m missing something, the ‘white mockery’ link you provided seems to be a genocidal anti-white rant that you yourself wrote. Not 100% sure what we’re supposed to get out of that.

        • Zorgon says:

          It comes just after the “Fnord” article Scott linked a few weeks back, so I’m guessing it’s one or more genocidal anti-white rants from elsewhere put through the fnord filter.

          I agree it’d be clearer if nydwracu would routinely link their sources, however.

          • Jake says:

            Haha, wow. That article really is the purest form of content-less outrage bait isn’t it? “Here’s a bunch of mashed together sentence fragments from all over the web, connected only by being bad stuff that THOSE PEOPLE are saying about US.”

        • Nornagest says:

          It reads to me like a fnorded compilation of several sources.

        • nydwracu says:

          People who want to find the sources can Google the phrases to find them, and people who want to dismiss it out of hand can complain that the sources aren’t linked.

          And, as it turns out, people who want to dismiss it out of hand can call it a “genocidal anti-white rant” before realizing that it’s a series of phrases from articles in popular news outlets and the lyrics of a song written by one of the most successful independent bands of all time — a band which happens to have connections to the Democratic Party.

        • peterdjones says:

          There are still ultra low status individuals, even if there aren’t any ultra low status groups….the most unpopular 5 or 10 kids in high school, for instance.

          Non affectionate mockery of the powerful by the powerless is a thing too….

        • It’s always less total work for the person who posts to supply citations than for readers to look for citations.

      • Ghatanathoah says:

        I suspect that there is white mockery of both types. The question is, which is mainstream and which is on the fringe?

        I suspect that good-natured mockery (i.e. Stuff White People Like) is mainstream and that poisonous mockery is marginalized fringe stuff that only occurs in specific protected spaces, like “Studies Departments” and extremist churches.

      • Nestor says:

        reminded me of Louis CK’s sketch on insulting white people “You can’t even hurt my feelings!”

        https://www.youtube.com/watch?feature=player_embedded&v=TG4f9zR5yzY

    • peterdjones says:

      As I tried to point out on the OB thread, mockery can range from cozy in-jokes to vicious put downs.

  2. Andrew says:

    I assume those wal-mart employees are getting $20 an hour because employees are scarce, not jobs.

    Also, I thought GiveDirectly had produced decent evidence that monetary windfalls are extremely beneficial. As in, years after the transfer they’re making and consuming substantially more. I can certainly see a monetary windfall being more effective when you’re eating-dirt poor vs can’t-afford-cable-tv poor, but I’m surprised that the effect apparently disappears entirely.

    • Samuel Skinner says:

      “Also, I thought GiveDirectly had produced decent evidence that monetary windfalls are extremely beneficial.”

      That is short term benefits; the company has only existed for 6 years. Technically it could be long term benefits as well, if the influx of money raises overall prosperity, but the evidence shows individuals do not generationally benefit from large scale transfer payments.

    • Oligopsony says:

      It could be that poor Americans are Deserving Poor Marshmellow Test Failer Scum and rural Kenyans are a statistically average sample of humanity that just happens to be stuck in rural Kenya, such that the latter are able to benefit more. It’s almost certainly the case that there’s more low-hanging fruit available to the latter group.

      • Jake says:

        I think the low hanging fruit aspect is a big deal. I often hear about people in the third world being given some seemingly tiny amount of money, say $100, and using it to start a business. In the US with a more developed economy most niches are sufficiently filled that some investment is needed to stand any chance of breaking through.

    • Kaj Sotala says:

      I can certainly see a monetary windfall being more effective when you’re eating-dirt poor vs can’t-afford-cable-tv poor, but I’m surprised that the effect apparently disappears entirely.

      Hmm. What does the impact of the Marshall Plan tell us about the impact of large-scale cash transfers to people who aren’t-quite-dirt-poor? Wikipedia seems to say that economists generally agree about it having been beneficial and of having sped up the European recovery after WWII, even though the recovery would probably even have happened (but more slowly) without it.

      • Ken Arromdee says:

        The Marshall Plan involved transferring money to people who suddenly became poor because of immediate external factors. People who became poor because of immediate external factors don’t necessarily get affected by money the same as the chronically poor.

    • James says:

      I see at least two plausible-sounding, tightly-linked explanations for this, and combined they convince me that this result makes sense.

      Cultural expectations: It’s pretty obvious how to improve your family’s life if you can’t afford to send all your kids to primary school, or buy the necessary capital to sustain yourself with anything besides farming; much less so if what we think of as “the necessities” are already covered. What’s the most wealth-enhancing way for an American family of four living on $25k/year to spend a $100k windfall, and how likely do you think it is that your typical American family of four making $25k/year thinks to do that and also actually does it, rather than buying consumer electronics and a nice car or saving it for a college education that their kids likely would have been able to at least attempt anyway? (Trade school? CFAR workshops? I actually don’t know, but I don’t think it involves an iPad or homeownership.)

      Bottlenecks: If you live in an underdeveloped country where almost everyone is dirt poor, chances are pretty good that the limiting factor in your lifestyle is money. If you’re a relatively poor adult in a wealthy country where lots of people are middle class or wealthier, many of whom came from relatively poor families themselves, it’s much more likely that you’re ultimately limited by your own abilities; otherwise you probably would have done well enough in high school, gotten loans and/or scholarships to go to college, picked a lucrative major out of fear of being poor yourself, found a well-paying job because you’re a capable person by hypothesis, and ended up not-poor. P(stuck in actual poverty trap | poor adult in poor country) >> P(stuck in actual poverty trap | poor adult in rich country).

  3. Carl Shulman says:

    “Carl Shulman, who wrote some excellent posts about parapsychology that inspired some posts here is back, with a suspicion that scientific fraud is more common than we are usually willing to believe.”

    In that post I cited a survey of psychologists: the median respondent estimated 5% committed fraud (mean 10%). I was mainly talking about how, when one takes selection effects into account that rate is enough for fraud to substantially affect literatures (as frauds can disproportionately get published, disproportionately dominate null fields, and can claim that their fake data were generated without p-hacking). I don’t defend any very contrarian view about fraud prevalence, relative to the literature, save for agreeing with the caveat that audit studies offer lower bounds and we should take into account that most escape detection.

  4. Stephen says:

    Since this is sort-of close to an open thread, I figure I should point out that the linked paper in your blog header (http://BIT.LY/1P64MAS) doesn’t work because it’s all caps, and bit.ly is case-sensitive. Luckily, there were only four letters in the link, which gives only 16 possible case-sensitive variations, so I tried them in succession to get the actual link. Turns out it’s http://bit.ly/1p64mAs, for anyone else wondering.

    (naturally, it took 14 tries to find the correct one)

    • Douglas Knight says:

      It’s upper case as a css effect, so if I copy and paste, it still works. uh…why didn’t you copy and paste?

      • Elissa says:

        Copy and paste gets all-caps for me. Maybe a browser thing? I use Chrome.

        • Douglas Knight says:

          Oh, that’s it. I’m using Firefox. I actually thought I had checked in Chrome because it was there that I looked at the google cache, which reveals the case.

        • Scott Alexander says:

          Aha! I hadn’t paid much attention because I figured who the heck types things by hand instead of copy-pasting. Now that I know Chrome mangles this somehow I’ve replaced it with a case-insensitive version.

        • Anthony says:

          Copy and paste, or right-click “Go to web address” work for me in Opera (12.17, not the chrome-plated versions). The right-click thing worked with the bit.ly address, too.

  5. Douglas Knight says:

    I expected your Cambodia link to be Jason Malloy from this week, because he mentions Pol Pot, unlike Lindsay. He also quotes Flynn estimating that the effect of the Khmer Rouge was 1 point. Which would not be measurable, even if we had tests from before the genocide. Malloy’s main point is to gather more recent tests.

    • Scott Alexander says:

      Hm, not sure how I missed that. One point would be very very low. If killing all the most intelligent people in a population only lowers IQ by one point a generation later, that, er, puts a lot of stuff about dysgenics into perspective.

      • g says:

        To be clear, one of the reasons why the estimate is only 1 point is that Pol Pot *didn’t* kill “all the most intelligent people” — the killing was based on occupation rather than intelligence tests, it wasn’t done with perfect consistency, etc. Flynn estimates a larger effect for actually killing the smartest quarter of the population.

      • >If killing all the most intelligent people in a population only lowers IQ by one point a generation later, that, er, puts a lot of stuff about dysgenics into perspective.

        Could you expand on how a lot of stuff about dysgenics gets put into proportion?

        • Scott Alexander says:

          A lot of people tend to worry that some social trend (like college graduates having fewer children) will cause a massive collapse in IQ by lowering the level of good IQ genes. But if the nightmare scenario – killing everyone with good IQ genes – is only able to lower IQ one point, it seems a lot less worthwhile to focus on the little things.

        • Ialdabaoth says:

          In fairness: there’s a difference between killing all the high-IQ people in one generation, and killing / suppressing the breeding of a portion of high-IQ people across multiple generations.

          • Jake says:

            Well sure, but Khmer Rouge killed something like 2 million people in a country of over 7 million. That’s a pretty gigantic effect compared to the fairly small correlation between low IQ and fertility seen in the US.

        • Douglas Knight says:

          Lynn’s estimate is a loss of 1 point per generation. Not a fast-moving problem, but it is happening every generation, unlike genocide.

  6. meyerkev says:

    So my thing with ZeroHedge is that they’re Chicken Little the website. The sky is always falling. And when the sky is falling, they’re right (Among other things, I believe that they were some of the first people to call out just how stupid the housing market was). But since this is the equivalent of a stopped clock being right twice a day, they’re not so much wrong as useless.

    With that said, yeah, Japan’s got some issues. Short version is that they did 20 years ago what China’s doing right now (And this should scare you). Non-performing loans covered up by artificially low interest rates, the Asian growth model, serious demographic issues, the lot. Which meant that they ended up with debt at about 200% of GDP. And zero growth while they worked the issues out of the system for about 20 years.

    So now they’re in this weird trap. As it stands, they can ignore the debt bomb as long as interest rates are practically zero. Sure, the principal is huge, but the interest is tiny, and thus once 40% of the country is on the retirement plan demanding pensions and medical care, they can more or less spend to their heart’s content without driving up the interest payments.

    But they can’t grow the economy without inflation (for reasons that I honestly don’t get) which raises interest rates. If they raise interest rates, then that enormous debt bill stops being ignorable. 250% GDP * 0.6% interest is doable. 250% GDP * 5% interest isn’t.

    And since the debt bill is SO huge, it’s unlikely that they can do what America did in the 1950’s and 60’s, where the economy/inflation grew faster than the debt did, so that even as debt exploded, debt as a percentage of GDP dropped.

    Which is why I wouldn’t be surprised if Japan defaulted at some point in my lifetime.

    • Papermoon says:

      Japan’s problem is not debt, it is economic stagnation and that is what it’s trying to address.

      The most important thing is that debt is not a bad thing and that big debt is not more of a bad thing. Especially in Japan’s case.

      Firstly it has absolutely no problem finding buyers for it’s debt (not really a surprise in a stagnating economy) and secondly it can of course purchase it’s own debt indefinitely. So it cannot default, full stop.

      The idea of “artificially” low interest rates are problematic as well, the only idea i have of “natural” interest rates is this:
      http://research.stlouisfed.org/publications/mt/20050301/cover.pdf
      And it obviously doesn’t apply in Japan’s case.

      When you say that you need inflation to grow the economy, what that means is that when interest rates are below the rate of inflation you get a “negative real interest rate”. This encourages borrowing and investment, basically it stimulates the economy. If inflation is very low then it compromises the governments ability to stimulate the economy.

      Debt is for some reason an emotional issue and a lot of people simply lash out at it. But an economy needs money and that money has to be supplied at a certain rate. If you think that rate is too slow then you speed it up, by printing money, issuing debt, etc.

      That’s what’s happening in japan now. It also happened in the US after the most recent crisis and all the same arguments about a debt bomb and artificial inflation rates proved false then.

      • anon says:

        Under what conditions can debt be bad, and why? Why are you confident Japan won’t reach such conditions once their demographic crisis hits?

        I’m neither anti-Keynesian, nor pro Keynesian.

    • Jake says:

      Fortunately while inflation raises interest rates, it also reduces the value of the debt, since it’s denominated in yen. Increased inflation almost by definition is helpful with debt repayment, whatever other issues it may have.

      • Anonymous says:

        If you have long term fixed-interest debt, then, yes, inflation makes it easier to pay off. If you have short-term debt, there’s no reason to expect inflation to help, because the interest rates will just keep up with the inflation. But, yes, it’s not clear it should hurt, either.

    • Anthony says:

      As I see it, the problem is that a) Japan has fewer and fewer people who are capable of being productive, and b) at very low internal interest rates, there’s very little appetite to invest in the additional capital needed to raise the productivity of the existing workers in Japan.

    • Douglas Knight says:

      When do you mean by calling the housing market? ZeroHedge was founded by bankers who lost their jobs in 2008, due to the subprime mortgage crisis. Isn’t that too late to call it?

  7. Douglas Knight says:

    I’m okay with people disagreeing on implications, but it’s annoying I can’t even get a clear picture on whether a country’s doing miraculously well or dystopianishly bad.

    They agree on what policies are right, they just disagree on why the example of Japan supports their claim. Since they agree on what’s ultimately important, what’s the problem?

    Shouldn’t you be more concerned that they disagree about what policies were followed than about scoring the outcome? I don’t see any factual disagreement. Noah Smith makes the hypothetical claim that they avoided deflation. That justifies almost any policy, but only if you agree that they were close to deflation. Zero Hedge claims that they have a time bomb. Wait and see.

    So in ten or fifty years, we’ll know whether Abe was a good or bad leader. But we still won’t know which policies he used, so we won’t learn anything about policies, if we can’t agree on what policies they used.

    • Jake says:

      What’s hypothetical about the claim that they avoided deflation? Check out this graph of Japanese inflation rates:

      http://www.tradingeconomics.com/japan/inflation-cpi

      You can set it to show different time frames. What it shows is that Japan was experiencing deflation for almost the entire period from when they really got hit by the Great Recession in 2009, until Abe was elected in late 2012 and pretty instantly started instituting policies that brought the rate way up.

    • ADifferentAnonymous says:

      Yeah, when I read ‘Keynesian dystopia’ I assumed it meant ‘dystopia of the kind Keynes predicted you would get if you listened to Milton Friedman’ and braced myself for disagreement on whether Japan was doing well or badly. I was kinda floored to see that the blogs disagreed on the outcome and the policy.

      Is Zero Hedge only predicting trouble rather than saying there’s trouble now? If so, it’s a bit cheeky to say that catastrophic results that haven’t happened yet ‘speak for themselves’ in the present tense.

      • Jake says:

        He is indeed only predicting trouble. Pretty much all economic indicators other than the debt have been doing quite well. Spending a ton of money on debt interest is not good of course, but trying to significantly pay down the debt would be REALLY disastrous. Already the biggest problem with Abenomics is being too concerned about the debt, which makes them take pro-cyclical actions like raising the consumption tax.

        • anon says:

          Why is there so much hate for Abe in the comments of the other article, then? Purely idiotic partisanship?

          • Jake says:

            There are two main reasons I’ve seen that people dislike Abe: First, he’s somewhat more nationalistic than Japanese leaders have been since WWII. He visited a shrine dedicated to Japanese soldiers who died in WWII, and is generally being more assertive and less pacifistic. Countries who have a bad history with Japan aren’t thrilled about this.

            The second reason people don’t like him, and I think this is more what we see in this article, is because he’s pursuing policies a lot of folks in the US and Europe are ideologically opposed to. Increasing spending, purposefully driving up inflation, etc. It’s mostly things that American liberals are doing or wish they were doing, and that American conservatives and pretty much the whole European policy elite are terrified of.

        • Paul Torek says:

          What Jake said. The second reason is what’s getting Zero Hedge in a dither, I think.

        • Douglas Knight says:

          he’s somewhat more nationalistic than Japanese leaders have been since WWII. He visited a shrine dedicated to Japanese soldiers who died in WWII

          I have no comment about the first sentence, but it is hardly supported by the second sentence. He is certainly not the unique PM to visit while in office. The longest-serving PM of the 21st century visited Yasukuni repeatedly. Lots of cabinet members visit. Between Koizumi and Abe 2, no PM visited, but that’s only six years (and six PMs). I think only one or two of them even tried to stop their cabinets from visiting. One of them didn’t visit as PM, but had earlier called for the Emperor to visit.

  8. ozymandias says:

    Of course, the fact that there are gender differences does not say anything about the *source* of the gender differences, and many radical feminist and pretty much all cultural feminist theories tend to be pretty into massive gender differences between men and women, so this is not exactly checkmate feminists even if it turns out to be true.

    That said, if it is true it does falsify some liberal feminist theories, particularly of the kind common on Tumblr.

    • zslastman says:

      Yep. Now Q the debates with everyone jumping to their favourite explanation of this data.

      It’s an interesting paper. They do several things differently from most studies that magnifies the effect. First of all, they dump the Big 5 in favor of a model one layer down. Which for instance splits extroversion into warmth/affiliation and dominance/venturesomeness. This gives bigger differences, because, for instance, these two factors cancel out to create about equal extroversion in men and women. You could argue this is cherry picking but actually the more complex model seems more intuitive and closer to how we think about people normally.

      They also do latent factor analysis, the idea being to control for the correlations between traits. This is kind of shaky I think. Observed variables are what we care about, and what’s exposed to evolution.

      And finally they use a multivariate distance instead of a single variable difference. Since the way we’ve carved up personality is arbitrary anyway, I think this is reasonable.

      The take away is that women are much more sensitive than men, and also somwhat warmer and more apprehensive, and less emotionaly stable and dominant. There are large (~70%) overlaps for everything but sensitivity. Give information on a persons personality, you can get their sex (or gender?) about 95% of the time, at least in this sample, which is about 10,000 people from the US in 1993. It all makes sense in light of stats on aggression and vocational choice. It also relieves me of a feeling of confusion I’ve had for some time. Apparently the sensitivity difference is robust in different cultures, but I’d be interested to see what traits remain different in say, Denmark.

      • Douglas Knight says:

        The interesting point in the paper is just pointing out that there are sub-factors of the big five that have bigger sex gaps than the big five. That’s a good observation, but it’s already published data and no math, so it’s a totally unacceptable paper.

        So the rest of the paper is devoted to another task: can we write a personality test to distinguish the sexes. This seems to me like a stupid question, but if people are really going around saying we can’t, I guess it needs to be done. And they limit themselves to building this test out of the accepted sub-factors. Most of the paper is devoted to this task and and to guessing how well their test would do. They didn’t actually give their test to anyone. Instead they use old data, but they don’t have all the raw data, just means and correlations, so I think that’s where the latent factor analysis comes in: just for predicting the results of their test.

        • Douglas Knight says:

          Nope, I was wrong. They use the latent variables to double their effect sizes on the individual traits, like Sensitivity/Tendermindedness, which moves from 1.3 std to 2.3. Color me skeptical.

        • zslastman says:

          I think it only seems like a worthwhile question if you know feminists who claim that people don’t differ substantially by gender, of which there are a few…

        • Douglas Knight says:

          Maybe it’s a good way of mocking such people, but it’s still a stupid question. Such people tend to either be dishonest or have reasonable positions to which this question is irrelevant. Indeed, the paper is aimed at Hyde, who has a quite reasonable position. Hyde’s paper is also a political move, but at least its point is interesting.

          Alrenous gives the example of the assertion that one can’t tell sex from conversation. If people actually went around saying that, this would be a relevant response. It’s probably true that propaganda encourages people to believe it, but that doesn’t mean it’s useful to rebut the particular point.

          I think Scott’s comment is good.

          In fact, the argument with Lewontin is also stupid. People argue that race can be distinguished by adding together lots of minuscule frequency differences as a prerequisite to arguing that interesting genetic differences between races exist. But while the two points are true, I don’t think that they have much to do with each other. Phylogeny is measured by genetic drift, which happens on one time scale, while selection can happen on many time scales, both faster and slower than drift.

      • Douglas Knight says:

        Re: sex difference in sensitivity/tendermindedness in Denmark,a paper that talks about tendermindedness and specific cultures, including Norway. But not both at once. It puts the tendermindedness difference at only .31 in America and .26 averaged over other places. On page 7 it lists differences for several traits across 26 cultures. Norway is middle of the European pack on all four traits.

        The .31 makes me I wonder if I’m reading the numbers wrong. Also, the sex differences on the big five look high to me.

  9. Protagoras says:

    For a long time the Princeton philosophy department didn’t interview job candidates (in defiance of what was the universal practice in philosophy departments hiring practices). I don’t know if they still have that policy (I’ve also been told it had more to do with some very important faculty members hating to do interviews and having enough status to get their way than the theoretical issues), but Princeton’s was also for a long time the uncontroversial best philosophy department in the world, and continues to be a contender for the spot, so the policy seems at the very least not to have hurt them any.

    • Gavin says:

      If you’re hiring someone to write brilliant papers and you can read the papers they’ve written, this seems like a good methodology. Most job applicants do not have track records anywhere near as clear or available as that–though perhaps this could be changed in the future. For tech jobs, github is starting to serve that purpose.

      • Anthony says:

        Except for entry-level jobs, every potential employee has a pretty long track record; the problem lies in accessing it. Corporate cowardice has made it harder to track down people’s records at previous jobs beyond whether they had them, but where it matters most, people are willing to talk off-the-record.

        The main purpose of the interview is to determine whether the person interviewed is sufficiently sane/well-socialized to be functional in the environment, presuming that their paper-trail indicates they’re competent.

      • Protagoras says:

        “The research may say that this is true in general, but it obviously doesn’t apply to my field for reasons.” Sure. To reply to the other defense of interviews in this thread, the anti-interview research has generally shown that interviews perform worse than assessment of available objective data, and that the combination of interviews with available objective data performs worse than the objective data alone. That interviews correlate with performance does not contradict this; it only shows that interviews are better than choosing randomly. The superiority of “structured interviews” is probably significant, but it’s not clear what it shows; it kind of seems like what makes an interview “structured” is that it is focused on eliciting specific objective information (which was probably available by other means), and so may be superior to unstructured interviews for the same reason relying on objective information is superior to interviews generally.

    • peterdjones says:

      Hmm but teamwork isn’t important in philosophy.

  10. Brian says:

    What’s the anti-tilt shift camera technique called? (I’m assuming it’s something other than “anti-tilt shift”)

    • Swimmy says:

      Maybe the technique used in the picture is focus stacking? It’s frequently used with images of insects.

  11. Stephanie says:

    Maybe I’m reading this wrong, but the thing about the cars:

    If all cars are programmed to go off the cliff (and why not just slightly to the side, why way overcompensate and go off a cliff?), then wouldn’t that be 100% of deaths on both sides because both cars would make that turn not just one? They were headed toward each other? The reason most don’t choose 100% death is because it is 100% death. The other way there is a chance for everyone to survive vs. everyone in it dying. Not more death, but less death. Both cars are facing each other; therefore both cars make the turn. Is there not a cliff for the other car to drive off of? Then should THAT be the car that overcompensates, as it is the car that doesn’t have a cliff to drive off of?

    Edit: I read the comment on your link and I agree with his comment. Having the cars try to decide which life is more important would actually cause more death than trying to protect itself. Like what if your car has five people in it, but the other one only has one? Is the car going to know that?

    Plus, I really can’t think of a situation that this would even apply. A car isn’t going to freak out like people do. It will be able to slow down AND more just enough over to not go off the road as well as not hit the other car head on. They might still hit each other, but it won’t be the same. A car isn’t going to panic.

    Car accidents would in theory be when the system is failing; therefore how it was programmed isn’t going to matter at that point, because the system is failing.

    • Anonymous says:

      Car accidents would in theory be when the system is failing; therefore how it was programmed isn’t going to matter at that point, because the system is failing.

      There are different levels of failure, and besides, the world is weird enough to throw obstacles at you quickly/unexpectedly enough that completely safe avoidance isn’t possible at your speed, but rarely* enough that said speed was still optimal.

      *It does seem kind of overly sensational (in a way analogous to Privileging the Hypothesis) to bring the trolley problem into early discussions of self-driving cars.

    • Jake says:

      Your objections are reasonable but they sort of miss the point. The situation with cars in the cliff is hypothetical – the point isn’t the realism of any details or what other options you would have in that scenario, it’s just a way of talking about the car system’s decision making. It doesn’t seem that crazy to me that there are different actions the car could take depending on whether it’s only priority was to keep it’s passengers safe or if it had some care for anybody else on the road.

      • Stephanie says:

        I think the problem with the “keep other people safe” programming would be (kind of like I added in my edit) that how is your car going to know there are people over there to protect to begin with? All it would know is that there is some obstacle: what if that happens to be a fallen tree?

        I can’t put my trust into a car that is going to randomly decide a tree is more important than my life. Or even that one other person’s life is more important than mine and my children when they are in the car with me.

        Plus, we have already be taught this in school. It is called defensive driving. You drive in ways to keep yourself safe; if you are constantly thinking of others you are more likely to get people hurt.

        • Jake says:

          You still seem to be somewhat avoiding the question by going after the details of the hypothetical rather than engaging with what it’s asking.

  12. CaptainBooshi says:

    A snide objection I’ve heard to the free market is that cable companies are capitalist and look how terrible they are. But one sign of how far cable is from an ideal market is the new poll finding over half of people would switch cable companies if they felt like they could.

    Well, to my understanding, this is the point of the snide objection. One of the failure modes for a free market is the most powerful actors muscling their way to an effective monopoly, which is the current state of the cable internet market. Personally, I wish I was even able to hate cable internet, but the only company close to where I live has straight up told us they’re not going to lay cable out to our area. I have satellite internet instead, 3 times as expensive, with a daily 450 MB cap, that loses all connection every time we have a storm, or even heavy clouds.

    Speaking of free markets failing, I saw an amusing article today. This paper apparently analysed a huge group of large companies, and found that the more you pay a CEO, the worse that company will perform. Not only do their companies perform substantially worse, they’re much worse at figuring out when to cash their stock options to make the most money, too, so it’s on both a personal and business level that getting paid more reduced their performance. The paper suggests that the reason this happens is overconfidence, that being paid so much more than everyone around them gives them the confidence to ignore information that contradicts what they know must be right.

    The sort of thing we’d never stop hearing about from feminists if it had happened to women: the impotence trials of 16th century France.

    I’d just like to note that while you always seem to be willing to extend the benefit of the doubt to people you almost completely disagree with, you also always seem eager to believe the worst about people you do agree with about many things. I personally believe you do this to signal to people you disagree with that you’re someone they can discuss things with (and you’re even willing to give them their worst assumptions about your side!), but I still find it very frustrating and annoying

    • James James says:

      “Well, to my understanding, this is the point of the snide objection. One of the failure modes for a free market is the most powerful actors muscling their way to an effective monopoly”

      Indeed, it is the point of the objection. Without government intervention (i.e. “left to the market”), consumers will not have choice. It’s not so much “powerful actors muscling their way” to a monopoly, but more that cables are a natural monopoly: big entry costs.

      (Big cable companies do indeed lobby, mostly to prevent pro-competition intervention. Lobbying in itself isn’t necessarily enough to gain any company a monopoly. Amazon lobbies about a variety of things, but they still have competitors.)

      The orthodox intervention is for the govt to either nationalise the infrastructure and allow a variety of companies to use it, or to force the owner to allow other companies to use it. In the UK, the govt owns the national grid, and allows lots of electricity companies to use it. British Telecom owns the telephone network, but must allow other ISPs to use it at a regulated price.

      Another example is cellphone numbers. The govt forced companies to set up a system for transferring a number between networks, so you can switch networks without being deterred by losing your number.

      “Free markets” is too vague. It can mean anarchy, or it can mean non-anarchy. I prefer “competitive markets”. If competitive markets are better than non-competitive markets, and the market is not becoming more competitive by itself, then I have no problem with the govt intervening to make it more competitive. A neoreactionary city-state would have no problem intervening. Non-competitive markets means someone is collecting economic rent, with a deadweight cost. Much better to eliminate the deadweight costs and have the city-state collect the rent through higher land prices.

    • Lavendar bubble tea says:

      “I’d just like to note that while you always seem to be willing to extend the benefit of the doubt to people you almost completely disagree with, you also always seem eager to believe the worst about people you do agree with about many things. I personally believe you do this to signal to people you disagree with that you’re someone they can discuss things with (and you’re even willing to give them their worst assumptions about your side!), but I still find it very frustrating and annoying”

      I’m kinda noticing/feeling that too :/ I caught myself thinking a few days ago “I think I’d get nicer/more open replies if I comment as a neo reactionary.” On one hand, I think it’s good to be able to have to defend your view point to people with similar viewpoints as way to stop group think/intellectual laziness. But, at the same time, I do feel like Scott may be assuming the worst of people/thought trends which are pretty similar/not totally alien to his moral system and that’s not very good for maximum intellectual honesty/engagement either.

      • anon says:

        I’ve been getting that vibe lately as well, last month or two. Not directed at me personally, but just in general.

        • nydwracu says:

          I would expect movements that have selection mechanisms for intelligence/patience/etc. [and possibly some degree of emotional stability? or something in that environment anyway] built into the foundational texts and that go out of their way to distinguish themselves from [and mock] movements considered similar to theirs with already-existing bad reputations to have better reputations than movements that try to build as broad a base as possible and don’t do much of anything to distinguish themselves from the enormous piles of blithering idiots who are already using their same exact banner, independent of ideology.

          Maybe if the feminists who aren’t blithering idiots broke with ‘feminism’, made their own label, built selection mechanisms into it, and started mocking the idiots, they’d have a better reputation?

        • Multiheaded says:

          In many feminists I’ve read, intelligence seems anti-correlated with patience and especially emotional stability. Same as with a lot of politics among intellectuals, at least until you get to the extreme extremes, where coherent theory often becomes lifestyle posturing anyway. (You’d probably agree that, say, Moldbug is a lot smarter than you – or most everyone – and an insufferable antisocial jerk prone to delusional flights of fancy.)

          CORRECTION: this only really seems useful for online stuff.

    • Scott Alexander says:

      That’s a good point, but then I think we should distinguish between two statements:

      A) Textbook ideal free markets work poorly
      B) Textbook ideal free markets might work well, but can be hard to get

      I would agree with B much more strongly than A. I get the feeling that people who bring up this example are either asserting A or else conflating the two possibilities, but maybe I’m wrong.

      I am pointing out that feminists like taking examples of historical horrors perpetrated against women (as women) out of context and saying they prove oppression, whereas here is an example of a historical horror being perpetrated against men (as men), and reality is that historical times were pretty horrible and alien to our morality for people of any gender.

      I am kind of past annoyance at how whenever I disagree with feminists I am accused of being mean or uncharitable to feminists and not giving them the benefit of the doubt. Maybe if you usually hang out in places where reactionaries are treated with total scorn and feminists are given a free pass, then me being neutral to both groups looks like privileging the reactionaries by figure-ground illusion. In favor of this hypothesis, actual reactionaries think I am consistently unfair to them. I usually figure when people on both sides accuse me of bias an equal amount, I’m probably in about the right place.

      • Jake says:

        I agree about the two statements, but an important corollary is that firms will try really really hard to transform their market from a perfectly competitive one to a broken, monopolistic one. The type of industry just determines how easy this is for them. The industries where markets work better are ones where it’s harder to accomplish this goal and become a monopoly, or at least a participant in an oligopoly.

        To make markets work better, the government then needs to intervene to prevent monopolies – by breaking up trusts, breaking up monopolies, by reducing barriers to entry, etc. Where these don’t make sense – mainly because an industry is a natural monopoly, the government needs to step in and either heavily regulate the monopoly, or nationalize it altogether.

        What this all comes down to is that markets are an incredibly useful tool, but we shouldn’t get too enamored with them and fall victim to the fantasy that ‘free’ markets in the sense of being free from government influence, will always magically produce the best outcome.

      • Anthony says:

        whenever I disagree with feminists I am accused of being mean or uncharitable to feminists and not giving them the benefit of the doubt.

        There was a comment a while ago about how feminists take a word, give it a reasonable meaning, and then proceed to use it in a completely different way. (Privilege meaning “shut up, white boy”, for example.) “Feminism” itself is one of those words. Modern feminism is a bunch of evil, small-minded, power-crazed, anti-male hyper-conformists, who will all fall back to “Feminism is the radical notion that women are human beings” whenever they’re called on their lies or their Maoist tactics.

        So, no, modern feminists don’t deserve the benefit of the doubt, and I’m glad Scott treats them badly.

        • Jake says:

          Or to put it another way, there are a lot of toxic, awful people out there, some of those people are feminists, and now that we have the internet those who want to listen to every toxic awful thing those people have to say are capable of doing so.

          Perhaps I like feminism better than you do because my impression of it is based on people I know who call themselves feminists, rather than what a small minority of them are writing on the internet.

        • ozymandias says:

          Seriously?

          Look. I googled “feminism” and these are the results I get:

          Feminist Majority Foundation. Sample issues include rape kit testing, Obamacare, an end to sex-segregated schools, and preventing housing discrimination against domestic violence victims. Evil power-crazed anti-male rants seem to be not in evidence.

          Feminist.com. Is sufficiently far from hating men that its most recent updated column discusses the boy crisis in education, giving men access to paternity leave, and the fact that men have shorter lifespans than women.

          Next up: the Guardian’s Feminism section. Does include articles about mansplaining and anti-MRA sentiment, which you could perhaps call strawmanny or ill-advised, but I am not entirely sure it reaches the “evil, small-minded, power-crazed, anti-male hyper-conformists” standard. Also discusses things like sexism against female celebrities and politicians, the importance of education for women globally, and the important role that men and fathers have to play in feminism.

          But hey. Maybe feminist.org is not run by real feminists? Someone should tell them.

        • Multiheaded says:

          Or maybe, just maybe, you’re some kinda rape apologist or something, I dunno. (I’m frankly too disgusted to read your fucking blog… sorry not sorry, dude. It appears to be some neutral curious stuff and some links to horrible woman-hating shit, I don’t know what to think.)

        • Ialdabaoth says:

          Damnit, Multiheaded, this game has *serious stakes*. Can’t you at least learn to *pretend* to play by the rules?

        • Multiheaded says:

          Ew. I was raised on the legacy of a higher civilization, yesterday’s tomorrow. I’m viscerally disgusted by savages, like racists and sexists.

        • Ialdabaoth says:

          Who have *weapons* that they can use to *remove you* from the playing field. Being viscerally disgusted is great motivation, but if it short-circuits your capacity to perform strategically it is *worse than useless* to you.

          Also, I think you are doing these people a great disservice. I’m a hair’s breadth from being one of them. The ONLY difference between them and me was some quirk of fate that caused me to see the fnords on both sides.

          Do you understand how *easy* it is to believe these sorts of things, and how *hard* it is to train yourself not to, without any kind of cultural backing?

          Even the most cursory look at history will show you that your beliefs are *not natural*. I happen to find them superior, so their unnaturalness and difficulty make them all the more valuable. But being able to conceive of a weak Other as a person is a mark of *privilege*. Stop treating it like a duty unless you’re willing to take the effort to prop up that privilege, first.

          • ACS says:

            Phrased more neutrally, they have words with which they can be somewhat unkind to you. If you are the sort of person who takes his principles seriously, the particular words with which they are somewhat unkind to you with will ask that you waste time reexamining principles that you actually do not have, but holding rhetorical violence and rhetoric-demanding-violence in the same frame does not seem justified.

        • Multiheaded says:

          Yes. You have a point. I need to beam out and delegate this to an away team.

          Do you understand how *easy* it is to believe these sorts of things, and how *hard* it is to train yourself not to, without any kind of cultural backing?

          god don’t talk about this i’m trying to forget

          omg ew i almost was one too

          seriously

          also, i honestly believe that calvinist!reactionaries and nerd!reactionaries are worlds apart, and the former are less threatening, as they seem to care/reflect less about inflicting misery efficiently or at a great distance. horrible to live around, of course, but they wouldn’t actively invent something like 19th c. race theory.

        • Anon says:

          There are a bunch of vastly different communities calling themselves feminists, so anything that starts with “feminists think/say/do…” will be often wrong. For example, a major feminist issue where I live has been better conditions/rights/legislation/etc for prostitutes. It would be hilarious to see the tumblr archetype try and twist around itself to support something like that. There are a lot of truces between communities (ie: dont talk about the prositution thing or the sex-positive and sex-negative groups will collapse the whole movement in civil war) and that kind of gives the appearance of unity, but in reality the values and ideas of groups who identify as feminist seem to contradict each other greatly. This serves to make the movement appear much worse from the outside. It’s basically a bigger, less focused OWS only the media hasn’t got a reason to tear into them. (I suppose the people who OWS targets can co-opt feminism for their own ends, so there’s that)

        • nydwracu says:

          calvinist!reactionaries and nerd!reactionaries

          ?

          (Nick Land and Gromar are the only reactionaries who will admit to Protestantism. I’m not sure if they’re Calvinists. But then there’s the Carlyle thing.)

        • Multiheaded says:

          I mean the subset of the Christian right who practice their extremely anti-leftist worldview in real life like Ialdabaoth talks about, as opposed to eccentric tech nerd bloggers on the Internet. And yes, I am saying that the tech nerds are more creepy than the weird inbred hick cultists.

      • Jake says:

        I think the ‘getting flack from both sides’ points towards you being not particularly biased one way or the other, but doesn’t really give any information on whether you, or one side or the other, is actually correct.

      • ACS says:

        Just a question: does the fact that a particular philosophy has a body count (and another does not) matter to you? Should it?

        • Scott Alexander says:

          Somewhat, but I am also extremely suspicious of people being willing to round off their enemies to philosophies with body counts, while being very defensive of their own non-identity with their nearest murderous cousins.

          Neoreactionaries would point out that progressivism (including communism) has the highest body count of any philosophy in history, whereas neoreactionaries (ie followers of Mencius Moldbug) have a body count of zero.

          Feminists would point out that illiberal people (including fascists) have a very high body count, whereas feminists have a body count in the low single digits (a couple of people committed suicide after being harassed by feminists too much, but nothing on a large scale).

          When people say they don’t want to kill anyone, I tend to believe them, rather than rounding them off to the nearest equivalent group that did kill people.

        • ACS says:

          It seems to me that that extends the principle of charity too far?

          When people suggest, for instance, that there will be violent rebellion, and it will need to be violently suppressed, and that this will be necessary in the long run, and that what they are arguing for is the best case scenario, I think that — at least at a meta level — it is appropriate to judge the likelihood of that actually being true independently of how politely they argue for the necessity of mass murder.

          Rational disagreement is fine, sure. But I am not sure that it’s meta-rational to have polite, rational disagreements with people who would use the state to murder me if it came to that: as important as it is to divide the world into “capable of making a reasonable, good-faith argument for their position” and “incapable of same,” it seems to be equally important (if not more important!) to divide the world into “willing to tolerate my murder or immiseration in order to ensure adoption of their position” and “willing to prioritize lack-of-murder over the specifics of their cultural program.”

        • Ialdabaoth says:

          Keep in mind, though, that this goes both ways. If I can wield sufficient power that the only way to stop me from murdering / immiserating others is to murder me, I can rightly claim that others are unwilling to prioritize lack-of-my-murder in favor of their preferred cultural implementation (which just so happens to not appreciate the misery I want to inflict on others).

          Politics, at its barest form, is often a matter of convincing people that *my* victims don’t count, and *your* victims make you a monster.

        • ACS says:

          That does, in fact, go both ways.

          I would not like anyone whatsoever to have the power to murder you for believing whatever awful things you happen to believe. I am willing to do whatever is necessary to keep that from being the case.

          In terms of immiseration, if there are things that are making you sad which are none of your business, or are no one in particular’s fault, then you will unfortunately have to live with them, because fixing your problem (with immigrants, with women not liking you) is (a) no one in particular’s problem, and (b) demands a solution which only particular people can offer. If you have problems which are caused by broad forces and can be solved by same, I’m perfectly willing to discuss that.

        • Jake says:

          But ACS, why only judge a philosophy by it’s intent towards it’s political enemies? What about a philosophy that advocates wars of conquest, while still being perfectly respectful of it’s political opponents back home. Or how about a philosophy that is against government subsidies for food or healthcare or housing, condemning it’s fellow citizens as surely as a clash in the streets.

          It takes a lot to justify violence, but surely you can imagine regimes sufficiently terrible that violent revolutionary action was called for?

          • ACS says:

            Well, not just by its intent toward its political enemies. But I do think that political enemies should judge political philosophies based on intent toward political enemies. I do not believe that it is possible to politely recommend my death.

        • Ialdabaoth says:

          I do not believe that it is possible to politely recommend my death.

          Whereas I am intimately familiar with how easy it is to politely recommend my death.

          Or rather, how easy it is to recommend my death in a manner that is accepted as polite by civilized company.

          You have to understand that “principles” are a veneer of civility over a fundamentally vicious game, and are therefore themselves nothing more than weapons to wield in that game.

        • ozymandias says:

          People politely wish my friends were dead all the time. Sometimes they write great big books about it, generally filed under “bioethics.” lbr here, wishing that people would die is only impolite if people recognize that their lives have value in the first place.

          • ACS says:

            What I mean is not that it’s not possible, but rather that it is not polite. I am agnostic as to the need for some hard boundary to tolerance, but “put down your gun before we start talking,” even if that gun is wholly abstract, seems to be a good general rule for discourse.

        • ACS says:

          You have to understand that “principles” are a veneer of civility over a fundamentally vicious game, and are therefore themselves nothing more than weapons to wield in that game.

          If I accept your conclusion and agree that principles are mere weapons, I do not understand on what basis you can object to the fact that tit-for-tat eliminationism (or, better, insisting that you renounce political violence or suffer same) is clearly the optimal strategy for dealing with people that hold your beliefs.

        • Ialdabaoth says:

          If I accept your conclusion and agree that principles are mere weapons, I do not understand on what basis you can object to the fact that tit-for-tat eliminationism (or, better, insisting that you renounce political violence or suffer same) is clearly the optimal strategy for dealing with people that hold your beliefs.

          Because tit-for-tat eliminationism is only an optimal strategy if you insist on playing by the same level of rules as your opponent.

          Ideally, you convince your opponent to embrace non-violence so that he is incapable of resisting when you enslave his children, torture his women in front of him, and then eliminate him with a laugh.

          And insisting that your opponent renounce political violence is seen as a crass grab for a monopoly on power by everyone who’s playing the game at the same level. The real trick is to claim that violence is inevitable, to signal honesty (the REAL trick being to signal honesty while playing at the next higher level (the REAL trick being to signal that you believe your opponent’s honesty while preparing to trap them at the next higher level (the REAL trick being to… you see where this is going) ) )

          • ACS says:

            You have apparently bitten too many bullets in an attempt to seem like a steely-eyed realist, and in doing so you’ve forgotten how games work .

            Ideology entirely aside, “cooperate within the context of a meta-agreement to cooperate; quickly punish defectors; quickly repair the meta-agreement when breached; forgive defectors and admit them to the consensus when the cost of readmission is lower than the cost of disorder” is a perfectly reasonable strategy. “Ruthlessly punish the opposition when in a position to do so” is not, as any number of states which do so demonstrate. Your thesis is that the war of all against all secretly persists even in the presence of a consensus otherwise, and your protestations of steely-eyed realism aside, this is simply not true.

            Would you mind clarifying the threat you’re under? (If you live in Mosul, I apologize.)

        • Ialdabaoth says:

          I grew up in a hyper-Calvinist Christian Fundamentalist sect with strong Prosperity Gospel overtones, obsessed with rooting out secret devil-worshippers in its midst. It’s colored most of my view of the world since.

          • ACS says:

            Re: Calvinism. Well, that seems sensible. People who hold unusual beliefs about which strategies for dealing with disagreement prevail tend, in my experience, to come from intellectual microclimates in which their beliefs actually do prevail.

      • CaptainBooshi says:

        “I would agree with B much more strongly than A. I get the feeling that people who bring up this example are either asserting A or else conflating the two possibilities, but maybe I’m wrong.”

        I also agree with B far more than A. I would say this example doesn’t even support A, since the cable industry is SO not an ideal textbook free market. In general, when I see this example being brought up ‘in the wild,’ it’s to counter the idea that free markets just work, and you shouldn’t ever mess with them. I do understand where you’re coming from, though, because that’s strictly anecdotal for me, and we clearly have very different experiences on the internet.

        “I am kind of past annoyance at how whenever I disagree with feminists I am accused of being mean or uncharitable to feminists and not giving them the benefit of the doubt.” It’s not that you disagree, or even are uncharitable. It’s that you consistantly portray them in the worst possible light, to the point of strawmanning them. I still believe it’s because you’re not interested in discussions with feminists, so it’s mainly signalling to the people you do want to have conversations with.

        • Anonymous says:

          This blog actually goes out its way to portray feminists in the best possible light. (It does so for groups like reactionaries too, but still, completely charitable and not highly critical at all). I think you lack context; you might not be considering this blog compared to say LW. 99% of feminists are not atheists. They can basically be described as crazy conspiracy theorists who deny human rights in favor of their own made up beliefs. You’re possibly not familiar with a context where a primary, overwhelming tone is that people are outright ridiculed and regarded as horrible for pushing religious beliefs, and this describes all mainstream political groups including progressives and feminists.

          You seem to be thinking that even in specific terms of “a few hundred people on the Internet” the feminists and reactionaries are being treated differently. But the reference and context for “a few hundred people on the Internet” for the blog host here is not the broader feminist/reactionary Internet (a vast subset of real life anyway) but actually a community where feminists are essentially no different from other anti-science bigots.

        • ozymandias says:

          Once again I point out the Google results for feminism and wonder which of us is having a selection bias problem.

          Also a lot of rude feminists are atheists, like, have you ever read FTBlogs

        • Multiheaded says:

          He means “not atheists” as in “believing in Evil Cathedral Social Science” and “not respecting Le STEM and its direct applicability to every single human problem”.

      • Douglas Knight says:

        First of all balancing complaints from both sides sets up extremely bad incentives.

        Scott, you treat SJW and reactionaries very differently. I’m not going to say that you treat one better or worse, or that that you treat them inappropriately, but if you don’t see this difference, I think you’re making a big mistake. You write posts about the inter-personal dynamics of SJW, while you write posts about the ideas of reactionaries. Mainly you talk about reactionaries being wrong, while you talk about SJW acting in bad faith.

        • ACS says:

          I think that’s the core of the problem: that on a two-axis model of “overall politeness” and “propensity to advocate for mass murder,” neoreactionaries are in the upper-right-hand corner and feminists are in the lower-left.

          I would be irritated by a world in which radical feminists had gotten all of their policy preferences enacted, but I would be dead in a world where neoreactionaries got theirs.

        • ozymandias says:

          Your model seems to predict the exact category of complaints we’re seeing: feminists going “Scott, you agree with us about everything, why are you being knee-jerk anti-us” and reactionaries going “oi, he misinterpreted our statistics, we are actually right.”

        • ADifferentAnonymous says:

          On the other hand, Scott believes that NR has a chronic wrongness problem and SJ has a chronic bad-faith problem, so what’s he supposed to write?

        • Ialdabaoth says:

          Heh. I think that’s actually what specifically helps me re-cement my pseudo-Progressive values:

          the Social Justice Warriors show me “my side” acting at my worst, and I find I can still justify the core principles while being appalled at their tactics. (“You’re not wrong, Walter, you’re just an asshole.”)

          The Neo-Reactionaries show me my enemy acting at their best, and I find I still can’t accept or even conscion their core arguments. (“I beseech you, in the bowels of Christ, to consider that you may be wrong.”)

          I feel like the best answer to “Nice is not Good” is “can’t you try just a little harder to be both?” – Neo-reactionaries seem to prefer Niceness at all cost, and don’t particularly seem to care deep-down about any kind of global Goodness. Social Justice Warriors have got it into their head that Nice is not Good, and thus allow themselves to be as Not-Nice as possible, claiming that this lets them better serve the forces of Good (insert dubious look here).

        • Multiheaded says:

          Makes sense, although as a Gloomy Cultural Marxist I also have to deal with the suspicion that SJ strategy is selected for ultimately making the machine run more smoothly by slightly alleviating some high-exposure things.

        • Ialdabaoth says:

          I will tell you right now, it isn’t. SJ strategy is selected for feeling like you’re striking out against a powerful enemy, when you’re really just striking out against someone more oppressed than yourself.

          The problem is that most SJ strategy is produced in America, and the right figured out a LONG time ago how to manipulate the left into eating itself and alienating its potential allies. (Karl Rove et.al. are the latest incarnation of this tactic)

          The best metaphor – and indeed the best actual implemetation – of this tactic was the 9/11 terrorist attacks. All Al Queda had to do was fly a few planes into a few buildings, ONCE. Guantanamo Bay? National illegal surveillance? Abu Graib? Extraordinary Rendition? Bullshit TSA security theatre? Crackdown on civil rights? Hypermilitarized police state? We did all that to *ourselves*.

          All someone has to do to destroy the Good is present a boogey-man mockery of evil, trot it out once in awhile in bad light, and then convince the Good that it exists everywhere, hidden in its midst. Then sit back and watch Good destroy itself through witch-hunts and hypervigilance.

          • How do you see SJ as a result of right-wing manipulation? I can believe the right might have wanted such an outcome, but I can’t imagine how they could have achieved it.

        • Multiheaded says:

          I don’t think this is quite what has been going on… (Although the Jacobinghazi thing did indeed seem to go down exactly like that.)

        • Ialdabaoth says:

          How do you see SJ as a result of right-wing manipulation?

          You start with, for example, audacious “political correctness gone amok!” claims. You ramp those up until your enemy is convinced (probably rightly) that the general public can’t tell the difference between a legitimate claim of oppression and moonbat-level craziness. You throw out terms and concepts like “reverse discrimination” and “men’s rights” and what-not, basically perverting your enemy’s tactics and making it harder and harder to tell the difference between your enemy’s legitimate tactics and your perversions of them. Then, once the sociopolitical climate is so thoroughly muddied that it’s impossible even for your enemy to distinguish between legitimate acts and illegitimate ones, you wait and let human nature take its course. In due time, your enemy will become the charicature of them that you’ve painted, because you’ve poisoned their error-correcting algorithm.

        • Douglas Knight says:

          ADifferentAnonymous, yes, he should write based on his beliefs, not based on balancing complaints.

          Ozymandias, my model? What do you mean by “knee-jerk”? Usually it means that people reject an example from a generalization without looking at the example. That doesn’t cover this post or Captain Booshi’s complaint, but I might include them, since Scott’s bringing up feminists seems pretty arbitrary. But people usually accept Scott’s concrete example and reject his generalization. Maybe promoting generalizations promotes knee-jerk reactions in the future, but it isn’t an example of Scott having a knee-jerk reaction.

        • suntzuanime says:

          I think that’s the core of the problem: that on a two-axis model of “overall politeness” and “propensity to advocate for mass murder,” neoreactionaries are in the upper-right-hand corner and feminists are in the lower-left.

          I would be irritated by a world in which radical feminists had gotten all of their policy preferences enacted, but I would be dead in a world where neoreactionaries got theirs.

          Eh, call me privileged straight cis white male scum, but a lot more feminists have advocated for my death than neoreactionaries.

        • James says:

          Do any neoreactionaries actually advocate, even half-jokingly, for outright killing or encouraging the death of any groups people? I’d believe it, sure, but I’ve also never encountered it before.

          On the other hand, I’ve seen plenty of “kill all men” “we’re just blowing off steam relax” from the more toxic SJWs I know.

        • Ialdabaoth says:

          Do any neoreactionaries actually advocate, even half-jokingly, for outright killing or encouraging the death of any groups people?

          Such rhetoric has been used in comments on this very blog, as a response to the “black criminal thug” problem. Example:

          One day he runs into George Zimmerman and under some set of circumstances, winds up bashing his head into a concrete pavement – not nice behavior, no? George Zimmerman does the good liberal thing (if you believe that liberalism is about “be nice”) and shoots and kills him – all good, right? One fewer “not nice” person out there to ruin the commons of being able to own property and not be violently assaulted?

        • suntzuanime says:

          My guess is it’s more a blind patternmatching “right-wing -> hate -> HOLOCAUST OH NO” than anything related to actual neoreactionary views. But if ACS has a specific reason to think they would be dead in an NRx utopia, I’m interested to hear it.

          EDIT: Do you have a link, Ialdabaoth? Of course not everyone that posts in the comments of this blog is a neoreactionary.

          EDIT2: If ACS goes around smashing people’s heads into pavement, I suppose that qualifies as advocating for their death, but they really shouldn’t do that.

          • ACS says:

            Rounding up the examples of neoreactionaries politely howling for blood of their political enemies is tiresome. You can feel free to google for it yourself, using whichever terms you feel most appropriate. If you can imagine a polite way to call for any random authoritarian massacre, you can be assured that it’s been done.

            Here, for instance, Moldbug condemns the Breivik massacre as immoral because premature, presuming that mocking anyone with any moral intuitions whatsoever is an adequate substitute for having them. If you’d like me to dig up apologia for slavery or Jim Crow Dollfuss or Qin Shi Huang or stoning homosexuals*, I’d be glad to pull those up for you as well. It’s not like they’re hard to find.

        • Oligopsony says:

          As a negative utilitarian I like to think I’m sympathetic to everyone’s death.

        • Oligopsony says:

          Showing concern over superweapon development is conspicuously status-seeking (hence status-lowering, for good reasons) behavior when superweapon deployment is implausible. For instance being concerned that feminist memes will result in inoccuous nerds being labelled creeps is at least relatively more plausible than that they will lead to the genocide of cis people.

        • suntzuanime says:

          It’s impolite to call for genocide, even if you are unlikely to carry out that genocide anytime soon. As ACS says, it is not possible to politely recommend your interlocutor’s death. People are offended at the rudeness moreso than they are fearing for their lives.

          (Yes yes, I am aware that attempting to apply politeness norms to anyone who is not a straight cis white man is tone policing and evil.)

        • Ialdabaoth says:

          I’ve noticed that when I go Dark, one of my first rhetorical tactics is to call for the genocide of my own kind, for the good of everyone else – and then demand that others acknowledge that it’s what they really want and/or really would be in their best interest to want it.

        • Anonymous says:

          ACS, you keep including asterisks and words like “here” that don’t refer to anything. Are you leaving out links? Try just pasting in the URL.

        • jaimeastorga2000 says:

          @Ialdabaoth: I don’t think I have ever seen neoreactionaries advocate genocide. The “closest” I have seen is a political stance that career criminals like Trayvon Martin should simply be hanged as justice for their past crimes, to prevent them from committing future ones, and to serve as a warning to other would-be criminals.

          From Foseti:

          A while back, I linked to a story about a guy in my neighborhood who’s been arrested over 60 times for breaking into cars. A couple hundred years ago, this guy would have been killed for this sort of vandalism after he got caught the first time. Now, we feed him and shelter him for a while and then we let him back out to do this again. Pinker defines the new practice as a decline in violence – we don’t kill the guy anymore! Someone from a couple hundred years ago would be appalled that we let the guy continue destroying other peoples’ property without consequence. In the mind of those long dead, “violence” has in fact increased. Instead of a decline in violence, this practice seems to me like a decline in justice – nothing more or less.

        • Zathille says:

          @ACS: I’d very much like to see quotes and links substantiating your claims. Perhaps it’s possible that what you interpreted as advocacy for violence isn’t so when considered in context? hard to say without sources.

        • Ialdabaoth says:

          The “closest” I have seen is a political stance that career criminals like Trayvon Martin should simply be hanged as justice for their past crimes, to prevent them from committing future ones, and to serve as a warning to other would-be criminals.

          Which, when combined with phrases like “the criminal class” and “breeding a race of thug lackeys”, is not exactly subtle.

        • Multiheaded says:

          A couple hundred years ago, this guy would have been killed for this sort of vandalism after he got caught the first time.

          …No?

        • dublin says:

          Blacks commit a disproportionate amount of crime -> punishing crime means disproportionately punishing blacks -> punishing crime is racist.

          Gotta love disparate impact reasoning.

        • Multiheaded says:

          Punishing crime is racist -> punishing crime means disproportionately punishing blacks -> [alienation, social disruption, destruction of communities, hopelessness, hatred, resentment towards state authority] -> blacks commit a disproportionate amount of crime

          Gotta love systemic oppression.

          (Before you get started with scientific racism: cf. the crime stats for Botswana or Ghana.)

        • Ialdabaoth says:

          I am coming around again to the idea that actual conversation on these matters is impossible, and that political power only EVER comes from the barrel of a gun, or its temporolocal equivalent.

        • Anonymous says:

          Zathille, here are the two posts by Moldbug on Breivik: 1 2 (bitly because the blog thinks I’m a spammer)

          I don’t think ACS’s characterization is fair.

        • Scott Alexander says:

          Yeah, I’m pretty sure it’s not meant to be fair.

          If it helps, assume the Reactionaries want something sort of like 2014-China but more centralized. It’s not quite right, but if otherwise you’re just going to assume they want to kill everybody, 2014-China will be closer to the reality than whatever else your feverish imagination will dream up.

      • Hainish says:

        “I am pointing out that feminists like taking examples of historical horrors perpetrated against women (as women) out of context and saying they prove oppression”

        I get the impression that feminists do this to counter the notion that women were never oppressed, that they simply didn’t flock to institutions of higher learning because they had better things to do, that they had loads of power but were just too lazy to use it, etc., etc.

  13. CalmCanary says:

    “10. Trust those who trust you and distrust those who are suspicious of others.

    20. I never trust a man who doesn’t have enemies.”

    It would appear that Nassim Taleb doesn’t think we should take his advice.

  14. - says:

    I expected better from Satoshi Nakamoto.

    Expected better based on what? The Bitcoin design which clearly has this property? Has something else been published under that name that I’m not aware of?

    • Anonymous says:

      I read that as sarcasm.

    • Scott Alexander says:

      What I mean is that many people whom I respect got extremely excited about the Bitcoin architecture saying that this had unexpectedly solved all the problems inherent in cryptocurrency and was totally un-exploitable. Apparently that wasn’t quite true.

      • tomas says:

        Whom are you referring to here, I’m really curious? I’m not aware of a single person remotely competent at mathematics who ever had a favorable opinion about Bitcoin.

        For a specific instance, someone everyone here might have read, I’d never classify Wei Dai’s remarks as anything other than highly critical and skeptical of the actual implementation of Bitcoin the whole time.

        I’m also unaware that Eliezer ever publicly said much about Bitcoin. If he did, and it was strongly favorable (because some popular statements are outright misconceptions, like the idea that Bitcoin invented a solution to the Byzantine Generals problem, that I doubt Eliezer communicated) that really affects my Bayesian judgment of Eliezer’s understanding of math more than any conclusions about Bitcoin.

        Even from people in widely disparate fields, essentially just random PhDs in physics or CS etc. blogging on the Internet, no one who again could be independently described as competent has been positive about Bitcoin.

      • ADifferentAnonymous says:

        It was in fact known from the very beginning that an entity such as the NSA could pull a ‘51% attack’ with lots of supercomputers, and while I lack the knowledge to comment on the likelihood of this happening, the gHash affair does not yield any evidence that it’s more likely. gHash is a pool that grew by attracting more and more miners, not by buying lots of supercomputers.

        The possibility of the NSA covertly setting up a malicious mining pool, on the other hand, may warrant consideration.

  15. Sniffnoy says:

    Regarding the Bitcoin thing:
    Isn’t this the kind of thing Peercoin was created to solve?

    Regarding interviews:
    I mean, you might think they’d be in favor of replacing interviews with tests, but depends on the type of test, because depending on the test they may call you a racist! (And by their own use of that word I guess they’d be right, but the claim that their use of the word actually agrees with the standard one is not something I’d agree with them on.)

    (See also: Tokenadult’s standard comment (here’s an instance) that he posts on any discussion about hiring on Hacker News.)

    I mean, the SJers don’t seem to be the people saying “Hey let’s only judge people on relevant characteristics, and implement blinding and debiasing measures to ensure that.” They seem to have an altogether different philosophy; I think they might well dispute the whole idea of “judging people on relevant characteristics”. And they seem to often insist that blinding and debiasing measures are essentially futile. I think if you could get them to agree with my “Hey let’s…” statement above, they’d be far more in agreement with you than they actually are. But it doesn’t seem to be something they’ve failed to consider; they’ve considered it and rejected it, and are in actual disagreement with you there.

    (I’d add here a paragraph about college admissions and Jews and etc. but you can probably fill it in. Well OK largely I’m leaving it out because the resulting comment was starting to sound uncomfortably nrx-y. 😛 And I don’t even have a source for the thing about admissions tests and Jews even being true, it just seems to be something that “everybody knows”…)

    • dublin says:

      On the one hand, SJs are naturally opposed to judging people on the content of their character instead of the color of their skin, since the legacy of slavery etc. has left blacks with deficiencies of character through no fault of their own. Raceblind procedures are racist, since reality is not raceblind.

      On the other hand, there are studies out there suggesting that given equal qualifications companies are more likely to hire whites than blacks, so raceblind procedures should improve the status of blacks relative to whites. SJs often worry more about what the actual effects of a policy will be on that status ratio then whatever bullshit philosophical justification they’re using to push it. So if they feel like they can’t get race quotas, they might settle for raceblindness as a consolation prize.

      • Jake says:

        “On the other hand, there are studies out there suggesting that given equal qualifications companies are more likely to hire whites than blacks, so raceblind procedures should improve the status of blacks relative to whites. ”

        This is the key point. Imagine that we live in a society where racial prejudice still exists and could effect hiring decisions (crazy I know). In that world, wouldn’t a good solution be to say “Well, blacks with identical qualifications are getting hired at a 20% lower rate than whites, so let’s just lower the bar by 20% for blacks to even it out.

        Of course neither the effects of racism nor of affirmative action can so easily be quantified, but that is the central justification. If it was possible to eliminate the myriad subtle effects of racism in all areas of life then affirmative action would be unnecessary, but since that isn’t possible we have to give advantages in the areas we can control to counteract the disadvantages that we can’t control.

    • James James says:

      “I’m leaving it out because the resulting comment was starting to sound uncomfortably nrx-y.”

      Perhaps you might be (somewhat) a neoreactionary but haven’t realised it yet?

      • Sniffnoy says:

        No, considering my positions more generally that’s nowhere close to true. Thing is they’re the main ones who go around saying the Things You Can’t Say, so if you want to say one of these things, you end up sounding kind like them, regardless of your other disagreements…

        Edit: Like, I’m probably further from neoreaction than Scott is, so…

    • Charlie says:

      On the one hand, it is a true fact, that I heard from a friend of a friend (a.k.a. The Atlantic), that ivy league admissions were changed because too many undesirables were getting into the old boy”s club.

      On the other hand, new, improved (now with 20% less nouveau riche!) college admissions do a surprisingly good job at predicting/detecting success. So I think this might actually be a point for the effectiveness of interviews, with screening out t3h j00z a mere justification.

      • Jack says:

        I don’t really buy that interviews are a relevant part of the admissions process, at least when it comes to finding and accepting people who will be successful. The most important elements seem to be (in no particular order):
        SATs, measuring native intelligence
        GPA, measuring conscientiousness, intelligence, and ability to work within an organization
        Extra-curricular stuff, which depending on the type and number can indicate motivation, ambition, and social skills

        Any school assigning much weight to the interview and application essay either has a homogenous applicant pool on the above measures, is dumb, or is trying to keep out t3h j00z.

        • Nornagest says:

          I wish someone had described it that way to me when I was going into high school. I’m not sure how much of it would have stuck without an adult work ethic, but it’s gotta beat the naive assumption that I ended up going with (“it’s all bullshit”).

        • Jadagul says:

          At the very elite schools, they basically have a line which is “good enough,” and basically corresponds to “everything above this line is statistically indistinguishable.” But they can’t admit all those people, so they have to look at something else.

        • Douglas Knight says:

          Jadagul, look at this graph of probability of admission to elite schools as a function of SAT percentile. Yes, 88th percentile is “good enough,” but it is very easily statistically distinguishable from 99th. It’s pretty clear from the graph that the schools think so, too.

        • Jack says:

          Nornagest: I feel exactly the same way.

          Jadagul: Good point. I think that falls under ‘homogenous applicant pool’. It’s also a possible justification for quota systems. If there are more qualified applicants than positions and they have to use crappy diagnostics like interviews, unconscious prejudice is likely to drive the decisions.

          On the other hand, why the hell should we have a system that only lets a (at best) randomly selected chunk of people from the top tier signal their competence? Do you think there are enough top schools that everyone who could get into one finds a spot?

        • Lesser Bull says:

          It’s Asians they try to keep out now, and mostly the extracurricular stuff does it, though probably interviews help too.

    • Tab Atkins says:

      I mean, the SJers don’t seem to be the people saying “Hey let’s only judge people on relevant characteristics, and implement blinding and debiasing measures to ensure that.” They seem to have an altogether different philosophy; I think they might well dispute the whole idea of “judging people on relevant characteristics”.

      This is another one of those “I feel like I’m taking crazy pills” moments I so often get when seeing discussions of SJ here.

      I’m moderately well known in the web tech conference circuit, and I’m friends with lots of people who run or are involved with conference selection processes. Many of these people are vocal feminists, and regularly over over how to get more equality (gender, race, history, etc) in their conferences. The most reliable and loudly trumpeted technique they’ve hit on is blinding, in various ways, most particularly the literal “remove names from the talk proposals while evaluating them”. Multiple people have reported success with this, getting vastly better gender, race, and “first time speaker” ratios when implicit bias is removed this way.

      The idea that anyone can dispute this just boggles my mind. Have you heard of implicit association tests, and how much they’re talked about in feminist circles?

      • Paul Torek says:

        This. So very much, this.

        I think SJ may be getting painted with the Flesh Man Fallacy brush. The same way that Rush Limbaugh and Sarah Palin are made to “represent” right-wing thought in left-wing circles, perhaps Andrea Dworkin is taken to “represent” feminist thought.

        • Oligopsony says:

          Moreover “Andrea Dworkin” doesn’t even represent Andrea Dworkin. (I suspect the same is true of Limbaugh and Palin, but they never seemed interesting enough to investigate.) ((Actually I suppose that the analogy is slightly imperfect in that Limbaugh and Palin are low-quality but in some ways representative while Dworkin is higher-quality but more niche – maybe one could sub in Moldbug or Jezebel as necessary – but this doesn’t interfere with the central claim, so I probably shouldn’t quibble.))

        • Tab Atkins says:

          I feel like most people who get mad at SJ here take their knowledge of what “SJ” is from Tumblr.

          I was going to say that this is like taking your knowledge of theism from Pharyngula or Dispatches From The Culture Wars, but at least those blogs mostly point at people who are actually in positions of power/influence (government, church, etc) and thus have a modicum of respectability to those they talk to. Tumblr is just full of teenagers.

          Learning about SJ from Tumblr is actually like learning about theism from Discovery Institute message boards.

        • Sniffnoy says:

          Sorry, but I think you’re just rounding off all SJ you consider ridiculous to Tumblr! (Also making an error in dismissing the Tumblrites in the first place, but that’s another matter). Tumblr itself is relatively recent, and I can’t say I was as ever far gone as the stuff you see on Tumblr, but that doesn’t mean there isn’t still nastiness out there that is harmful and is less ridiculous-seeming and easier to fall into than Tumblr; rounding it off to Tumblr is a mistake.

          And even groups nowhere near as bad as Tumblr still often have terrible norms and habits of discourse. If you disagree once or twice you may get a polite counterargument, but if you continue to disagree? Not to mention the rampant pattern matching and rounding off to a position or argument they’re familiar with. And even where things aren’t nearly as bad as on Tumblr I’ve noticed a general lack of charity. And I can tell you that the places where I’ve seen this have not all been “full of teenagers”.

          Maybe your experience has been different — but that just means you’ve avoided the problem, not that it isn’t there. Or maybe you dismiss these people because you automatically accord them no legitimacy and thus don’t consider them a problem; but, well, that’s like me making my old mistake of assuming the old-school sexists can’t really be a problem because I accord them no legitimacy. Let me just remind you of this old post of Alicorn’s. Even if we take everything there and credit it to Tumblr, it’s still causing problems.

          I guess now I’ve drifted into “why you shouldn’t dismiss Tumblr even if it is ridiculous”, which I could rant about for a long time, so let me just make one more point — let’s not forget that being able to somewhat-credibly call someone racist or sexist is certainly a form of influence. Sure, maybe the threat of that from someone you consider ridiculous wouldn’t cow you, but it’ll get a number of us to snap into line — because, you know, we don’t want to be evil, and we know we have to shut up and listen, because our common sense is no good, it’s racist and sexist, so we are really interested in learning all the things we need to do to not be evil. I’ll stop there.

        • Tab Atkins says:

          Sorry, but I think you’re just rounding off all SJ you consider ridiculous to Tumblr! […] Maybe your experience has been different — but that just means you’ve avoided the problem, not that it isn’t there. Or maybe you dismiss these people because you automatically accord them no legitimacy and thus don’t consider them a problem;

          That’s possible! I mean, this is a hard thing to separate out; it’s very easy to go No True Scotsman and defend my group by just throwing out the ones I don’t like, but it’s also legitimate to argue that not everyone who *claims* to represent my group really does.

          For example, radfems and lesbian separatists claim to be “feminist”, but they’re of the “men should basically be killed” crazy sect. Calling that “feminism” might be historically correct, but it’s a world of difference from the “let’s fix all the bugs in culture that penalize people based on race/gender/etc” feminism, which is pro-white and pro-men as well, when those groups are unfairly penalized for the race or gender. Lumping them together isn’t useful; it turns “feminism” into a terrible label, like “demotist” is by lumping together liberal democracy and communism.

          So, for one, that kind of problem definitely crops up. I’ve seen in this very blog comments talking about radfem as if it was a mainstream view, rather than a bugfuck-nuts offshoot.

          That said, within things that are close enough to be usefully lumped with “my” feminism, there are definitely some unfortunate types. There’s the SJW template that manifests strongly on Tumblr now, where people use “you’re privileged!” as a way of bullying and feeling powerful, rather than as a way to actually tear down the structures of privilege. There’s Shanley, who’s a firebrand (which are necessary to achieve change!) that I think has gone past the point of usefulness. There’s plenty of others.

          I don’t think I can legitimately distance myself from them; I think an honest “feminism” grouping would include them. But I am willing to call them out as bad, and I think it’s legitimate to complain about people generalizing from them to the whole cause.

          • I have some evidence that the anti-male streak is at least somewhat central.

            I decided long ago that I wasn’t going to identify as a feminist after reading Mary Daly, who had a wide streak of misandry. So wide that if I was to get any good out of her books, I needed to read them with the interpretation that when she said “women” she meant people, and when she said “men” she meant some sort of evil non-sentients.

            If she was accepted by feminists, I didn’t want to join that group.

            When Daly died, there were a lot of feminists who wrote eulogies, and a lot of feminists who called her out for transphobia, but I didn’t see anyone complaining about the misandry. I didn’t look for such, but I’m assuming that I had a reasonably random sample, so it was probably rare.

            The transphobia (which I didn’t notice at the time) presumably came from the misandry.

        • ADifferentAnonymous says:

          Tab: have you read Scott’s post “Weak Men are Superweapons”?

          tl;dr is that sometimes complaining about harm caused by bad members of a not-all-bad group tars the entire group, but you still sometimes have to complain about the harm, but you can also abuse this mechanism, so the whole situation is hard.

          It is indeed a pretty crappy state to have terms like ‘SJ’ and ‘feminist’ include both tumblrites and the speaker selectors you deal with, but linguistically separating them may not be as easy as it sounds. I worry that any term that doesn’t originate with credentialed feminists of some kind is going to read as a slur.

        • Tab Atkins says:

          Yes, I have read that post, and that’s pretty much exactly the reasoning driving me to combat these kinds of assertions.

          A lot of it is context, of course, as you imply. I’m outspokenly feminist on Twitter, for example, so I could probably complain about bad aspects of feminism and get something useful out of it; my history implies that I’m not crypto-attacking feminism as a whole. On this blog, on the other hand, where explicitly anti-feminist groups like NR are talked about seriously and at least somewhat kindly, and many commenters are loudly anti-feminist, it’s difficult to distinguish “constructive criticism” from “weak-man attacks on the greater concept”.

        • Tab Atkins says:

          Nancy Lebovitz:

          I have some evidence that the anti-male streak is at least somewhat central.

          I decided long ago that I wasn’t going to identify as a feminist after reading Mary Daly, who had a wide streak of misandry. So wide that if I was to get any good out of her books, I needed to read them with the interpretation that when she said “women” she meant people, and when she said “men” she meant some sort of evil non-sentients.

          If she was accepted by feminists, I didn’t want to join that group.

          You’re quite right. Mary Daly was part of “second wave” feminism, from which today’s radfem and similar groups have descended. Second-wave feminism was definitely anti-male.

          “Modern” feminism is third-wave feminism, which was started at least partially as a reaction against second-wave’s prejudices.

          There’s been further developments bubbling out of that as well; third-wave started in the 80s, and it’s evolved quite a bit in the decades since.

        • Sniffnoy says:

          (Finally getting back to this after several weeks…)

          For example, radfems and lesbian separatists claim to be “feminist”, but they’re of the “men should basically be killed” crazy sect.

          Now I think you’re being unfair to the radfems! 🙂 At the very least they point out concerns the libfems don’t. I’m sure the libfems have answers to these concerns, but as someone who isn’t intimately familiar with these arguments, I don’t feel comfortable dismissing the radfems out of hand.

          So, for one, that kind of problem definitely crops up. I’ve seen in this very blog comments talking about radfem as if it was a mainstream view, rather than a bugfuck-nuts offshoot.

          Part of this might be variance in environments, but let’s not forget the other factor I’ve already mentioned — once you’ve entered the trap, it becomes difficult to distinguish between different strains of feminism, because you’re required to obey all of them. (As well as whatever superficially pro-woman nonsense can pass itself off as feminism.)

      • Armstrong For President 2020 says:

        It seems like the disconnect here comes from a source deeper than sampling bias; you’re judging SJ/Feminism (and presumably the rest of the Intersectionality cluster) by the ideas which have the most cachet with high-level insiders, whereas outsiders tend to judge it by the results they see when those ideas get put into practice.

        The philosophy of identity-over-competence has been pushed in ordinary people’s faces for decades via Affirmative Action; if your goal was to turn whites into crypto-racists with a single policy, you really couldn’t do better than that. High-visibility media stunts like Occupy’s “Progressive Stack” and the constant hand wringing over Straight White Male overrepresentation (especially in places like the tech industry where whites are underrepresented in favor of asians, or universities where men are underrepresented in favor of women) don’t really counteract that message.

        I’m not one of the people who believes that revealed preferences are the end-all-be-all of motivation. The fact that we often fail to live up to our ideals doesn’t mean they’re unimportant. But with identity politics, it’s pretty clear by now that no-one with any shot of making policy is particularly interested in judging people by “the content of their character.”

        • Tab Atkins says:

          It seems like the disconnect here comes from a source deeper than sampling bias; you’re judging SJ/Feminism (and presumably the rest of the Intersectionality cluster) by the ideas which have the most cachet with high-level insiders, whereas outsiders tend to judge it by the results they see when those ideas get put into practice.

          You seem to have missed the part of my post where I’m talking about real-life selection processes that actually happened. Thus isn’t theoretical noodling on what might work, this is literally what actual people I know have done when organizing their conferences.

          The philosophy of identity-over-competence has been pushed in ordinary people’s faces for decades via Affirmative Action;

          No, that’s the dishonest gloss given by people who oppose it. Google unconscious bias, implicit association, the multitude of studies showing identical resumes being judged significantly differently based on what names are put on them, etc. Forcing quotas is one way to attempt to overcome these problems; when you must hire a minority or a woman, you’ll judge them properly, and your likely to get someone you would have hired anyway if not for the unconscious biases. It’s not ideal, but it’s a simple idea that’s easy to implement and enforce and achieves roughly what it sets out to do, which are good qualities for a government policy.

        • suntzuanime says:

          Affirmative Action is more or less the opposite of the blinding procedures you praise: I don’t see how you can say “no guys, SJ is actually all about blinding” and simultaneously defend Affirmative Action.

        • Armstrong For President 2020 says:

          Real life selection processes, for a certain value of real life which primarily concerns conference speakers. When we’re talking about real life selection processes such as those seen in academic/corporate/government work, the selection mechanisms tend to either be quotas or about-to-be-struck-down-as-disparate-impact.

          In terms of implicit association biases, that would be more convincing as an explanation for gaps if the ‘gaps’ themselves held up to real scrutiny. After accounting for IQ, the Black-White income gap is reduced to about 2%. Something similar happens to the Women-Men income gap when you control for time invested into a job (since women generally work fewer hours, take more time off for their reproductive health, and quit earlier in their careers). The bias on identical resumes is only relevant when resumes are actually identical; typically, they’re not and the bulk of these ‘gaps’ is driven by typical circumstances.

        • Oligopsony says:

          Nobody is claiming SJ is “all about blinding.” Sniffnoy said that it seems like SJ is opposed to blinding on principle, and Tab pointed out that they in fact aren’t. One can endorse a number of policy tools as appropriate for different situations!

        • Tab Atkins says:

          @suntzuanime:

          Oligopsony got it in one. “Blinding” isn’t the goal, “fairer/more accurate representation” is. Blinding is one method, but as I explained further in my comment, quotas are another method (a flawed one, but it has other qualities that make it attractive). They are indeed exactly opposite in method, but their effects are similar – they get women and minorities of appropriate skill into jobs and other positions when implicit biases would have excluded them and selected less-skilled white males instead.

          Armstrong for President 2020:

          In terms of implicit association biases, that would be more convincing as an explanation for gaps if the ‘gaps’ themselves held up to real scrutiny.

          You’re moving goalposts now, talking about pay gaps when I was talking about representation. I don’t have personal examples of pay differences, and I’m not prepared to debate pay numbers right now.

          The bias on identical resumes is only relevant when resumes are actually identical; typically, they’re not and the bulk of these ‘gaps’ is driven by typical circumstances.

          This doesn’t make sense. The resumes were identical to control for confounders; the identicalness is not itself important.

          Are you trying to somehow say that in the real world black/female/etc resumes are actually worse? That’s missing the point; even if we grant the hypothesis that the average resume from those groups is worse than the average white male resume, the point of the study is that equivalent resumes will be judged harsher if you’re black/female/etc than if you’re a white male, so given two people with equivalent experience, the white male will be chosen more often than the other.

        • Alexander Stanislaw says:

          @Tab Atkins

          The argument is that a black person with an identical resume to a white person or asian person is statistically likely to be worse at the job/position (in part due to the existence of Affirmative action itself). In other words race adds more information in addition to the resume. I don’t know if this is true in general but it is certainly true of medical students. Black medical students from the same school tend to have lower board scores and tend to drop out at much higher rates. Mostly because they have lower GPAs and MCAT scores coming in.

          Also, regarding resumes have you seen this? Distinguishing “people are irrationally prejudiced” and “people are correctly using race as a statistical variable” is difficult and more work needs to be done. Certainly some people are unreasonably prejudiced against blacks and other minorities, but the resume studies are not enough to demonstrate that this is a chronic problem.

          I didn’t always think this but JJJ convinced me a while back.

        • Tab Atkins says:

          Also, regarding resumes have you seen this? Distinguishing “people are irrationally prejudiced” and “people are correctly using race as a statistical variable” is difficult and more work needs to be done. Certainly some people are unreasonably prejudiced against blacks and other minorities, but the resume studies are not enough to demonstrate that this is a chronic problem.

          I have not seen that, but it looks interesting, and I’ll read it!

      • Sniffnoy says:

        Ugh, what a mess I’ve made.

        OK: My statement was unfair, or at least misleading. It is not in fact the case that SJers will uniformly oppose blinding measures, nor is it necessarily the case that they will in substantial numbers oppose blinding measures at all. I didn’t actually assert above that they do, but whether I intended it or not, my comment does seem to strongly imply it, so other people can hardly be blamed for reacting as if I did. (And I might well have intended it at the time; I don’t remember offhand.) So, I would like to apologize for the mess I’ve made with my comment above.

        I would like, however, to assert the following weaker statement: Though they might endorse blinding measures, I am doubtful that there is not a substantial portion that would endorse them for different reasons than I would, or possibly, judging by the post, than Scott would (an ethic of judging based on relevant criteria rather than irrelevant ones). Moreover, the nastier ones will go after you for endorsing their policies for the wrong reasons.

        • Douglas Knight says:

          Sniffoy, no, your comment did not imply in any way shape or form that they oppose blinding in general. I am extremely disappointed in your retraction.

        • Sniffnoy says:

          I am extremely disappointed in your retraction.

          <shrug> Whatever — you can be disappointed in how you put it if you like; point is, my claim should be clear now. I mean, it’s not really a retraction, is it? I’m still saying, after all, that “if you could get them to agree with my ‘Hey let’s…’ statement above, they’d be far more in agreement with [Scott] than they actually are”, and that this is a position they’ve rejected, not failed to consider (but not that they have in general rejected blinding as a useful tool for their own ends). And that therefore, if Scott wants to encourage them to use it even more, he’s going to have to do so pretty differently. I still think my original comment was confusing, but, whatever; if you weren’t confused, well, great.

          (Actually, strictly speaking, Scott originally wrote not about blinding, but leaving out interviews altogether. There’s a reasong I wrote “blinding and debiasing” rather than just “blinding”; I probably should have written “blinding, debiasing, etc.” to indicate the general flavor of stuff I was going for — I didn’t mean to single out blinding in particular; simply not doing an interview, while it could be called a form of blinding, would not, I think, generally be considered to fall under that.)

          (Meanwhile, maybe I should have actually stated the college admissions point explicitly, as instead someone else did so downthread but way less politely than I would have, and making additional assertions I wouldn’t have…)

        • Douglas Knight says:

          All those weasel words in your new statement. “am doubtful that there is not a substantial portion”? Why not just say Tab Atkins! He 100% fit your original statement. Fuck disclaimers.

        • Tab Atkins says:

          How do I “100% fit [the] original statement”?

        • ADifferentAnonymous says:

          Seems pretty clear to me that Tab’s goal of “fairer/more accurate representation” means a state where irrelevant considerations are ‘effectively ignored’. There are no doubt some out there whose idea of ‘fairness’ means proportionate representation irrespective of merit, but I don’t think you have grounds to assume that’s Tab’s position. If you disagree, we can ask him.

          It’s certainly true that Tab is willing to endorse policies that result in deadweight loss if the observed disparities are justified, but I think this is because he’s confident they aren’t justified, not because he wants the deadweight loss.

        • Douglas Knight says:

          Tab Atkins, are you in favor of blind testing for admission to college? blind testing for civil service jobs, including fire fighters?

        • Tab Atkins says:

          It’s certainly true that Tab is willing to endorse policies that result in deadweight loss if the observed disparities are justified, but I think this is because he’s confident they aren’t justified, not because he wants the deadweight loss.

          Correct.

          (Though I also think it’s reasonable to assume that some representation differences are *caused* by the representation itself; I’ve heard plenty of stories from women and blacks about, as children, assuming that they couldn’t be X or Y when they grew up because they never saw any X or Y person that looked like them. That could mean that there really aren’t enough qualified people from a given group to hit good demographic balance, but taking on some deadweight loss by hiring them anyway could encourage skilled people of that demo to actually pursue that career. In other words, taking short-term hits might be worthwhile for long-term societal optimization. But this is a much more subtle thing, and I don’t think it’s necessary to lean on this to justify better representation; in the areas that I have personal experience in, skilled people are already routinely passed over in favor of more People Like Me, and so getting a fairer representation mix would actually be a positive in the short-term too.)

        • Tab Atkins says:

          Tab Atkins, are you in favor of blind testing for admission to college? blind testing for civil service jobs, including fire fighters?

          If there is a simple, easy way to blind the admission/hiring processes that doesn’t lose other useful qualities, yes. That’s not always possible.

          (The fact that blinding is hard is part of the reason why I think quota-based stuff is sometimes worthwhile even when it’s not ideal, as I explained above. It accomplishes something similar – forcing you to hire, say, the best black person, rather than filling the role with another white person who isn’t as qualified but fits the interviewer’s biases. Quotas are simple and widely applicable in most situations, without requiring significant changes in existing hiring processes. They’re thus cheaper to implement and easier to enforce, both qualities that are pretty worthwhile in a required-by-law initiative.)

        • Douglas Knight says:

          There are simple methods, like SATs for college. Do you really not know that’s what I’m talking about? Do you really not know that the vast majority of affirmative action is applied to college admissions?

          I chose firefighters because there have been several examples of their tests have been in the news in the past few years, but maybe you don’t know about them.

          By the way, hiring quotas are illegal.

        • Tab Atkins says:

          There are simple methods, like SATs for college. Do you really not know that’s what I’m talking about? Do you really not know that the vast majority of affirmative action is applied to college admissions?

          You asked me a general question, I gave you a general answer. Don’t get angry about specifics when neither you asked about nor I answered about them.

          For the specifics of college admission, I don’t know how well SAT works as an admission criteria. I haven’t studied it seriously, and a quick skim of Google results appears to be mixed, with SAT being reported as a weak to moderate predictor in different things.

          I did not defend particular current uses of quotas (though I’m not necessarily declaiming them either).

          I chose firefighters because there have been several examples of their tests have been in the news in the past few years, but maybe you don’t know about them.

          I am not familiar with firefighter hiring practices.

        • Ialdabaoth says:

          in the areas that I have personal experience in, skilled people are already routinely passed over in favor of more People Like Me, and so getting a fairer representation mix would actually be a positive in the short-term too.

          Positive in what sense? In my understanding (especially when thinking Darkly), it doesn’t MATTER who’s better at what – and in fact, racism and sexism are MORE important, the more innately skilled your underclasses are.

          Life is ultimately about competition, and most status games are winner-take-all. Having an easy way to signal collusion (say, based on skin color or genital shape) means that you can coordinate to maintain your advantage, and leverage it to prevent opposing groups from ever gaining advantage. The more innately talented the oppressed groups are, the more advantage you gain from oppressing them, and the more important it is to maintain the narrative that they aren’t actually talented. So as a cis het white male, why wouldn’t I want to maintain racism, sexism, and all the other discriminatory -isms? What possible advantage is it to me, if the world is improved but my power over it is diminished? Ruling in Hell is always preferable to serving in Heaven, because at least when I’m ruling I can continue to secure my rule. When I’m serving, no matter how nice Heaven is, I have no guarantee that it will stay that way, and that I won’t wind up serving in Hell the moment things go south.

        • Douglas Knight says:

          What I meant when I said that you 100% matched Sniffnoy’s description is that you are in favor of “better ratios” and appear interested in blind methods when and if they achieve ratios that you think are better and not at all because they might help avoid your biases. It’s possible that you just happened to have avoided all the well publicized examples that might force you to confront this conflict, but I think it is a hard coincidence to believe.

          You are welcome to create a test more predictive than SATs, but blind SATs are certainly more effective than anything else widely deployed. Do you live in California? Admission to public universities used to be, basically, SATs+race. Now that they aren’t allowed race, it’s SATs+a bit of race picked up from the essay.

          ADifferentAnonymous, OK, I asked him. Were you surprised? Did you change your mind? Do you think I should change my mind?

        • Armstrong For President 2020 says:

          So what alternate universe have we passed into where there are tons of qualified blacks being passed over for incompetent whites?

          We do remember that whole 85-mean-13.5-SD thing right? That means that the odds of a black american having over a 125 IQ, about the average for medical doctors PhDs and high-level executives, are only 0.15%. Comparatively, 4.75% of non-hispanic whites, 9.18%* of East Asians and 25.1%* of Ashkenazi Jews have the same score. And those differences only amplify the further out you go!

          This really isn’t rocket surgery here; people are not equal, in fundamental and easily measured ways. Any fair system will return unequal results, and equal results will only be returned by unfair systems.

          *Ish. I’ve never been able to find any good information about either East Asian or Jewish standard deviations on IQ tests; I’m using the 15 point white SD as a fudge, but for all I know it could be several points in either direction for either of them. Either way, we’re still talking about prevalence differences of orders of magnitude here.

        • Tab Atkins says:

          Ialdabaoth:

          I’m going to assume that you’re just pretending to be a sociopath in that comment, ironically or whatever. I don’t think it deserves any further response.

          Douglas Knight:

          What I meant when I said that you 100% matched Sniffnoy’s description is that you are in favor of “better ratios” and appear interested in blind methods when and if they achieve ratios that you think are better and not at all because they might help avoid your biases.

          I have no idea what you’re talking about, or how you have somehow read this into what I’ve written. I have said nothing of the sort.

          You are welcome to create a test more predictive than SATs, but blind SATs are certainly more effective than anything else widely deployed. Do you live in California? Admission to public universities used to be, basically, SATs+race. Now that they aren’t allowed race, it’s SATs+a bit of race picked up from the essay.

          …and? What am I supposed to infer from this?

          You also appear to have somehow come to the conclusion that I am against blind college admissions. Stop, go read my response again. I’m not willing to argue with someone who deliberately misreads me.

          ADifferentAnonymous, OK, I asked him. Were you surprised? Did you change your mind? Do you think I should change my mind?

          What do you think you asked me? It certainly wasn’t “do you think ‘fairness’ means proportionate representation irrespective of merit?”, or any question which tracks to that.

        • ADifferentAnonymous says:

          @Douglas Knight I think it’s pretty clear that Tab’s position is “Hey let’s only judge people on relevant characteristics, and implement blinding and debiasing measures to ensure that”, with quotas endorsed as a dirty-hack debiasing measure (but only insofar as they are counteracting a bias!).

          I’ll grant you that his further position of taking temporary deadweight loss to help the next generation realize its potential is a departure from the above statement, but it’s a lot closer to it philosophically than it is to holding equality of outcome as a terminal goal, which I take to be the philosophy sniffnoy was complaining about.

          So I think you should change your mind about what Tab’s philosophy is, and I think you should pursue your legitimate empirical disagreement with him.

          @Ialdabaoth I’m not sure what this thread needs is an exploration of foundational issues with the concept of pursuing net social positives, but the short answer is that it’s cooperating in prisoners’ dilemma and we’ve somehow done an awful lot of it to get where we are today.

          (fixed name error)

        • Douglas Knight says:

          “empirical disagreement”? What are you talking about? The prevalence of affirmative action? of tests?

        • Tab Atkins says:

          Armstrong:

          We do remember that whole 85-mean-13.5-SD thing right? That means that the odds of a black american having over a 125 IQ, about the average for medical doctors and PhDs, are only 0.15%. Comparatively, 4.75% of non-hispanic whites, 9.18%* of East Asians and 25.1%* of Ashkenazi Jews have the same score. And those differences only amplify the further out you go!

          If we grant that whatever IQ numbers you’re using are accurate and usefully track performance at various jobs, you’re still committing a rather basic mistake of conflating population distribution with individual values. This whole time I’ve been talking about removing the effects and consequences of implicit and societal biases on individuals; my argument for better diversity in itself is just an observational one, that when I’ve seen selection processes that purposely blind themselves, they come up with better gender/race/etc ratios than an unblinded one does, suggesting that we leave a lot of value on the table by default due to biases. This suggests that we can target that second-level effect, the increased diversity itself, as a more visible and easier-to-purposely-optimize thing, and get the underlying effect (increased competence) to come along with it.

          ADifferentAnonymous:

          I’ll grant you that his further position of taking temporary deadweight loss to help the next generation realize its potential is a departure from the above statement, but it’s a lot closer to it philosophically than it is to holding equality of outcome as a terminal goal, which I take to be the philosophy sniffnoy was complaining about.

          Indeed, it’s just investing in the future. Spend local fitness now for greater global fitness in the future. I put it in as a parenthetical, and emphasized that it’s a much less important argument, because it’s much more difficult to justify. Ignore it if it’s a problem; I just didn’t want to seal off my arguments in a way that would make me look goalpost-moving if I poked in that direction later.

        • ADifferentAnonymous says:

          Tab contends that companies today generally exhibit a bias against underprivileged groups in their hiring, such that a modern hiring committee with a judiciously-set racial quota would often be a more meritocratic proceeding than the same hiring committee without such a quota. I take it you disagree with this, or else I am very confused.

        • Tab Atkins says:

          Tab contends that companies today generally exhibit a bias against underprivileged groups in their hiring, such that a modern hiring committee with a judiciously-set racial quota would often be a more meritocratic proceeding than the same hiring committee without such a quota.

          I never actually said that. Nothing I said was about influencing the makeup of the hiring committee.

          (I don’t necessarily disagree with what you said, but it’s not part of my point, either. Plus, evidence suggests that minorities are also afflicted with similar implicit biases to majorities, because “majority is awesome!” is part of the shared culture that both are soaked in, so more diverse hiring panels may not have a significant effect on hiring, or at least less of one that I think we could achieve through more direct methods.)

        • nydwracu says:

          If there exists bias against certain underprivileged demographics which leads to the selection of suboptimal applicants of overprivileged demographics, why hasn’t this been investigated and solved by the things doing the selecting? Especially in corporations: isn’t this exactly the type of problem the profit motive is supposed to solve?

        • Tab Atkins says:

          If there exists bias against certain underprivileged demographics which leads to the selection of suboptimal applicants of overprivileged demographics, why hasn’t this been investigated and solved by the things doing the selecting? Especially in corporations: isn’t this exactly the type of problem the profit motive is supposed to solve?

          I know you got into this discussion in another thread, but the answer seems quite obvious: because people are lazy and stupid, and it’s easier to not do it. Corporations are pretty interesting lifeforms with the ability to pretty decently wring profitable behavior out of their lazy and stupid cells, but they’re not perfect.

          There are some examples of people doing this, and professing getting a benefit out of it, but that seems to have gone pretty much unheeded, due to the aformentioned “stupid and lazy” thing.

        • Oligopsony says:

          I am aware of some evidence that more diverse companies perform better across a number of market-relevant metrics, which could point to some money on the ground (albeit the kind that may take effort to pick up,) although I don’t recall offhand what confounders, if any, had been controlled for.

          (On a lower-n level, Alan Greenspan, not much of a SJW, famously acquired a mostly female staff that he saw as arbitraging discrimination elsewhere.)

        • Douglas Knight says:

          ADifferentAnonymous, that has nothing to do with this thread. I suggest you go back and read Sniffnoy’s comments repeatedly and address the question of whether any of them are clearer to you than the others. In any event, tremendous disagreement with you prevents me from taking your advice.

          I certainly think that racial quotas filled by precedence on blind tests are better than blind lotteries for choosing firefighters. But, again, that’s illegal.

        • ADifferentAnonymous says:

          Sorry, by ‘a hiring committee with a quota’ I meant ‘a hiring committee whose decisions are required to satisfy a quota’, not a committee whose membership satisfies a quota.

        • Tab Atkins says:

          Ah, ok. Then yes, that’s one of the things I’ve said.

        • Sniffnoy says:

          Oh my, this blew up while I was away… I’m just going to respond to two things…

          This whole time I’ve been talking about removing the effects and consequences of implicit and societal biases on individuals; my argument for better diversity in itself is just an observational one, that when I’ve seen selection processes that purposely blind themselves, they come up with better gender/race/etc ratios than an unblinded one does, suggesting that we leave a lot of value on the table by default due to biases. This suggests that we can target that second-level effect, the increased diversity itself, as a more visible and easier-to-purposely-optimize thing, and get the underlying effect (increased competence) to come along with it.

          This is a decent argument that diversity can serve as a measure, but just in case you’ve forgotten

          (I am not saying you are necessarily wrong; I am saying that if you have not accounted for Goodhart, you are making a mistake.)

          (Also it’s only really an argument that the current state of affairs is suboptimal; trying to read a target out of it seems like something to be wary of.)

          I’m going to assume that you’re just pretending to be a sociopath in that comment, ironically or whatever. I don’t think it deserves any further response.

          This is a good idea, Ialdabaoth does this fairly often.

        • Tab Atkins says:

          This is a decent argument that diversity can serve as a measure, but just in case you’ve forgotten…

          (I am not saying you are necessarily wrong; I am saying that if you have not accounted for Goodhart, you are making a mistake.)

          You are very right to bring that up. That said, I think it doesn’t get to apply in force here; if you have a quota, you’re still looking to get the best out of each required category anyway. That is, you’re not *replacing* the “I’m looking for the most competent people” metric, you’re just throwing an additional requirement on there.

          (Also it’s only really an argument that the current state of affairs is suboptimal; trying to read a target out of it seems like something to be wary of.)

          Yes, you can’t really read the conditional probabilities that way, if we’re being strict about things and pretending that we can only assume a uni-directional link without further evidence. But I suspect that in real life we can do so. Here’s a supporting anecdote (the post is long and mostly irrelevant; search for the paragraph starting with “I’ve told this story before” and read from there).

        • Sniffnoy says:

          You are very right to bring that up. That said, I think it doesn’t get to apply in force here; if you have a quota, you’re still looking to get the best out of each required category anyway. That is, you’re not *replacing* the “I’m looking for the most competent people” metric, you’re just throwing an additional requirement on there.

          Well, you could argue that you’re not changing it much (is this what you’re arguing? I’m having a hard time telling), but there is going to be some cost for not “just looking for the most competent people” (although that assumes one was doing that correctly, which as you have pointed out, one probably wasn’t).

          Yes, you can’t really read the conditional probabilities that way, if we’re being strict about things and pretending that we can only assume a uni-directional link without further evidence. But I suspect that in real life we can do so. Here’s a supporting anecdote (the post is long and mostly irrelevant; search for the paragraph starting with “I’ve told this story before” and read from there).

          It’s a very interesting story — I’ve seen it before, actually, but thank you for reminding me of it. It’s an interesting strategy and it seems to have paid off here, but I must admit I’m not entirely at ease with it. One would hope that there would be some way to modify it so as to get the benefits of it without having to pay the corresponding costs. But it’s possible that such a thing might not exist — it’s hard to get people to truly put in as serious an effort when nothing is actually at stake — and that something like this is the best we can do.

          I’m wondering if maybe this is getting to the heart of our disagreement here. I have no problem using such positive-discrimination methods — so long as there is a solid argument for doing so, that it’s really the best way, that you can’t avoid it and get the same good results. To my mind, the fact that a method uses any sort of discrimination is a significant cost that needs to be overcome; these aren’t a style of solution you just toss around casually.

          As such, to my mind, arguing for positive discrimination bears a burden of proof. So I expect people arguing for it to make solid, explicit arguments that shows they’ve seriously thought about it, have anticipated objections, and really don’t see a better way — or at least lampshades the gaps in the argument and asks us to spot them the point, to let us know they’re not an idiot.

          But this is not what I see. What I see is the people making flimsy arguments and then papering over the gaps by calling evil anyone who tries to point out the holes. Sometimes these arguments are backed up by data, but, well, data is just the start of an argument, it’s not an argument in and of itself. I mean, honestly, I don’t know statistics. I don’t know how to judge whether you’ve done your statistics right. So I will spot you the data. What I want to see is solid argument structure, because that seems consistently to be lacking! (Then maybe once that’s down I’ll worry about the solidity of the data.)

          And, you know, in a lot of these cases, I can pretty easily see how to take these people’s arguments and improve them and fill in many of the holes myself. But when you’ve annoyed me by making a flimsy argument and acting as if there are no real issues in it, then, well, I don’t want to make your argument for you. I want to disagree with you out of spite, in the hopes that maybe it’ll inspire you to stop making such flimsy arguments. Unfortunately, this never seems to work.

          I’m hoping this maybe makes it a bit clearer where I’m coming from?

  16. In the UK the “last mile” between the exchanges and the customer are forcibly opened up. We have a lot of competition over this distance, and I believe that because of this we’re actually better off than lots of Europe (and far better off than the USA) when it comes to broadband pricing.

  17. suntzuanime says:

    A poll asks consumers whether, in the case of a sudden crisis, they would want their self-driving car to save their own life or maximize the total number of lives saved. For example, if it was about to hit an oncoming car and couldn’t get out of the way, should it veer off the road and down a cliff (100% chance of killing self, 0% chance of killing others) or hit the oncoming car (75% chance of killing self, 75% chance of killing others)? Respondents overwhelmingly in favor of cars programmed to protect their driver alone – which makes no sense, since presumably everyone’s car will have the same programming so this kills extra people for no reason. Immanuel Kant is not amused.

    Three concerns: an emotional one, a decision-theoretical one, and a methodological one.

    Emotional: If I buy a robot car, I damn well want it to be loyal to me. I do not want it to consider whether killing me would serve the “greater good”; if it takes this into consideration, it is not really my robot, it is society’s robot. If your car kills you, this is the same sort of violation as when your phone tells the FBI what whorehouses you visit, except much greater in magnitude. If you value owning technology, rather than being owned by it, take a stand now.

    Decision-theoretical: Unless you ban loyal cars, bad actors will buy loyal cars instead of disloyal ones. If you do ban them, bad actors will hack their robot cars not to kill them. Why support a norm that protects bad actors at the expense of norm-followers? That undermines the meta-norm of norm-following, which is a really important norm.

    Methodological: I couldn’t find the actual survey at your link, but if the example question is representative, there might be a confounding factor: people tend consider a certainty of a bad thing worse than two 75% chances of a bad thing. This might be less about the specifics of robot car ethics and more about people not shutting up and multiplying.

    • Anonymous says:

      Also, Scott’s write-up at least seems to ask specifically about YOUR car: it is rationally selfish to want your car to protect you, even if you want cars in general to be safer overall.

      I didn’t find this study very surprising: cars advertise safety for their passengers, not in general, and the popularity of tank-like cars (or at least, the perception UKers have that USers like driving tank-like cars) would seem to fit with a model that people are willing to escalate overall danger to increase personal safety.

      • Nornagest says:

        SUVs became popular in the US because station wagons (that’s “estate car”, for you Brits) turned unprofitable in the Eighties thanks to poorly-thought-out EPA regulations, but there was still a need for cars big enough to haul around >2 squalling brats. Minivans counted as trucks by EPA standards and could dodge the fleet mileage regs, but those quickly picked up a reputation for being fit only for harried suburban housewives, which is death for any kind of mass marketing. Then some bright boy came up with the idea that you could sell what’s fundamentally a minivan by jacking it a foot off the ground, dressing it up to look kind of like a truck, and showing clips of it tackling the easier parts of the Rubicon Trail in between Oprah segments.

        It worked. SUVs had been serving the rural market for a while by then (the Chevy Suburban’s actually quite old), but that’s how they really took off in Middle America. In a few years they’d became cool enough that even people without families drove them, until gas prices spiked in the late 2000s, aerodynamics and ergonomics slowly pushed them back towards the minivan form factor, and people started realizing that it was kind of a dumb idea anyway.

        Safety issues had almost nothing to do with it outside of post-hoc rationalization. The Ford Explorer was the most popular vehicle in the US at the same time it was getting raked over in the press for rollover issues caused by its factory tires’ unfortunate tendency to convert themselves into ballistic shreds of disincorporated rubber while screaming down the highway at 110 KPH.

    • Erik says:

      suntzuanime’s emotional concern applies to more than just technology. It’s also the case for most lawyers: when you hire a lawyer, the lawyer is supposed to work specifically for you, which means keeping your secrets even if those secrets could be used “for the greater good” by revealing them in court and screwing over your case.

      In some places, a similar rule applies to doctors: if you’re admitted to the hospital after doing something illegal, you can inform the doctor what you were doing (to aid in your treatment) and the doctor is forbidden to report this to the police.

      Priests, of course, must have the same rule about the confessional if the confessional is to get any traction.

      • Fnord says:

        I believe it’s generally thought that lawyer/client, doctor/patient, and clerical privileges are overall utility maximizing. Even if in one specific case utility would be maximized by a doctor sharing information about a patient’s crime, if doctor/patient privilege weren’t the rule patients would simply lie. We still wouldn’t find out about crime AND patients would get inferior treatment.

        That’s a somewhat similar argument to the decision-theoretic one suntzuanime posits for “loyal” self-driving cars, but I think the case for self-driving cars is significantly weaker, given the difference between what it takes to defect.

        • Erik says:

          Going back to technology, I’m going to seriously support the principle that my device should answer to my will. This isn’t even an issue for ‘stupid’ devices – when I buy a saw, for example, no reasonable person suggests that the saw should be GPS-enabled and tied into the National Property Registry and have retractable teeth so that the saw can lock itself down when it detects it’s outside of my property to prevent me sawing down my neighbor’s postbox, because this is a terrible way of preventing me from sawing down my neighbor’s postbox.

          The complaints against Digital Rights/Restrictions Management systems seem similar too. Many of those partly convert “your” computer into “your and Sony’s/EA’s/whoever’s computer”, and weasel around “sale”, claiming that you haven’t really bought the thing, you’ve entered into a licensing agreement which may be terminated and updated and generally renegotiated in ways that invite lawyers. Which ties into many of the problems of haggling, something most societies moved away from for good reasons, such as haggling being time-consuming, energy-draining, and unpredictable.

          I worry “societal cars” might end up with similar problems of the programmed car behavior being different from region to region, varying by type and number of passengers, having stupid exploits, etc.

          Still, self-driving smart cars seem like a great tool for reducing car accidents, so let me pretend I’m Scott Alexander and can put a semi-tangential section header here before veering back to the point about cars:

          II.

          I have heard anti-deathist people from LW, MIRI and similar places sometimes complain that fiction depicts immortality as a bad thing, making the general populace suspicious of their work and ideas. For most of my experience with fiction, though, I would add a strong caveat: Costly and/or scarce immortality is depicted as a bad thing.

          For example, if the immortality treatment requires some very rare resource and costs two million dollars per year of life (extendable indefinitely), so only a rich minority of people can afford it, then yes, your society is likely to be pretty badly screwed. The entire GDP of the US (2013) then buys immortality treatment for only eight million person-years; most places aren’t as rich as the US, and one can’t spend the entire GDP on immortality treatments, so in all likelihood it’s going to be less than 2% of people getting it, and a lot of the ones who don’t get such treatment are going to be very, very willing to go to extreme lengths to get their hands on some of it. After all, if they lose, what’s the worst that can happen – they die?

          Some numbers: A quick glance at Financial Times indicates there are about two hundred thousand people with assets over 30 million dollars, which seems about the order of magnitude where your interest exceeds two million dollars a year and you can get “true” immortality by buying the immortality treatment indefinitely.

          That’s assuming a cooperative world society where the immortality treatment is in the public domain and can be performed with simple tools and little training, it’s just costly because of the resource requirements. Suppose instead that instead of being a pill containing costly materials, the treatment is a complicated tweak to the human body (still providing one year per operation, extendable indefinitely). A doctor instead needs twenty years of training and a custom-built laboratory to perform the treatment, and very high initial aptitude is required to qualify for the training in the first place. In this case the marginal cost per treatment is far lower, but the initial investment in being able to dispense the immortality treatment is higher.

          What’s going to happen to the first few doctors who complete their training? Preposterous levels of security against preposterously strongly-motivated kidnap and brainwash attempts, I assume, not only on the doctor but on their friends, family, and potentially their entire hometown or worse.

          Then there’s the case where the immortality treatment is a strategic secret. Wouldn’t do to have those guys becoming immortal, after all. Combine with the previous case for strikes on doctoral training centers, and so forth.

          Fiction usually stays interesting by covering some sort of conflict. Fiction usually stays conveniently short (or hides the length in appendices and supplements) by not covering the worldbuilding in detail. This seems to me a quite straightforward reason to expect that whenever a work of fiction deals with immortality at length, it’s not going to be the good sort of “solved” immortality where everyone largely lives forever. Some works have that (John Wright’s Golden Age comes to mind), but there it’s a background element.

          III.

          So, while I support worldwide immortality as a general principle, I’m willing to defend the worries about what one might gently call “implementation difficulties” resulting in a worse world situation by introducing a massive new cause of conflict.

          I think a similar attitude is justified towards smart cars as a means of ending road deaths (as a subset of deaths in general): great once the kinks are worked out, but fractional implementation seems unlikely to result in proportionately fractional good.

          For example, if only some of the cars on the road are smart and supposed to minimize loss of life in an accident, their attempts at counting life in the other car might get abused by the guy who buys three crash test dummies and drives around with them in his stupid car to get a higher ranking in the smart cars’ algorithms. This could be fixed once all cars are smart and properly count (and communicate) number of passengers to be valued, but along the way, I foresee a bunch of semi-fixes like making it illegal to drive with crash test dummies in your car, then some other guy discovers that if he puts his dog in the front seat, smart cars will count it as a person. The legislative authority gets smart enough to not play coding-by-exception, so they pass a general law against spoofing the smart car life valuation algorithms. The lawyers are invited to resolve exactly what constitutes “spoofing”. Appeals of edge cases go to the supreme court. Smart car manufacturers try to update their algorithms at the same time as the legislators and judges update the legal landscape. Angry dog owners complain that they can’t drive their dog around in their two-seater. Some bright senator gets the idea of having cars use marital status as a tiebreaker if there are equal numbers of people in both cars when deciding which one to sacrifice. It’s shot down as discrimination by marital status. Conspiracy theorists get it into their heads that the government wouldn’t bother to deny it unless it were true, start spreading urban legends about how else the smart car life valuation algorithms discriminate. A bereaved widower files a lawsuit asserting that his pregnant wife should have been counted as two people when determining whether and who should have swerved. The smart car debate gets abortion politics all over it.

          IV.

          For the record, I don’t have a car, and I don’t feel a need for a car. I bike to work. I used to take public transport an hour and a half every weekday for several years. Public transport around where I live has been great in my experience, and public transport crossed with Uber is sort of the model that I’m hoping a citywide network of smart cars can eventually converge on: signal your destination, get in a car going that way, swipe card to pay fee, travel, disembark.

          I think developing the socialized smart car system from something like Uber, the public transport fleet, or a new project might be the best way to go about it, gradually obsoleting existing private cars. The suggestion that has seemed implicit in the discussion so far – putting utilitarian restrictions into otherwise private smart cars, a sort of socialization by degree – sounds to me as though it has a lot of potential to lead to interesting times, with complaints about slippery slopes and a lot of anger about perceived or actual infringements on people’s property and other rights.

        • Tab Atkins says:

          Note that all these moral dilemmas are neatly side-stepped by Google whenever they get mentioned, by simply stating that the cars are too dumb to do that, and instead they just follow traffic law to the best of their ability, and anyway the robot cars are much less likely to get themselves into such a situation in the first place as they are much safer and more defensive drivers than the humans they’re replacing.

      • Stephanie says:

        Good point about all those professionals. I hadn’t thought to make that comparison.

  18. kylind says:

    On Vanderbilt: At the height of the family, the fortune was about 200 million dollars in the late 19th century.
    That would make it about 5 billion in today’s dollars (and not 300 billion).

    • Douglas Knight says:

      Where do you get your inflation numbers?

    • Scott Alexander says:

      Judging by the CPI calculator, you seem to be right. But the article says $300 billion, and Wikipedia says $100 billion. I’m going to put this one down to “inflation is hard to calculate”

      • Eric Rall says:

        The higher numbers are scaled with GDP growth (to highlight his wealth relative to the economy as a whole) rather than the more conventional purchasing power estimates.

  19. US says:

    “a growing body of evidence that job interviews are terrible and strictly inferior to judging candidates based on their accomplishments or test scores”

    ‘Accomplishments’ are much worse predictors than are (IQ) test scores and my reading of the literature indicates that they are probably not much better than structured interviews in most cases (mainly because the validity of GPA as performance indicator falls over time – unlike IQ test scores which tend to retain their validity over time). Some data:

    “the validity coefficient for unstructured interviews as predictors of job performance is around r = .15 (range .11 – .18), while that for structured interviews is around r = .28 (range .24 – .34). Cook (2004) calculates the overall validity of all interviews over three recent meta-analyses – taking job performance as the common denominator of all criteria examined – to be around r = .23.”

    “meta-analytic estimates provided validities for biodata [‘accomplishments’] in the region of .25”

    “In the past ten years meta-analysis has provided compelling evidence for the validity of GPA in occupational settings. Most notably, Roth and colleagues reported corrected validities above .30 for job performance […] and .20 for initial level of earnings […] the highest validity was found for job performance one year after graduating, with validities decreasing thereafter”

    I have written more about this stuff here and here. Generally it’s worth noting that it is really hard to find consistently valid, stable, and solid predictors of job performance, in the sense that even the best predictors leave a lot of performance variation unaccounted for. It’s important to note that whereas past job performance might well be a reasonable predictor if you could somehow get ‘the true variable’, it is often not easy for potential employers to correctly estimate this variable, e.g. because some of the indicator variables used to assess it may be problematic – for example “an assessment of the existing evidence suggests that the reference is a poor indicator of candidates’ potential.” (another quote from one of the linked posts).

    • Anonymous says:

      >‘Accomplishments’ are much worse predictors than are (IQ) test scores

      I disagree. Accomplishments are an indicator of future accomplishments being likely.

      >mainly because the validity of GPA as performance indicator falls over time

      There are accomplishments other than GPA, and which mean much more than GPA.

      • US says:

        “>‘Accomplishments’ are much worse predictors than are (IQ) test scores

        I disagree.”

        You should not confuse this for an opinion – I don’t really have strong opinions about these things. It’s not an opinion, it is the result of decades of research on the topic. IQ performs better than any other performance indicator.

        “mainly because the validity of GPA as performance indicator falls over time”

        The original comment included a separate section on GPA and test results, and apparently when making a few changes to the original comment I muddled things up a bit. I’m well aware there are other types of accomplishments, and you would be aware that I was aware of this as well if you’d followed the links and read my posts about the book I read about this literature (I explicitly made clear in my comment, via the links, that I read a book on this literature. I’m puzzled why you’d think a ‘past performance is a good indicator of future performance’ (paraphrasing)-argument would add any value at all).

        My argument is r=0.28 from metareviews on structured interviews, vs 0.25 for biodata. Your argument is: ‘Accomplishments are an indicator of future accomplishments being likely.’

        You can probably combine biodata with GPA to improve upon the value of a composite ‘achievements’ variable – which was my main reason for including the isolated GPA results in my comment, though as I mentioned it got a bit muddled up – and it does make some sense to do this if one wants to have a discussion about the value of a variable like ‘accomplishments’, but we dont actually know very much about the relative values of various combinations of performance indicators; there does not seem to be much of a literature on this stuff, so the effects are largely speculative.

        A few things we do know is that biodata is mostly used in a very inefficient manner by firms (‘gut feeling’ vs ‘systematic and validated scoring mechanisms’), references are pretty much useless, and GPA is not that great once the applicant is no longer in his early twenties (but does retain some value even so); all of which makes the ‘past achievements’ variable much less useful than you’d think. It’s true that there are other variables than GPA, but you’re incorrect that they “mean much more than GPA.” I actually implicitly told you that in my first comment; most of the time the other variables (biodata) seem to be even less useful than GPA is.

        • Oligopsony says:

          I feel like this may be Bad Rationalism here but regardless of the data, I can’t quite get myself to believe that IQ is more predictive than past performance. To generalize from one example: I test decently enough but am basically a fuckup in every meaningful way. I think the implication of “IQ is more predictive than past performance” should suggest that my consistently having fucked things up in the past is just a series of flukes and I should start being decently successful any day now, which is flattering but… just wildly implausible. Maybe I’m misinterpreting you, though.

        • ozymandias says:

          I agree with Oligsopony for p much the same reason.

          I have an absurdly high IQ, but I’m a fuckup for reasons that do not show up on IQ tests. I think those reasons are probably relevant for people who would want to hire me for jobs, and they should pay attention to those reasons and not just my absurdly high IQ.

          • Protagoras says:

            I’m with the O people, despite my username starting with a P. I’d suggest forming a support group, but committing to useful ideas like that is the kind of thing I’m pretty bad at following through on (due to ADHD and depression and probably other issues).

        • Ialdabaoth says:

          Fourthed, but I think we’re all somewhat unusual.

          More specifically, I think that for every one of us, there’s at least two people who “got lucky” and can’t sustain it – so statistically, it’s true that high IQ is a better indicator of future performance than past performance.

          (I must be doing better, in that I did not immediately trot out the “if you aren’t wildly successful, then you’re clearly lying about being smart” meme.)

        • US says:

          “I feel like this may be Bad Rationalism here but regardless of the data, I can’t quite get myself to believe that IQ is more predictive than past performance”

          ‘Regardless of the data…’ That does sound a bit like bad rationalism 🙂

          As I mentioned I don’t have strong opinions about this stuff, and I was quite surprised to learn just how well IQ performed compared to other metrics when I read the book (though I had read stuff on related topics before, so it wasn’t a completely new insight that the variable is very useful). But the data is quite clear; not only is IQ a really good predictor, but it also seems to explain much of the same variation other variables do – meaning that most other variables don’t add very much once you’ve included IQ.

          It’s not outrageous to contend that past performance might somehow be as informative, perhaps even more informative, than IQ, if only we knew how to get at it and identify the variation that is important. A big part of the problem is how to access and assess the data; you can’t include Everything That Person Has Ever Done in your ability assessment, so you have to restrict the analysis to specific variables related to past performance and instrumentalize these. People do that, and when they do apply such performance indicators and compare them with IQ, IQ performs better than the other performance indicators.

          Much of the problem relates to the operationalization of past performance; it’s not trivial to figure out which variables to include and which weights to give them. People are different, so even the same variable may sometimes tell you different things about different people. The Complete And True Information About Past Performance is not available, e.g. because the competitor may have an incentive to lie to you about which of his soon-to-be-former employees are high performers and which are not, or because people lie about their job history. Firms also sometimes act in a manner which may not lead to good matching by lying about what the true job function is, or what the true career trajectory of people getting the job may look like. Figuring that ‘you’ll know it when you see it’ and not going about analyzing the past performance data in a rigorous and systematic manner – which is probably quite common practice in the real world – incidentally drastically lowers the value of past performance indicators and makes them even worse.

          You may be able to collect multiple different types of variables (somehow combine Big Five traits, references, and biodata into one metric) about this past performance stuff and compare them with IQ and get better validity – but as I mention there does not seem to be much of a literature on this stuff despite the potential importance of such effects, and so it’s not really clear such combinations would do better. What is clear is that gathering such combinations of different variables and combining them analytically into useful performance metrics is likely to be much more expensive than are IQ tests.

          ““IQ is more predictive than past performance” should suggest that my consistently having fucked things up in the past is just a series of flukes and I should start being decently successful any day now, which is flattering but… just wildly implausible. Maybe I’m misinterpreting you, though.”

          No, it should suggest nothing of the sort. You’re just one individual and IQ doesn’t explain everything. Plenty of other stuff’s important as well, one point is just that it’s hard to figure out what that ‘other stuff’ is at the individual level. You may have fucked up despite a high IQ, but compare 10.000 people with an IQ of 130 with 10.000 people with an IQ of 70 and you’ll see large between-group differences in their probabilities of fucking up (and large differences e.g. in the kinds of careers the people who do not fuck up can aspire to).

          “I’m a fuckup for reasons that do not show up on IQ tests. I think those reasons are probably relevant for people who would want to hire me for jobs, and they should pay attention to those reasons and not just my absurdly high IQ.”

          So most employers do. That is, pay attention to the other stuff as well. Because they know the other stuff’s important as well. IQ is the best variable, but it leaves a lot of stuff out as well, and combining multiple variables is likely to be useful. Though we don’t know that much about such aspects, some studies have found that other variables (e.g. conscientiousness) do have incremental validity over IQ; if this is the case, including both variables in the assessment is likely to be smarter than limiting the analysis to just one of them.

        • Armstrong For President 2020 says:

          Height correlates well with health and other life outcomes, but once you start getting a few standard deviations out you end up with some seriously unhealthy people.

          There are a lot of similarities between height and intelligence, particularly that both are highly polygenic, so it shouldn’t really surprise us if geniuses and giants are generally playing by the same set of rules here.

        • US says:

          “once you start getting a few standard deviations out you end up with some seriously unhealthy people.” – you think the same applies to people with high IQ? Do you have good reasons for thinking that (i.e. studies/books)?

          I ask because I am not familiar with data indicating this to be the case, and I don’t like to rely on evidence like ‘mathletes seem quite odd’. Evidence may exist and I’m certainly not ruling it out, but mathletes may be odd (to go with the example at the link), if they are odd, for all kinds of reasons unrelated to their intelligence; they’re a highly selected sample of individuals which may have a lot of things in common aside from just being smart. I’m member of a chess club. I can tell you that selection effects are certainly relevant there. ‘Odd’ high-IQ people may be a small minority of the very smart people, but for various reasons – e.g. ‘most people like to think that there’s a downside to being smart, because otherwise life would be unfair to people who aren’t’ – we notice the odd ones but not the normal ones. Lots of very smart people may both be unaware that they’re smart and quite normal. I don’t know, but I don’t like to rely on impressions about these sorts of things, because such impressions may be wrong. For example:

          “There are widespread myths about the psychological vulnerability of gifted students […] In fact, a comprehensive survey of the research on this topic finds no evidence that gifted students are any more psychologically vulnerable than other students”

          From this link.

        • ozymandias says:

          Gifted is an IQ cutoff of 130. I don’t know about other people but I am very much above 130. (Not sure where; I broke the test in kindergarten and haven’t been tested since.) It is possible all the hyperfunctional people in the 130-145 range are compensating for us fucked up people in the stratosphere.

        • US says:

          “I broke the test in kindergarten and haven’t been tested since”

          In that case I’d be cautious about drawing strong conclusions about your present day IQ – tests display quite a bit of variation both from test to test (even though they tend to measure similar things and correlate quite a bit, the same individual may in some cases score substantially different scores on different tests) and when taken at different points in time; the robustness of early-childhood estimates in particular have been questioned as IQ does not seem to stabilize until later on, perhaps at the age of 10 (“The consensus in the literature is that cognitive abilities are established relatively early on in life – IQ, for example, is known to stabilize by about 10 years of age”, p. 4 here). Tests taken before IQ has stabilized may be somewhat unreliable.

          Age-related variation in scores may be quite large, and of course they are not the only source of uncertainty; for example problems like faulty scoring by psychometricians analyzing the data may also cause substantial problems – see e.g. the ‘Errors of measurement’ part of this book (p.138->). Rees and Palmer found that: “Between 6 and 12 years of age, and again between 12 and 17 years of age, 30% of the subjects changed by 10 or more IQ points.” (described here, p.458).

          A thing which is in particular worth having in mind in this context is that the confidence intervals are highly asymmetric in the tails (see page 143 of the book to which I link above), so that you’re more likely to score lower than higher on a second test if you’re at the top the first time, and likely to score higher than lower if you’re at the bottom: “on the WISC-IV, the 95% confidence interval given for a Full Scale IQ of 60 is 57-75”, and “on the Binet-5, the 95% confidence interval for a Full Scale IQ of 132 is 127-135″”.

          I’m not familiar with the specific test you took and I have no idea if this observation is relevant in your case or not, but ‘breaking the scales’ may not be a big deal, depending on the type of test taken, as different tests have different sensitivities; for example the Danish Mensa admissions test only yield reliable estimates up to an IQ of 135 (corresponding to a sensitivity cutoff roughly at the 1% mark).

          Please don’t read the above comments as me implying that you’re ‘stupid’. It’s probably rather that I think you may be overestimating the predictive value of the test results you got when you were a child.

          It may be true that the value of IQ changes at the very high end of the scale, but I have not come across evidence to that effect and before I see such evidence I’ll consider it a speculative proposition and nothing more.

          “Is anything known about what proportion of high IQ people are knocked out by depression/akrasia/whatever?”

          Presumably, but this is not something I know anything about. Wikipedia gives this.

  20. Mercer says:

    On the chimps study: is it possible that humans are trying to second-guess too much? The beauty of Nash equilibria is that they only require a very basic model of the other player’s intent, not a full theory of mind. Perhaps if humans thought they were playing against robots/computers/chimps they would be better at this? Though the chimp accuracy level sounds incredibly impressive!

    • Jake says:

      I think this might be a situation where verbal reasoning hurts. A Nash equilibria seems like the kind of thing it’s easier to grasp intuitively than it is to put into words. So a chimp just looks at the situation and makes a judgement, while a person who isn’t explicitly familiar with Nash has to try to reason it out more explicitly.

      I would actually expect that language caused a lot of problems like this, but it’s just so incredibly useful that it caught on anyway.

  21. “A snide objection I’ve heard to the free market is that cable companies are capitalist and look how terrible they are. But one sign of how far cable is from an ideal market is the new poll finding over half of people would switch cable companies if they felt like they could.”

    The “ideal market” seems to play a conceptual role for many libertarians similar to the role played by the “ideal state” for many communists. As with the communists, it seems worth thinking hard about why the ideal often isn’t achieved. A standard answer is often “regulation”, but it’s hard to believe that’s the core issue with the cable companies. Rather, infrastructure costs create natural economies of scale that lead to monopolies or duopolies, not true competition. Without major changes in technology the notion of an ideal market in cable is a mirage.

    • Jake says:

      Yeah I also thought that was a weird statement. Just because cable is far from an ideal market doesn’t mean it isn’t part of capitalism. Markets work really well in a lot of situations, but there are also a lot of situations where they sort of break down. This of course includes any industry that’s a natural monopoly – like cable or anything else infrastructure related.

    • ADifferentAnonymous says:

      Pretty sure most free market types acknowledge that natural monopoly (that is, fields where things like infrastructure costs tend to create monopolies) is a potentially valid reason for regulation. Milton Friedman even said (in his free-market opus Capitalism and Freedom, I think) that it might be better to have the government run such industries(!).

      Cable is pretty clearly one of the most naturally-monopolistic fields, so the fact that it’s not an ideal market doesn’t tell us that much.

    • Anthony says:

      Natural monopolies usually aren’t, once you actually look at them. The reason that the U.S. had a telephone monopoly nearly everywhere was that AT&T lobbied and bribed local governments, and eventually the federal government, to grant them exclusive franchises pretty much everywhere. Those poles carry telephone and power; there’s no reason they couldn’t carry a second telephone line (or that other poles could be set up), except that local governments under the influence of Progressive ideology wanted the goodies that AT&T promised them in exchange for eliminating “inefficient” competition.

      Mexico had competing phone companies over 100 years ago – my great-grandparents had an Ericsson and a Bell telephone before they fled the Communists. If it could work in Mexico in the early 1900s, it could work in the much richer U.S. of the mid-20th century or the present.

      The actual “natural monopoly” is the poles; which aren’t really, but cities may want to regulate them to limit the visual clutter of having lots of poles to carry lots of utilities. But given a network of “telephone” poles in a city, there’s no reason that multiple cable companies or multiple telephone companies, or multiple internet providers couldn’t share those poles.

      • Jake says:

        Really there just isn’t as hard and fast of a line between ‘natural monopolies’ and not. It’s more of a spectrum where the larger the cost of entry and the greater the economies of scale, the more likely an industry is to be taken over by a monopoly or oligopoly.

        Your example of the government putting up the poles but then letting anyone use them is a good example of government intervention to improve the market. If they didn’t provide any poles the most likely scenario is for a firm to put up it’s own poles, and then dominate the industry through it’s control of that scarce resource. Of course another company could put up it’s own poles, but that would take huge amounts of resources before they could be at all competitive with their opponent’s more established network. By providing public goods in the form of the poles, the government can reduce the barriers to entry and make it more difficult for a monopoly to form.

        The problem in the US is that our ‘free’ market for cable produces the worst of both worlds. We’d be better off if the government just took over and ran the whole thing – as a number of municipalities have done to great effect, or if the the government intervened to produce more competition as you describe. I’m fine with democratic accountability for my internet, and I’m fine with market accountability, but you need at least one.

  22. Andrew C says:

    Both Noah and Zero Hedge and entirely right here (never thought I’d type that). Japan has been engaged in expansionary fiscal policy combined with tight money for decades. Realizing that this sort of thing is stupid is why in the US we mostly have neo-Keynesians instead of Keynesians, who only talk about the need for fiscal expansion when the monetary authority is constrained by not being able to move it’s interest rates below 0%. Recently Japan has changed it’s policiesr and the economy is doing much, much better but the government still has a massive pile of debt from the previous policy. So while Japan is clearly doing better, it’s not clear whether we can say it’s objectively doing well. \

    • Salem says:

      Pretty much this.

      Japan under Abe is doing (roughly speaking) the right thing, for the moment. But Japan also has to deal with the legacy of 2 decades of failed economic policy (“the lost decades”) which are definitely a Keynesian dystopia. Moreover it will be very hard for Japan to rectify things in the long term, because there are toxic long-term political effects from continually running massive deficits and continually engaging in massively wasteful/corrupt spending. As Hazlitt wrote “Today is already the tomorrow which the bad economist yesterday urged us to ignore.”

  23. B.B. says:

    Scott Alexander said:
    Job Interviews Reward Narcissists adds to a growing body of evidence that job interviews are terrible and strictly inferior to judging candidates based on their accomplishments or test scores. I wonder if the social justice people can be directed to attack companies using job interviews?

    It is the social justice people who have pushed for relatively more subjective hiring criteria like job interviews over more objective measures like tests of cognitive ability because they are supposedly racist. Blacks are more narcissistic than whites and homosexuals are likely more narcissistic than heterosexuals. Since they are at the top of the social justice advocates hierarchy of the oppressed, biasing hiring criteria in favor of them would be a plus. Sure, Asians are disadvantaged by such a process, but since they are a high-achieving group, they are essentially dishonorary whites anyway. They are already discriminated against in elite university admissions by well-meaning racial egalitarians, so why not in job admissions as well?

  24. Vanzetti says:

    Say it after me: psychology is NOT a science. 🙂

  25. Jake says:

    The two posts on Abe and the Japanese economy aren’t totally contradictory. The Noah Smith article is right that Abenomics seems to be improving the Japanese economy, and the Zero Hedge one is right that this is coming at the expense of continuing to grow the already large Japanese government debt.

    One thing Zero Hedge might want to keep in mind though is that the Bank of Japan now owns 20% of all Japanese government debt, and is currently buying about 70% of newly issued debt every month via Quantitative Easing. So while the debt is growing, the Japanese government is essentially borrowing most of that money from itself. The main potential downside of this strategy that people worry about is that creating all the money needed could cause inflation, but since Japan has been struggling with DEflation for the last 20 years that’s less a downside and more a bonus.

    Of course really they’d be better off dropping the charade and just letting the Central Bank deposit money directly in government bank accounts – which would produce the same improved financial situation for the government and the same monetary effects, but without the side effects of distorting asset prices and increasing wealth inequality. Plus it would make it more obvious to people how government spending really works – that a government that operates using it’s own currency has it’s spending restricted only by the inflation rate, not by any actual fiscal constraint.

  26. Hainish says:

    From the sex differences paper: A corollary is that, when investigating sex and personality as predictors of a given outcome (such as health, self-esteem, and so forth), cleaner and more meaningful results are likely to obtain if personality is measured at the level that yields the most clearly sex-differentiated profiles.

    Can someone explain why this quote isn’t as horrible as it appears on first impression?

    • Nathanael says:

      It is as horrible as it appears.

      The author admits in one of his responses that his goal is to torture the data in some fashion which will show big sex differences. (He doesn’t put it that way, of course, but that’s what he says about his non-standard methodology.)

      I actually bothered to do the cleanup work of reading these papers. They’re the same bilge we always see.

      Here in reality: No, there are no important psychological differences to speak of between “men” and “women”.

      Almost every paper which claims to find such differences (with the exception of some which only claim to look at socially trained behavior in one society) is being misleading in a very particular way: they are lookin at the difference between the “average man” and the “average woman”, but they are not looking at the variance among men, or the variance among women. The variance within one gender is invariably far, far larger than the difference between the “average man” and the “average woman”.

      Any psychological differences between the “average man” and the “average woman” are therefore *irrelevant in all practical terms* as they can have *no political or clinical significance* — with individual variation being far larger, paying attention to the difference between the averages is always an error and highly misleading.

      I’ve looked at far too many papers like this. They all use exactly the same method of lying with statistics; it appears to be the favored method of sexists who really, really, really want pseudoscientific justifications for sexism. Honestly, I wish they’d mix it up a little and try another method of lying with statistics… it’s boring to see the same one over and over.

      • Watercressed says:

        >Any psychological differences between the “average man” and the “average woman” are therefore *irrelevant in all practical terms* as they can have *no political or clinical significance*

        Even small differences in the mean that are << than individual variation will cause significant differences on the proportions of people far from the mean.

      • Alexander Stanislaw says:

        The variance within one gender is invariably far, far larger than the difference between the “average man” and the “average woman”.

        I’ve never understood claims like this. Consider a one dimensional problem.

        Population A has a uniform distribution in X over the interval [0,50].
        Population B has a uniform distribution in X over the interval [50,100].

        Difference between the averages = 50
        Variance of X in both populations = 208.33 !

        That’s what happens when you compare quantities that don’t have the same units. Maybe you meant standard deviation. I still don’t think this is a useful way of comparing populations.

        Consider the following hypothetical. Suppose men masturbate on average 8 times per month with an s.d. of 10. And suppose that women never masturbate. The difference between the averages is smaller than the s.d. And this means nothing.

        Consider one last problem:

        A has a Gaussian distribution G(-1, 2)
        B has a Gaussian distribution G(1, 2)

        They look like this. I think there is certainly is difference between those populations.

        What we are usually interested in when comparing populations is the degree of overlap. As Vaniver points out this gets even messier in the multivariate case. Glancing at the averages and variances is quite primitive in the one dimensional case but it really breaks down in the multivariate case.

        Edit: And as Watercressed reminded me above, even with a high degree of overlap between populations, differences in the means can still lead to significant differences on the tail ends. For example if men are somewhat more aggressive than women, then they will be significantly overrepresented among the “super aggressive” category of people.

        On a less mathematical note. To me this paper is interesting because it asks the question – to what extent can we separate men and women on the basis of personality? Is there a set of questions we can as to get a good separation? I can’t imagine anyone answering no to this question – but the authors of this paper actually try to quantify what those questions are and what the degrees of overlap are. To no one’s surprise casual sex, masturbation and aggressiveness are near the top, but there are some other interesting variables as well.

        • Douglas Knight says:

          Actually, this paper does not use casual sex, masturbation, or aggressiveness, but only uses 16 subfactors of the 5 factor model.

        • Alexander Stanislaw says:

          Darn. I misread your other comment and confused the two papers.

      • Douglas Knight says:

        they are lookin at the difference between the “average man” and the “average woman”, but they are not looking at the variance among men, or the variance among women

        This paper measures differences in units of standard deviations, known as the z score or Cohen’s d. If you don’t know that, it’s hard to believe anything you say about it.

      • Anonymous says:

        Well yeah, Nathanael does not really know what he’s talking about. 95% predictive accuracy is not something to dismiss for any reason provided here, for instance. However the whole field of study this relates to could be criticized from more basic principles. We did have discussion on this blog not too long ago about how the Big Five might be pretty terrible and almost all social psychological research has lots of issues, replication and such.

        These researchers did not use “non-standard methodology” in terms of what you would see in actual mathematics, fields like machine learning and so on. If anything they applied techniques that other psychological researchers are inexplicably decades behind on.

        This is a point in favor of all psychological research along these lines just being bad, including that which yields the result that men and women don’t differ in personality, because the underlying data is invalid. So more studies would have to be constructed from the ground up, but it’s not a unique criticism for what these researchers did.

      • Scott Alexander says:

        The entire paper is about degree of overlap and uses measures derived from standard deviation. Either you haven’t read it or one of us is very, very confused.

  27. Lavendar bubble tea says:

    I just find the Finnish memes really really funny despite not knowing anything about Finish culture. My first thoughts were along the lines of “WOW! Is there really a social culture like this? This is so fun! Oh my god, it would be so super fun to do a Nordic LARP of a society really like that and I so want to try the bicycle thing. I wish I had the skill for it.” …this is coming from a person with a history of essentially being a hermit for two years due to panic attacks and social anxiety. I don’t understand where these feelings are coming from. (Then again, my reaction to the HIV blood homeopathy stunt was “Wow! So glamorous. This person sure knows how to pull off a secular ritual!” (Consuming a dangerous/tainted substance without fear before a crowd to prove the power of a new paradigm. Come on, that’s basically a secular re imagining of the acts of some Christian saints. It’s really stunning from a metaphorical/theatrical l viewpoint)

    *Edit for typing error

  28. Kaj Sotala says:

    Re: the car survey, I wonder what the responses would have been to “in the case of a sudden crisis, would you prefer a self-driving car to concentrate on saving its driver’s life even if this caused it to collide with the car you were in, or to minimize the risk to the other traffic?”

    Framing makes a big difference. Once you change the wording so that the person being polled is the one who gets a personal benefit from the “maximize lives saved” policy, I would assume the answers to change a lot. (Ideally this should have been asked in a more impersonal way from the start, the current framing makes me think that the results are entirely worthless. People will endorse different rules for themselves than for people in general.)

    • Kibber says:

      > People will endorse different rules for themselves than for people in general

      Frankly, I thought that was kind of the point of the study.

  29. Anonymous says:

    I AM LEGITIMATELY ENJOYING NANOCYBORG UBERHOLOCAUST (what is wrong with me? I honestly do not think I should be liking this…)

    It honestly seems like pretty decent background/white noise type stuff to listen to when focusing on other things, in much the same way I occasionally use a rain-sound generator.

    I am not sure I would classify it as particularly metal. And it is only the second weirdest metal I have ever seen, if you class it as that, the weirdest being
    Heavisaurus, the Finnish children’s dinosaur-themed heavy metal band (example: https://www.youtube.com/watch?v=cjJNPSdvjp0)

  30. Ken Arromdee says:

    The homeopathic HIV virus is nonsense.

    1) Homeopaths claim that a tiny dosage of X cures X, not causes X. Failing to get sick would not prove that homeopathy doesn’t work–homeopaths *expect* that you wouldn’t get sick from it.

    2) Because the probability of getting HIV from a properly diluted homeopathic HIV concoction is so miniscule, the danger of the process is dominated by the danger of some kind of human error in diluting and administering the blood. This danger cannot be reduced below a certain level and permitting it may be unethical all by itself.

  31. Michael Edward Vassar says:

    The people making the decisions in the companies are narcissists and only respect other narcissists. Don’t want to trust non-narcissists with power.

    • This is in regards to the interviews link?

      That’s hard to distinguish from narcissists being good at influencing normal people.

    • Ialdabaoth says:

      I understand why non-narcissists would only want to trust non-narcissists with power, but why would narcissists want to only trust other narcissists? It seems like narcissism gets fed best when surrounded by non-narcissists.

      • When you put it that way, it does seem odd, but here’s an extended discussion of sociopaths preferring to promote other sociopaths, and commenters saying “now I understand my career”.

        • Ialdabaoth says:

          Heh. That pyramid keeps coming up more and more. (I had a near-identical observation when I was growing up, except I called the “clueless” “tools”. I still think “tools” is a better term for them.)

    • Scott Alexander says:

      I don’t actually think there’s inter-narcissist unity in the manner you seem to suggest. In fact, I would expect narcissists like other narcissists less, on the grounds that “she wants to talk about herself all the time, instead of talking about me!”.

      • If the Gervaise Principle is correct, the situation may be that narcissists (or at least sociopaths) fear and despise conscientious people as a bunch of tiresome moralizers and possibly as loose cannons.

        I have a friend who believes he lost a job because he wouldn’t pad his expense account, so the other employees didn’t trust him.

        • Scott Alexander says:

          Narcissism and sociopathy are different in ways that are important here.

        • Ialdabaoth says:

          I have a friend who believes he lost a job because he wouldn’t pad his expense account, so the other employees didn’t trust him.

          {look of painful solidarity}

        • dublin says:

          And they’re right to do so.

  32. roystgnr says:

    Everyone’s car will *not* have the same programming, because at least at first nearly everyone’s car won’t be driven by a computer program at all, it will still be driven by a person. “Should my car prioritize my life when it and an identical car are endangered” is a very different question than “should my car prioritize my life when some jerk endangers it”. Superrationality only works when you know the other players are superrational too.

    The remaining human drivers have another consequence too: if it turns out that robot drivers are become safer than people (a priori likely in the long run, and there’s already a little bit of supporting evidence for Google’s cars), then at first the best way to reduce the total number of accidents will be to increase the rate of adoption of the new safer vehicles, even if that means giving in to human selfishness and limiting the total safety of each vehicle.

  33. At this point I think we just need to admit that everything about sex in medieval times was horrible and oppressive and terrifying no matter what gender you were.

    Yet more slander against the medievals by referencing events that didn’t take place in the Middle Ages.

    • Deiseach says:

      Beat me to it: the 16th century is not the Middle Ages, it’s Early Modern Period. And anyway, I like the 13th century, so using “mediaeval” as an insult doesn’t work to make me think “Ah yes; this is indeed a Really Terrible Thing that could only happen in those Really Terrible Times, unlike the homelife of our own dear queen”.

      Also, I stopped reading that article the moment I hit the bits about “innocent men” and “greedy, bored women”. I’m a “just the facts, ma’am” kinda gal – don’t try and make my mind up for me before you’ve demonstrated that yes indeed, all the men were innocent and all the women were greedy bored bitches screwing over the poor nice guys for no reason other than they wanted to move on to a new sugar daddy/boytoy.

      • JTHM says:

        Perhaps my values are just too new-fashioned, but I’m pretty sure that impotence is in no way a moral fault, and thus that no guilt is possible.

        So yeah, they *were* all innocent, functional genitalia or no.

  34. pwyll says:

    Re using test scores instead of interviews to hire employees: yes, it would result in a more accurate hiring process. But no, employers aren’t going to switch to that… because the supreme court has already decided that’s raciss. Thus the emphasis that Microsoft and Google put (or used to put) on “puzzles” – basically IQ tests in disguise.

    It’s a typical perversity in the discourse that companies are legally barred from using the most effective methods available to screen employees… and then are castigated for not using said methods.

    • pwyll says:

      Perhaps we should just fill job openings by lottery, since certain fire departments were forced to do so when people complained that the tests were racist?

    • ACS says:

      Just to clarify: I work at Google. Google has never relied on puzzles, and in fact relying on puzzles is banned because it empirically produces garbage results.* I have no idea how that got started and no idea why people keep repeating it.

      • Douglas Knight says:

        Did you leave off the justification for the empirical claim? The obvious guess is that google used to use puzzles. One place it may have started it he billboard recruitment puzzles.

      • pwyll says:

        Google has never relied on puzzles … I have no idea how that got started and no idea why people keep repeating it.

        People keep repeating it because it used to be true. (Thus the “used to” in my post.) I’ve heard this directly, in anecdotes from friends, and this article has more examples. It’s true that Google has in more recent years moved away from general puzzles, towards more domain-specific puzzles… If they can use educational credentials to restrict the range of their applicants, then it’s more probable that generic brainteasers may lose their power. (Here’s a demo of the problem of restriction of range.)

    • Multiheaded says:

      I get the feeling that many left-libertarians/market anarchists would tolerate racial discrimination if reactionaries would in turn support workers’ ownership of the means of production… (Ha ha.)

      • Ialdabaoth says:

        I would love to see that world, but only after a VERY clear understanding of the difference between ‘consumer goods’ and ‘means of production’.

      • Alrenous says:

        It is not illegal to start a co-op. Are you it expecting it to outcompete hierarchical firms by hiring superior workers, by superior worker morale, by efficiently routing around executive meddling? If so, why don’t you start one?

        It is illegal to assert your right of free association, regardless of the justification.

        • Multiheaded says:

          I’d expect a co-op to “outcompete” hierarchical firms by having the Visible Foot of Government on its side… to make an individual’s share less alienable in the face of concentrated capital’s bargaining power, have global treaties against the threat of capital flight, give workers an option like UBI so they wouldn’t fear telling an efficient sociopath boss to fuck off, provide an attractive model of workplace democracy in state-owned enterprises which would compete with co-ops for labour –

          – exactly like I wouldn’t need military training to defy a bunch of Mad Max raiders as long as I have law enforcement and effective gun control and a peace-loving society that gives a fuck on my side. I’m (basically) not an anarchist, but even anarchists must believe in a monopoly on force on some level (outside of historical or fictional scenarios with endless “free” land full of “free” resources), since it’s the only thing that would keep a state of anarchy from being overrun by a primitive order. They just talk about it less, and in a different way. Why should it be any different with capital?

          It is illegal to assert your right of free association, regardless of the justification.

          Because there is direct and glaring historical evidence of the externalities.

        • Oligopsony says:

          There’s evidence (beyond just the microeconomic fundamentals that would suggest it) that worker-managed firms actually are more productive; I can go digging if your skeptical. Strategic disadvantages include the things mentioned by Multi, as well as high risk aversion, which a UBI (which if we’re in Mutualistan could be nicely implemented as a firm capitalization differential) would also help with.

          And yeah, I wouldn’t consider racial discrimination a real social problem if some groups didn’t have access to more structural power than others (this is more or less what “you can’t be racist against white people” means.) Aesthetically distasteful, sure, but so is picking your nose.

        • Multiheaded says:

          And yeah, I wouldn’t consider racial discrimination a real social problem if some groups didn’t have access to more structural power than others (this is more or less what “you can’t be racist against white people” means.) Aesthetically distasteful, sure, but so is picking your nose.

          Hell, what if in a Weird Eutopian Communism it could be a legitimate largely-accepted subculture, like kink is in SJ spaces today? You can already have raceplay within a non-sexualized BDSM context; what if it was also non-eroticized? Of course, people like me might reflexively react like a thoughtful 14th c. Catholic would react to today’s Tumblr-brand polyamory… but I would at the same time feel its consistency with my understanding of emergent objective morality… which leads to the question: could one explain poly to a reanimated Meister Eckhart?

          I can go digging if your skeptical

          Please do anyway, if you could spare the effort; I’d be interested too!

        • Oligopsony says:

          Start with this meta-analysis or this obviously biased but comprehensive and available report. There’s a bunch more just by going through Google scholar with an obvious search like {cooperatives productivity} and with many studies usefully breaking things down by different kinds of employee participation and/or incentives – the evidence for full employee ownership is more ambiguous than for mixed models. I have uni library access so let me know if you want a pdf of anything in particular.

        • Alexander Stanislaw says:

          And yeah, I wouldn’t consider racial discrimination a real social problem if some groups didn’t have access to more structural power than others (this is more or less what “you can’t be racist against white people” means.)

          Really? It seems undesirable to me to have a country that has two racial groups of equal structural power that discriminate against each other. Cooperation would be preferable, but when one side cooperates and the other side defects that is the worst of all regardless of whether the defector has less structural power.

          Also I find the term “structural power” disturbing for two reasons. First of all power is multidimensional not linear (similar privilege is not a scalar quantity – two people can be privileged over along different axes). And secondly it allows you to divide groups of people in an unnatural way. As Ialdabaoth illustrated here. By lumping low status men/women and high status men/women together, the woman/man is able to justify the harm they inflict on the low status man/woman.

        • Oligopsony says:

          I agree with you on all substantive points here.

        • Alrenous says:

          Multiheaded,

          Just looks to me like you have a government fetish. So the advantage of co-ops is
          purely psychological, for the benefit of the worker. If this is true, you should have a severe propaganda advantage. +5 to rhetoric. As workers defect from status-quo employment they way the (Comanche?) captives defected to the native lifestyle, it will drive up the price of labour until what-Multiheaded-calls-capitalism locally collapses.

          No, anarchists don’t believe in the monopoly of force. Your statement just looks like a puzzling non-sequitur to me. Certainly there’s a theoretical issue with border-level security, but unknown unknowns hardly constitute hostile known knowns. What keeps an anarcho-capitalism from being overrun by a primitive order is the several police firms which provide security insurance.

          That said your co-op friendly government has the exact same problem. If co-ops do not outcompete hierarchy without a hand on the scale, the co-operative government (nice halo effect you have there) will be conquered in short order by a capitalist one. Either that or you’ve solved the anarcho-capitalist border-security problem.

        • Erik says:

          It is illegal to assert your right of free association, regardless of the justification.

          Because there is direct and glaring historical evidence of the externalities.

          Expand on this a bit, please? Because if you’re referring to not getting hired as an externality, I think you’re abusing the term, and if you’re referring to some other externality, I’m not seeing it.

    • Oligopsony says:

      that’s raciss

      my favorite racist thing by far is dismissing concerns about racism, and then phrasing it in dialect

      • Anthony says:

        So is that high-status or low-status mockery?

        • Oligopsony says:

          Reactionaries are mocking high-status liberals by calling them low-status blacks, then I’m mocking them in a way that I hope will be status-lowering, and then your post may or may not be doing the same to me, and then this post here is probably doing something it that area, maybe even to itself, I’m not entirely sure.

          (I also think the concept of “status” as used by LW types is probably really overextended, but that’s a story for another day.)

      • Nornagest says:

        The first thing that came to mind when I read that wasn’t dialect but rather “cis” in the sense of “not trans”, which seems to be on its way to becoming a term of abuse in certain circles. That probably says more about the amount of Tumblr I read than about pwyll’s intentions, though.

      • pwyll says:

        It’s not so much dismissing concerns about racism… The supreme court was right, using aptitude tests for employment is racist. The problem is that the very concept of “racism” is an ugly, blunt, instrument that was used to ban one of the most useful tools an employer has for the very difficult problem of hiring the best employees. When Neoreactionaries spell it “raciss”, one connotation is to remind the reader that “racist” is more a term of abuse than an argument.

    • Douglas Knight says:

      Griggs isn’t the reason that employers use degrees instead of IQ tests because Griggs said exactly the same about degrees.

      Why are employers afraid of IQ tests? I think that is a great mystery. Probably they just do the same thing everyone else does and are afraid of making waves by using tests. But some of them ask lawyers and the lawyers quote Griggs. This is nonsense on two grounds: the one I stated and because Griggs is superseded by a law that says the same thing. Are the lawyers reading justice department tea leaves, but refusing to say it, or are they all making the same error?

      • Armstrong For President 2020 says:

        The law isn’t composed of words on paper; the Anglosphere tradition is that case law and common practice tend to carry the day. The EEOC is more than willing to use demographic composition itself as evidence of discrimination, judges really don’t like to hear arguments about group differences, and even if you do win your case on the merits the process itself is onerous and exposes you to political attack.

        Anyway, if you prefer to read about this sort of stuff in the original legalese this dude seems to have written a bit about it in the ‘pre-employment testing’ tag. The comments section seems pretty calm and intelligent too from the one or two posts I’ve read.

        • Douglas Knight says:

          Yes, the right thing is to predict the behavior of the EEOC and see that they prosecute disparate impact cause by tests and only rarely disparate impact caused by use of degrees. But no one does that. While your link is a good exception, they usually invoke Griggs, which is the wrong answer from all angles. Indeed, the first comment in your link invokes Griggs to defend a practice (“high school or equivalent”) that has nothing to do with Griggs without giving a second thought to what EEOC actually does.

          But once you are thinking about the EEOC, you should think about what it does, not what it says it does. Not everyone who uses a test gets sued.

          I didn’t know that the EEOC prosecuted strength tests.

  35. Anonymous says:

    The papers concerning barriers to employment (like showing that black-sounding names were discriminated against compared to white-sounding names on identical resumes) also claimed that the effects disappeared after personal contact has occurred. Interviews are good/bad/indifferent depending entirely upon what conclusion you’ve already decided on.

  36. Leonard says:

    presumably everyone’s car will have the same programming

    I would not presume that. I mean, isn’t the whole premise of the question that there is a choice? I.e. that two kinds of cars exist? Why would someone ask me “would you prefer all cars to be like A or like B?”, in the backasswards form of “would you prefer your car to be like A or like B?”

    Also, the idea that your own buying choice changes the reality around you is drastically counterfactual. I.e., if I buy a car right now with antilock brakes, this does not mean that suddenly all other cars have antilock brakes. So, even in a future hypothetical, I don’t assume that my buying a car means that suddenly everyone else’s cars get my software.

    • Alrenous says:

      Isn’t this whole thing an already-solved problem?

      Your insurance company asks what kind of self-driving software your car has. It bills accordingly. It will have or rapidly gain the actuarial data to discriminate between startlingly fine distinctions. E.g. it will have a nice table of prices for no-fault insurance based on software.

      That is, in the unlikely event that the bureaucracy doesn’t meddle. For example, certain programs won’t be able to get insurance, because they tend to get you charged with criminal negligence – there’s no real immediate need to regulate, driving without insurance is already illegal.

  37. Leonard says:

    allowing something to be mocked is a form of countersignaling proving high levels of respect for that thing.

    That’s gay.

    By which I mean, of course, that homosexuality is highly respected in our society, and it was even more highly respected twenty years ago.

    Seriously, allowing something to be mocked may be a form of countersigning. But actually mocking something (and getting away with it) shows superior status. Just because you’re allowed to try does not mean you succeed.

    • ozymandias says:

      Already covered in the Hanson post: low-status and high-status people can get mocked, medium-status don’t, for the same reason that rich people and poor people dress like poor people.

      • Eric Rall says:

        “An American is a person who isn’t afraid to criticize the President but is always polite to traffic cops.”

  38. While I’m all in favor of Cats and Dogs Snuggling, Nader and Norquist seem to be making the argument that we should increase the minimum wage so that we can stop spending money on social welfare programs. I’m going to guess that you aren’t in favor of this.

  39. Shmi Nux says:

    > allowing something to be mocked is a form of countersignaling proving high levels of respect for that thing.

    Yeah? Try mocking Jews (being non-Jewish), or worse, the Holocaust, and see what happens.

  40. Alrenous says:

    The absolute worst case for Cambodia is: Pol Pot killed a full third and killed smartest first. This would depress mean IQ by about ten points. As per Flynn in your edit, these assumptions are wildly unrealistic.

    Re: Robin Hanson

    One easy test: in our society, dominant groups are ones that people are allowed to insult and lampoon. For example, you can get in a lot more trouble for making negative general statements about blacks or women than about whites or men.

    Circular reasoning. Why do I have to point out circular reasoning?

    Please note that disproving the support does not support the opposition.

    You have to be careful about asserting or believing the counter-intuitive, because scholars get no status points for believing the obvious.

    Thus the circle is complete: both sides of the debate have a study to cite and can accuse the other of being “science denialists”.

    I’ve reviewed the literature enough that I can easily explain how both are true. Indeed I think I’ve seen that exact study before from GirlWritesWhat.

    In most individual measures of personality, women overlap men by about 65%. If, however, you multiply these odds together across a mere five personality items, you get an overlap of approximately 8%. Lots of women are roughly as dominant as men, but to find a woman that’s as dominant as a man and as cold and as insensitive as a man is very rare.

    In personal conversation, total personality matters most, which is why it’s so easy to tell apart most men and women you’ve met. (Ref: department of the obvious.)

    • ozymandias says:

      I find your last paragraph contrary to my experience. Online, I fairly regularly am blindsided by people’s genders. (Offline, of course, there are physical cues and so confirmation bias sets in.) It is possible this is because pretty much every place I’ve participated online is heavily gendered– Less Wrong and environs are ~80% male, slash fandom is essentially devoid of cis men, feminist communities I have been active in have had about twice as many cis women as cis men– and thus attract outliers (i.e. female LW participants are very masculine, male slash fangirls are very feminine).

  41. Vaniver says:

    I hate this sort of thing, because it’s clear that I would need to study large amounts of very complicated math and develop strong opinions on statistical minutiae before I can have a reasoned opinion on this subject, plus after I do that whichever side I disagree with will still tell me I’m wrong and dismiss the other side’s paper based on the first nitpick they can find.

    Don’t worry, it’s pretty simple: univariate bad, multivariate good.

    Suppose there’s a river that runs northwest-southeast through a city. Men live on one side of the river, women live on the other side of the river. But you don’t know about the river- you just have latitude and longitude for where women and men live.

    It looks like the male mean is north of the female mean- but there’s a bunch of overlap. It looks like the male mean is east of the female mean- but there’s a bunch of overlap. But when you consider the two together, and look for the best dividing line between the two, you discover that if you run the line northwest-southeast you get perfect separation between men and women.

    The paper noticing almost no overlap is one that relies on multivariate analysis.

    • Ialdabaoth says:

      So framing this more narratively:

      If you drew out a nice multidimensional graph with various personality traits as the axes and data points as individual personalities, you would discover various close clusters of “personality archetypes” which tend to be distinct to a particular gender. While those archetypes themselves may be distributed rather haphazardly throughout the space, if you find a point near one of the particular archetypes’ epicenters, you can make a pretty reliable guess about the gender of the person represented by that point.

      Is that about the gist of it?

    • dublin says:

      This is more or less the same bullshit logic people were using to say that race didn’t exist in genetic terms, right?

    • Scott Alexander says:

      So basically you’re saying this new paper is correcting a Lewontin’s fallacy?

      That’s neat, but I worry that both approaches have some merit. For example, if there are 100 traits on which men and women are very very slightly different, that could produce a very large multivariate difference, like say “less than 10% overlap between men and women”, and that would sound very important, but it would still be the case that in real life, the differences between men and women might explain nothing because they’re very very similar on each specific trait.

      Or even worse, what if men are very different on one trait everyone agrees they’re very different on (for example, aggression)? Then it might be that the multivariate difference is driven almost entirely by aggression, and looks very large, but it gets interpreted to mean that men and women are different in general.

      Also, since your river analogy was good, do you mind explaining what they mean about “latent variables”?

      • anon2 says:

        Without background on what you might call complicated mathematics it’s tough to really get the point across briefly but a latent variable is something that is not directly measured.

        Imagine you wanted to know the ability of people to shoot three point shots in basketball. But, you only measured subjects shooting free throws. You do have other good evidence to think these two things are strongly correlated. The latent variable is three-point shooting while free-throw shooting was measured.

        Also, the paper you linked is a couple years old and there have been responses since then, including from the original authors it seems, might be worth a look.

        http://www.epjournal.net/wp-content/uploads/EP1110671076.pdf

        http://www.epjournal.net/wp-content/uploads/EP1110671076.pdf

      • Nathanael says:

        For starters, Lewontin was correct and it isn’t a fallacy.

        Statistics is hard sometimes. The Wikipedia article titled “Lewontin’s fallacy” has some good discussion of why Lewontin is largely correct.

        When trying to analyze *gender* issues the same is true.

        But there’s something much worse going on: why would you even want to try to guess people’s sex by psychological methods? The excuse with trying to figure out “geographic ancestry” is that it may help predict medical problems. But for anything like that (medical) where you really need to know someone’s sex, there are much more obvious methods which are very reliable, such as DNA testing or looking at gross anatomy.

        And these studies aren’t good for anything else (other than “try to guess sex using psych test” games), for exactly the reasons Lewontin explained.

        I am suspicious of “pure research” as an excuse; all research is motivated, and the motivations of research like this are blatantly suspicious.

  42. I feel like once we reach the point where corporations are going behind government’s back to try and come up with novel lucrative climate change solutions, the fight against climate change is in damn good shape.

    Sorry, too late. A rather bad case of climate change is already guaranteed to happen, and history will eventually conclude that it couldn’t have been any other way in a world that was too democratic (i.e. we had a system of governance that allowed ordinary humans’ strong preference for short-term partying to be reflected in the policies chosen).

    • drethelin says:

      considering the history of earth, rather bad climate change was guaranteed regardless of what we did.

      • Yes, but significantly later, and by then we would have developed the technological ability to fully control the planet’s climate according to our will (or alternatively would have killed ourselves with advanced technology on the way there).

        But I should add the caveat to what I said that it’s not *completely* impossible that we’ll develop super-powerful technology in time anyway, while also managing to not kill ourselves with it first. But both of these tasks would have been easier if we hadn’t caused severe climate change and associated large-scale political disruptions to start getting underway already.

    • Paul Torek says:

      We probably are doomed to significant warming, but the paper wasn’t very convincing. “They also factored in the ability to govern each, the ethics and public acceptance,” says the description. Public acceptance? Nobody asked the public to accept CO2 increases; those were simply inflicted. If a geoengineering scheme is ruled out because of public disagreement, the deck is being stacked.

  43. Douglas Knight says:

    I recommend the original Gender Similiarties paper. I find the text unreadable, but the compilation of studies is interesting, particularly for its diversity. The largest sex differences are: throwing 2 standard deviations, masturbation .96, tendermindedness 0.91, attitude to casual sex .81, helping while aware of observers .74, mental rotation .7, grip .66, sprint .63, body esteem .58, physical aggression .54, assertiveness .51, activity level .49, spelling .45, computer self-efficacy .41, language .40, spacial perception .44, smiling while aware of observers .46, but you should also read all the items that are similar.

    • Nornagest says:

      Throwing. Huh.

      I wonder if the difference is anywhere close to that big in trained athletes?

    • Nathanael says:

      The gender similarities paper is excellent.

      Even most of the identifiable differences appear to be socially constructed — certainly body esteem (this is wildly culturally dependent), and also physical aggression. (This is documented in experiments, but also in real life. The number of delinquent violent girls has been increasing for several decades now. We’ve needed more jail cells for women and fewer for men.)

      Some of the other differences are just really weak correlations, or they’re from methodologically questionable studies — there may be a difference there, or there may not.

      A few of the differences may actually be solid and durable results which seem to hold even after you control for culture, education, and upbringing as much as you can. Masturbation rates seem pretty likely here. (There are some good theories on this, including that it’s harder for women to figure out *how* to masturbate — who knows, maybe this’ll change over time too.)

      But it’s a very short list. Even in the case of mental rotation, where the correlations are pretty strong, individual differences dominate over gender differences, and it’s still not clear that there’s anything but social effects here.

      The only two things studied in that paper with correlations big enough to be worth paying much attention to (d of more than 1.00) are actually *physical* differences (throw velocity, throw distance).

      The similarities, by contrast, are pretty much *all* solid (culturally independent) results and pretty well proven.

  44. Pingback: June 19th 2014 | Wot I Dun

  45. Shenpen says:

    I have never seen a company who would not have job interviews – how do you even decide without an interview if a candidate has the right kind of personality, charisma looks and hygiene to charm customers and to motivate coworkers, or at least be able to work with human beings?

    • EoT says:

      I tend to agree. I would never hire someone for a public-facing job without interviewing them first.

      The big problem here is that pretty much everyone manages to look decent, dress well, and be clean for a job interview. It’s not too much harder to come across as affable and cheery for the short time a typical job interview takes.

      There should be certification programs in things like hygiene, friendliness, manners etc. I imagine it would involve people secretly surveilling you at random times for some long period of time. If you can fake being presentable and friendly for, say, six months, you can probably fake it for however your job lasts.

  46. Paul Torek says:

    “you can’t actually get HIV by swallowing it.”

    Unless maybe you have, I dunno, bleeding gums.

  47. pwyll says:

    Regarding Noah Smith vs. Zero Hedge on Japan’s economy: I’d say Spandrell gets in the last word: http://darkmatterjournal.com/DarkMatterJournalTexts/SpandrellBabies.pdf

    • Nathanael says:

      High birthrate promotion is one of the dumbest policies in the world, so I really don’t respect that piece, even though he’s fairly reasonable apart from that.

      Honestly, we don’t need more people, we’re over carrying capacity on the Earth as it is. If an individual country is worried about having too few young people, fergodsake just allow immigration and deal with it.

      • Ialdabaoth says:

        we’re over carrying capacity on the Earth as it is.

        Absolutely untrue; the Earth can easily sustain upwards of 20 billion people. Just not with our current practices.

      • Armstrong For President 2020 says:

        Yeah, who needs birth rates above replacement when you can just import a new population from abroad!

        After all, it’s not like immigrants have significantly different intellectual and physical abilities which might make it difficult for them to perform the kinds of highly skilled work a first world society needs to survive… oh wait.

        Well at least since they have the same values and cultural assumptions they can be easily integrated into a society without destroying it’s sense of trust… oops, nope that doesn’t work either.

        At the very least we can all agree that there is absolutely no risk of bringing disease or crime from their home countries along with them… damn, I really thought I had a winner there.

        Remind me again, why should we be replacing our native populations with immigrants rather than just doing what human beings are literally made to do?

    • Hainish says:

      That article: So many dog whistles, so little time.

  48. It wouldn’t surprise me a bit if smart people have average depression rates– it’s just that smart people talk about their depression more– and more vividly– than everyone else does.