Links For July 2014

The Koran talks about a mysterious immortal figure named Khidr who travels the Middle East and hangs out with prophets. And the passages about him seem to have inspired some of the Jewish folk tales about Elijah I heard growing up.

Scientists find the gene that causes an entire family to be morbidly obese. Uninteresting in that most people probably don’t have it. Very interesting in that it’s yet more proof that obesity can have genetic causes.

This is more levels of hype inversion than I like in my stories. Scientists Prove God Exists says an ABC article, which then goes on to say that ha ha, of course scientists didn’t prove God exists, they were just making a joke for a snazzy headline. All that really happened was scientists proved Godel’s ontological proof of God’s existence was correct. But, uh, if a proof of God’s existence is correct, that should mean God exists. I feel like the article somewhat overlooks this important point.

A while back we discussed ability of wealth or poverty to continue over generations, with some interesting papers on slavery as examples. I recently found another one that agrees that past levels of slavery are not related to lower incomes, but are related to greater income inequality, presumably through decreased education of black people. There’s a lot of stuff I don’t get here – instead of measuring income inequality and assuming it was racial, why didn’t they measure income of blacks directly? Also, how does this square with our last paper that found that descendants of enslaved and free blacks had equal outcomes within two generations of emancipation?

Less Research Is Needed: an article on how much the author hates the phrase “more research is needed” and how in some cases it can be used to make debate interminable so that the “wrong” side of a controversial question can never be proved.

Peter And Jane Go To The Art Museum

Reddit has a really good post by a Chinese citizen about their perspective on the Tiananmen Square incident. It would have been worth it if all I’d learned was the phrase “the events of May 35th”, which is how they get past censors screening for the date “June 4th”. But instead, I get a really complicated picture of the forces at play which almost make me feel sympathetic for the Chinese government. See also the post just underneath on the revival of meritocracy and Confucianism in China (possibly exaggerated).

I know nothing about this and it is probably bunk, but with that disclaimer: Fluid Tests Hint At A Concrete Quantum Reality. The ripples of certain kinds of oil droplets precisely reproduce a lot of the weirdest features of quantum mechanics on a macro scale. The explanation isn’t anything weird about probabilities, just some unusual interactions between the droplets and its own waves. If particles produce waves in space-time with the same kind of properties, that would go a long way to explaining the quantum world.

More on the debate about whether marijuana use causes schizophrenia: schizophrenia and cannabis use seem to share common genes.

Chimpanzees don’t like Western music, but do like music from Africa and India.

Solve all tornados by building a 1,000 foot high wall across the Midwest. As a bonus, keep out White Walkers. However, I for one am not anxious to trust our country’s safety to anyone with the photoshop skills displayed by the demonstration picture. WHAT ARE THOSE BOATS EVEN DOING?

United States renaming street with Chinese embassy after imprisoned Chinese dissident. Sounds sort of like something a four year old would do. Reddit suggests China rename street with our embassy to “Edward Snowden Lane”.

Neo-Nazi hipsters considered more hateable than regular neo-Nazis or regular hipsters. I feel bad about sharing this article, because it’s clearly one of those “Look at the people who are different than you! Mock them!” type pieces. But to be fair, these people are pretty mockable. And I was tickled by the sentence “In February, Tim and Kevin started Balaclava Kueche, Germany’s first Nazi vegan cooking show.”

I am enjoying Fake Liberal News Site Twitter. The big question is which is more on target – @vauxnews (“The president’s plan to circumvent today’s Supreme Court decision is not just legal, it’s brilliant. And he’s handsome. So, so handsome”) or @salondotcom (“Could this Baptist YouTuber that freaked out over “Ancient Aliens” be the new face of the religious right?”)?

Speaking of Vox, here’s there article on how the New York Times predicted the assassination of Archduke Franz Ferdinand would be good for European peace.

From the Department Of What Now, Motherf@#&kers? : sex differences in mental rotation, a skill generally associated with mathematics ability, are greater in nations with greater gender equality. Offered explanations aren’t bad, but poor nonrandom sampling limits ability to draw many conclusions.

ISIS’ Plan For Global Domination (supposedly). Is it wrong to want the terrorists to win so we can have a country called “Qoqzaz”? Also, I imagine two ISIS members daring each other to try to draw Khurasan bigger and bigger, then laughing and keeping it on the map because they’re not going to achieve global domination anyway.

I was linked to this interesting but hard to believe paper on how the requirement that psychiatrists report homicide threats to the police significantly increased homicide rates, presumably because homicidal people were less willing to talk the problem out with their psychiatrist. I’m doubtful for many reasons – what percent of murderers see psychiatrists, what percent of them would bring up their homicidality even without the ruling, what percent of psychiatrists would be able to treat them effectively, and what percent of homicidal people even know what the laws on mandatory reporting are?

I’ve had some patients ask me the best way to disguise their scars. This is definitely the best way.

Practice Makes One-Third Perfect, The Other Two-Thirds Is Talent. Interesting example of scientific failure here: people found that people who were good at things had practiced longer than people who weren’t and so assumed that lots of practice (rather than talent) led to success. More sophisticated investigation suggests that talent leads to minor success which leads to motivation to continue practicing which leads to lots of practice which leads to success.

We know the recession is officially over because Dubai has started building crazy huge useless buildings again.

Not only is truth stranger than fiction, it has better monsters. Here’s the Black Swallower. Make sure to read the section that tells how known specimens came to be collected.

Sky Kingdom is a Malaysian cult which is best known for having a giant two-story teapot in the middle of their compound.

Telegony is the ancient and medieval idea that a woman’s children could inherit characteristics not only from their father, but from all the woman’s previous sexual partners. It was seriously defended right up until the real mechanisms of genetics were pinned down in the late 19th and early 20th centuries. I wonder how much influence that had on ideas of sexual purity.

From there I did some Wikipedia link-clicking to learn that the Telegony is also the name of a sequel to the Odyssey, and that in fact there is a whole Epic Cycle of which the Odyssey and Iliad are only a part. And it ends with all of Odysseus’ sons hooking up with all of Odysseus’ sexual partners, which I guess isn’t especially weird for a Greek myth.

The latest development in the brave new post-Bitcoin world is crypto-equity. At this point I’ve gone from wanting to praise these inventors as bold libertarian heroes to wanting to drag them in front of a blackboard and making them write a hundred times “I WILL NOT CALL UP THAT WHICH I CANNOT PUT DOWN”

Ozy and I are staying in the Mount Washington Hotel in the White Mountains right now, so here’s a New Hampshire hotel story for you. When the Supreme Court ruled in favor of eminent domain for creating useful public buildings like hotels, the residents of a town in New Hampshire where a Supreme Court justice owned property organized a movement to seize his house and turn it into a hotel called The Lost Liberty Inn.

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

116 Responses to Links For July 2014

  1. Chris Billington says:

    A droplet bouncing on the surface of a liquid has been found to exhibit many quantum-like properties, including double-slit interference, tunneling and energy quantization.

    I was about to flaunt my credentials as a quantum mechanic and express my disdain for this popular science article presenting these things as if they were even remotely specific to quantum mechanics (they’re not – water waves have all of the above properties too and it has nothing to do with quantum mechanics. They are simply properties of waves, of which quantum particles are but one example).

    But then I read a bit further and…yeah…wow. A classical analogue to pilot wave theories. That is really cool, and definitely makes me update on the plausibility of Bohmian mechanics.

    Oh wait, I just remembered nonlocality. The oil drop system is obviously local, and so it can’t possibly model the nonlocal aspects of Bohmian mechanics. The pilot wave in Bohmian mechanics need not be any more nonlocal than normal quantum mechanics though, if I understand correctly: being governed by the Schrödinger equation, it can develop entanglement the same as wavefunctions in normal quantum mechanics. But you have to make the particles’ motion nonlocal too, having it depend on what the pilot wave is doing at other locations in space — something which the oil drop is clearly not doing. I suppose the oil drop analog is only for the simplified case where there is no entanglement (you can’t entangle water waves with each other, they’re just classical waves), so they don’t come up against this.

    But since entanglement, not wave behaviour, is really the thing that makes quantum mechanics different from classical mechanics, my mind is relatively unblown over this.

    • whales says:

      Why would you have updated on that? What the equations say is uncontroversial. It’s like Ringbauer et al.’s simulation of closed timelike curves from a little while ago, or the Solano group simulating the complex conjugation of a wavefunction. As cool as those experiments are, finding a mapping from a real system onto those equations doesn’t really say anything new about the nature of quantum mechanics.

      I do like that it gives a nice way to visualize (an aspect of) Bohmian mechanics. People seem to have more trouble with it than with the others.

      • Chris Billington says:

        Nature seems to like re-using ideas all over the place. If someone shows a classical system that has a classical particle interacting with a classical wave in just the right way, it makes it a lot more plausible that that interaction would occur in another system.

        Also their analog system isn’t that complex. Nature seems to like simple things. The more complex their classical analog, the less plausible it becomes that it might show up elsewhere. Showing that this behaviour can be exhibited by a really simple system makes my prior higher for it occurring at the fundamental level. I guess one could say something about Solomonoff priors here.

        (To contrast, as it happens, I have a classical analog to Bohmian mechanics right in front of me (my laptop), but it’s so complex that there’s a low prior for nature having an analog to general purpose CPUs simulating Bohmian mechanics in its fundamental laws. Something akin to water drops bouncing off waves is much more plausible)

        So basically it’s that they’ve shown the correct particle-pilot wave interaction (except without nonlocality) can happen as a consequence of a really simple system, as opposed to a complex one, and as opposed to just asserting the interaction in Bohmian mechanics without deriving it from something more fundamental, both of which have lower priors.

        • whales says:

          Fair enough. I guess I was thinking that “particle guided by a wave” was already fairly natural and explicit in the guiding equation, and wouldn’t have judged this analogue any more fundamental than what’s suggested by various derivations of the guiding equation.

    • Chris Billington says:

      Also, they say on the topic of entanglement:

      But in the pilot-wave version of events, an interaction between two particles in a superfluid universe sets them on paths that stay correlated forever because the interaction permanently affects the contours of the superfluid. “As the particles move along, they feel the wave field generated by them in the past and all other particles in the past,” Bush explained. In other words, the ubiquity of the pilot wave “provides a mechanism for accounting for these nonlocal correlations.” Yet an experimental test of droplet entanglement remains a distant goal.

      as if they can model entanglement without nonlocality. No. They will not be able to do this. Entanglement in quantum mechanics isn’t just a correlation between two particles which you don’t know about yet — the nature of the correlations depends on what type of measurement you perform later. You cannot reproduce quantum statistics with hidden variables unless you have nonlocality. It’s a theorem. They will only be able to simulate this in an analog if the measurement basis is known in advance, which is totally cheating and not really demonstrating entanglement at all.

    • Will says:

      There is (limited) non-locality in their system. The pilot wave moves at the speed of sound in the liquid, the droplet moves substantially slower. Its like quantum mechanics \phi(x)\phi(y) would be non-zero even outside the sound cone for arbitrary observables, but the commutator would be 0. The pilot wave is a non-local hidden variable in Bohm’s theory (its how Bohm’s theory works)

      Its unclear to me how/if collapse would occur, but I never understood multi-particle Bohmian mechanics anyway. I wouldn’t update towards Bohm- it doesn’t play well at all with quantum field theory.

  2. Protagoras says:

    I located the paper on Goedel’s ontological proof, and it looks like the authors are claiming to have shown that Goedel’s proof was valid, not that it was correct. That is not at all surprising; I’d be pretty shocked to discover that Goedel proposed an invalid proof myself. But it is unclear whether all the axioms Goedel employs are reasonable; they do not all appear self-evident. Nor is it particularly clear what they mean, and so it is also unclear whether the tendentious “God exists” is in fact a reasonable interpretation of what the conclusion of the argument actually says.

    • orthonormal says:

      Right. Everyone in logic agreed that Gödel’s version of the ontological argument was a valid proof, given the premises. It’s just the same problem as with the informal ontological argument: the premises are overpowered and unjustified.

    • Paul Torek says:

      Do you happen to know – What is the relation between those premises and the S5 modal logic system? S5 is already overpowered, I suspect.

      • Protagoras says:

        I think Goedel used S5, but he also used some other axioms that look much more questionable to me. I really don’t think S5 is to blame.

        • gwern says:

          FWIW, the machine-checked-proof paper notes that you apparently really do need something as powerful as S5 to make it work.

        • Paul Torek says:

          Vilhelm S had a link to a blog post by Neelakantan Krishnaswami, which answered my question. I meant to reply to myself but I forgot. Anyway, what you said, about the other axioms. In spades.

    • Josh says:

      The ontological argument is nothing but semantic trickery. The terms are defined in such a way that the desired conclusion is true by definition.

      1. God is defined as the greatest conceivable being.
      2. “Greatest conceivable being” is defined in such a way that existence is a necessary condition, i.e. something can’t be the greatest conceivable being unless it exists.
      ∴ God exists by definition.

      • nydwracu says:

        That’s not original to Gödel…

      • Ken Arromdee says:

        I don’t think that proves the greatest conceivable being exists. Rather it proves “every member of the set ‘greatest conceivable beings’ exists”.

        Normally “X has property Y” and “every member of the set ‘all X’ has property Y” are synonymous, but when Y is existence, they are not. The set could be empty, yet it could still be true that all members of the set exist.

        • Josh says:

          Yeah, but I can conceive of the greatest possible set of greatest conceivable beings, and a set that contains elements is greater than an empty set, so this set cannot be empty.

          Come to think of it, a set containing many elements would be greater than a set containing a single element, so I think I just disproved monotheism also.

        • Ken Arromdee says:

          Josh: That doesn’t work. It’s sets all the way down.

          In other words, you’ve then proven “every member of the set ‘all sets which contain….’ is not empty”.

          It’s really just a trick of language. I phrased it as sets, but you don’t have to. “All unicorns have one horn” means “all unicorns that exist have one horn”. Proving that God exists with that kind of proof would lead to “all gods which exist, exist”.

        • Sniffnoy says:

          Existence isn’t a predicate, guys…

          (Yes I am aware that more sophisticated versions of the argument (e.g. Gödel’s) get around this problem. Still, you are talking as if it is one.)

        • Josh says:

          @Ken, yeah I know. My post was tongue-in-cheek.

  3. K says:

    Think maybe the practice article link is broken?

  4. Sniffnoy says:

    They don’t seem to have drawn any actual border between Khurasan and Qozqaz…

    • Andy says:

      It’s there on the original, but not on the translation, which appeared to have been done pretty crudely.
      Original: Quartz article
      I can’t find a way for that border erasure to happen during the translation Photoshop process, so I’ll go out on a limb and assume the Daily Mail sucks at that sort of thing. I’m also quite annoyed that they didn’t translate the text at the bottom-right of the original.
      This whole flap has been a great illustration of a maxim employed with great frequency by one of my cartography professors: “Every map is an argument.” Most of the classic English-language examples of this date from WW2 German propaganda in the United States. I’ll be adding it to the “Pacific Radiation Map” in my developing rant on “maps can be used with false context to instill fear and panic, and this is a very very very shitty thing to do, SO DO NOT DO IT AND DO NOT FALL FOR IT.”
      Also, a useful heuristic is “if it’s on ABC or the Daily Mail, it is probably closer to false than true.” The map was released by a group of online Islamic State fans and doesn’t seem to have any official standing. ABC got it via Third Position, an American Nazi group, which should have been a giant red flag to them too, but… ABC sucks, and the Mail sucks. I don’t know why I was surprised by that.

      • AF says:

        The translation they did do is pretty terrible. The original map just has the standard Arabic names for those regions, which someone decided to translate letter by letter, adding a few new letters along the way.

        The map would look a lot less foreign and scary if they wrote “Europe” instead of “Orobpa,” Ethiopia instead of “Habasha,” Anatolia instead of “Anathol,” etc. These words have common English translations that should be used.

        Scott, there’s only one z, but if you want to call the Caucasus “Qoqaz” you just need to speak Arabic.

        As for the text in the bottom right, the large block is the profession of faith: “There is no God but God and Muhammad is the Messenger of God.” The same thing is on the flag of Saudi Arabia. Underneath that it says “The Caliphate.”

        • Andy says:

          As for the text in the bottom right, the large block is the profession of faith: “There is no God but God and Muhammad is the Messenger of God.” The same thing is on the flag of Saudi Arabia. Underneath that it says “The Caliphate.”

          Thanks! Though I do appreciate knowing the Arabic names for places, and I thought the Arabic names were appropriate. Naming something on a map is a way of taking power over it, the transliterations of the Arabic names I thought were more appropriate than the English names for a map from the ISIL fans’ worldview. I’d probably do the same if I were studying ISIS/L/Islamic State from an academic viewpoint.

        • AF says:

          Andy – That’s a good point that naming places involves claiming power over it. I should not complain about examples like Ethiopia where the Arabic word actually is different.

          For most of these, however, the standard English name is already a better transliteration than the one they give. The Arabic word for Europe is a transliteration of Europa – there’s no reason to put both a b and a p at the end. The middle consonant in the Arabic word for Anatolia is much closer to a t than a th. Yemen is closer to the Arabic than Yaman. With these changes, the Daily Mail map sends a different message than the original.

          Also the Daily Mail thought Khorasan wasn’t big enough and decided they needed to add India to it…

        • Anonymous says:

          Also the Daily Mail thought Khorasan wasn’t big enough and decided they needed to add India to it…

          For shame! I just noticed that Daily Mail and Third Position are using different maps. I wonder if the Daily Mail made its own version or found another version of the map somewhere. I’d bet on the latter -it seems like less work than the Mail remaking it. I’m now picturing multiple groups of ISL/IS fans around the world producing fanmaps and imagining themselves as governors of a province, like the Southern lunatics who kept trying to add Cuba to the United States before the Civil War.

  5. lmm says:

    A lot of QM is really just wave behaviour (e.g. the uncertainty principle applies perfectly well to a classical wave packet). There are times when I think the best first-order approximation would be to just forget any notion of particles. is not at all what you were talking about, but kind of amusing, at least if you like Hitler jokes.

    • lmm says:

      More seriously, Look Who’s Back connects much more closely to this stuff than I originally thought.

    • Luke Somers says:

      Like I always say (when it comes up): everything is always waves. Sometimes, waves act like particles.

  6. Elizabeth says:

    “” used to be a Livejournal-clone blogging platform, and I used it when I was twelve to enthuse about Runescape dragon fanart and complain about school. It’s always pretty weird when people link to it.

  7. anonymous says:

    The link for “practice makes one-third perfect…” doesn’t work.

  8. Samuel Skinner says:

    I am deeply amused that the Chinese plan mentioned is either similar to what neoreactionaries want or an attempt to make the Cathedral explicit, depending on the details. Anyone think the Chinese will be idealistic/nutty enough to try to pull something like this off?

    An example of what is mentioned in the link:
    “Today, many scholars are advocating a return to this old system, but with improvements based on modern technology and new ideas. There are many various proposals, for example, the scholar Jiang Qing proposes two houses, with an upper meritocratic house selected by examination, performance statistics and interviews, and a democratically elected lower house. The lower house can propose and write legislation, but the upper house holds ultimate veto power. The general idea behind all these proposals are the same, which is to select leaders who are moral, knowledgeable and highly capable, but with the emphasis on morality. With modern technology, leadership select can be vastly improved compared to ancient times. For example, popular opinion on issues can be tracked in real time through extensive online polling, big data allows for performance to be tracked with objectivity and precision, telecommunications can greatly facilitate interviews and provide insights on a candidate’s life history. I think this ancient idea, revitalized by modern technologies, can be hugely promising.”

    • bad at pseudonyms sorry says:

      Mencius Moldbug always had shades of Confucianism to me, but despite his moniker, he apparently hasn’t read any Confucian philosophers.

      • AWJ says:

        I have Moldbug pegged as more of a Legalist than a Confucian. Between Moldbug and the Book of Lord Shang there’s a remarkably similar blend of viciousness and contrarianism (shorter Lord Shang: “all those things the Confucians claim to be virtues are actually BAD, and rulers should stamp them out!”)

    • lmm says:

      The idea that you want moral leaders is pretty uncontroversial. If you’re not going to elect them democratically then exams seem like the obvious alternative. Do the similarities run any deeper than that?

      • Samuel Skinner says:

        It depends entirely on what one means by moral. If you mean not corrupt, than sure. If you mean people who see something wrong with shooting protestors in Africa to make sure the mines and farms run on time or- heaven forbid- utilitarians who think funds should be spent helping the worlds poorest, not just citizens, than no, moral isn’t necessarily in the best interests of the state and citizens.

    • peterdjones says:

      Aristocratic upper houses arrent entirely new, although examination entrance might be. The semi-reformed House of Lords is sum meritocratic … distinguished figures are given peerages so they can within in it.

    • Oligopsony says:

      IMO, if you want your legislators to be moral, you want aristocracy or sortition, not meritocracy. Anything that allows behavior to influence who’s in power will select for those who want power the most.

      • Zathille says:

        Does that not assume wanting power the most is an immoral trait in itself? Of course, it could be that it is associated with bad behaviour, but I’d say it’s less about wanting power itself and more with wanting power ‘at all costs’ (And so, making tradeoffs that privilege gaining power at costs that may damage others more than acceptable)

        On a tangent, I tend to be suspicious of those who seek a ‘more moral leadership’, partly because certain moral values, while almost universally laudable in general, may be a dog-whistle for certain in-groups and the like. I guess it’s a matter of being suspicious of it being used as a Motte and Bailey at this point.

        ‘Whose morality?’ Might be a pertinent question, perhaps.

        • Andy says:

          ‘Whose morality?’ Might be a pertinent question, perhaps.

          A very pertinent question. Being infected with current US politics, when I hear “we need more moral leaders!” I generally assume someone’s talking about Judeo-Christian morality, in which case a big chunk of my friends are headed to the stoning fields.
          But I think choosing leaders who are strong in Consequentialist morality would be pretty much the ideal. Scott Alexander for Congress? A wonderful fantasy, almost too sweet for words…

        • Multiheaded says:

          ‘Whose morality?’ Might be a pertinent question, perhaps.

          Even further, I think all three of us would agree that ultimately the material aspects of how the system exists and recreates itself are so vastly important, we’d expect them to usually outweigh the difference between a moral administration and an evil/corrupt one. For example, William III of Orange is usually hailed as an agent of progress and a huge step towards a bourgeois state in Britain, but I distinctly remember reading that he was really detestable as a person and a sociopath/sadist of some sort (beyond allegations of homosexuality, that is). Can’t find the link, sorry.

      • Alexander Stanislaw says:

        Sortition sounds like a really cool idea for a system of government. It achieves one of the goals of democracy in a very pure way – it ensures that the government is representative of the people via random sampling. The fact that no one has tried it successfully in modern times probably means there are some implementation issues but I still like the idea.

        It would be uninteresting to come up with a list of pros and cons of sortition, so instead I’m going to ask the question – what would a society based on sortition look like? First off, there would a different sense of civic duty – instead of the duty to vote, there would be the duty to take up a position as a politician if you are chosen. Secondly, the sorts of issues that get discussed would probably become more biased towards issues that affect more people, rather than issues that have an effect on a smaller number of people. Right now, because of the way the lobbying system works, smaller groups who have strong interests (eg. the corn industry) can get legislation passed in their favor better than larger groups with weaker interests (eg. people who buy foods with sugar)*. This would be reversed under sortition. As for political factions, they would probably be recreated even without a party system. However, there would probably not be two factions in general.

        *If this is too opaque in the US, the corn industry benefits from several policies which have the combined effect of making corn more profitable and making sugar more expensive for Americans. The corn industry gets its way because they have a strong incentive to keep these policies in place, but regular Americans don’t have much of an incentive to oppose them since the cost to each person is very little.

        • Samuel Skinner says:

          What would it look like?

          Long term limits (because people need to learn on the job)
          Staggered elections like the senate (to prevent everyone from learning at once)
          Much greater focus on political personalities (there is going to be a much greater spread in intelligence and experience under such a system so the best are going to starkly stand out)
          More power to other branches of government
          More idealism and less pork (because individuals don’t have to worry about reelection so they can sponsor things that don’t immediately benefit constituents)
          Stronger lobbyists (because they know the proper way to write laws)

          A major change is that there will be a lot bigger focus on consciousness raising and education by political movements. You want everyone to have at least heard of what you believe in so that when someone is chosen for government it is easier to make a connection with them.

          A big change would obviously be political movements. I don’t see republican/democrat divide surviving because it is an artifact of first past the post. Any idea what the political breakdown would be?

  9. suntzuanime says:

    Telegony sounds like the sort of thing that would be invented in a hurry when your child looks more like your supposed-ex than your husband.

  10. Anonymous says:

    But, uh, if a proof of God’s existence is correct, that should mean God exists

    Wrong. It means what the theorem defines as God must exist when the preconditions set forth by the theorem are true.
    I read that theorem (which Gödel never published, btw) and a phisical interpretation is far from trivial. Also the modal logic it uses is weird and rarely used and a few passages in the proof felt fishy.

  11. Avery says:

    The cookbook is an April Fool’s joke, btw.

  12. Vilhelm S says:

    Btw, Neelakantan Krishnaswami recently made a nice blogpost about Gödel’s ontological proof.

  13. Alejandro says:

    Also, I imagine two ISIS members daring each other to try to draw Khurasan bigger and bigger, then laughing and keeping it on the map because they’re not going to achieve global domination anyway.

    This would also make perfect sense when giving “ISIS” the meaning that comes first to my mind when reading it: the incompetent spy agency in the outstanding animated series Archer.

    • Scott Alexander says:

      I figured – hence the “supposedly” – but I still found it interesting,

      • AWJ says:

        I found it even more interesting that Business Insider chose to source the map in the form of an embedded tweet from “ThirdPosition”, a white nationalist feed.

        Didn’t Business Insider already get some negative attention back during the Zimmerman trial for linking to Stormfront?

        • Andy says:

          That’s the same source as ABC and the Daily Mail. It’s an older map from some ISIS/L fans, but “Source Zero” for the current outbreak seems to be Third Position.

  14. There’s a lot of stuff I don’t get here – instead of measuring income inequality and assuming it was racial, why didn’t they measure income of blacks directly?

    They did. The measure Bertocchi & Dimico call “racial inequalty” is defined as horizontal inequality, which is normally proxied in contexts such as these as ratio of group income to group income. The whole paper seems like retreading the obvious, but it has to be understood in the academic context of debates surrounding the institutional origins of inequality that got kickstarted by Acemoglu.

    There’s a lot of stuff I don’t get here – instead of measuring income inequality and assuming it was racial, why didn’t they measure income of blacks directly? Also, how does this square with our last paper that found that descendants of enslaved and free blacks had equal outcomes within two generations of emancipation?

    The two are, what is the word, “orthogonal” ? The first paper is only interested in whether the “degree” of slavery in the past correlates with inequality today and does not really concern itself with the intergenerational within-group trajectory of income.

  15. MugaSofer says:

    Solve all tornados by building a 1,000 foot high wall across the Midwest.

    So many Pacific Rim jokes …

  16. Q says:

    The article hypothesizes, the monkeys do not like the western music because of the regular rhytm, which to them sounds like fist-chest banging of males fighting for dominance. What is human equivalent of music we would not like, because it interferes with our survival cues ?

    The best example which comes to my mind is an opera Salome from Schoenberg, where the constant high pitched tone in the backround makes it really unpleasant to listen to. And that means – what ? That the listener fears he has ear-ringing, the first sign of becoming deaf ? Or siren ? Or both ? Any better example of unpleasant features humans do not normally incorporate into music (unless they are Schoenberg) ?

    • Creutzer says:

      Not that that makes it any more pleasant, but “Salome” is by Richard Strauss, not Schönberg.

      I can’t actually think of anything other than a high-pitched whine that is likely to be universally considered unpleasant by humans… And I don’t actually know why high-pitched whines are so bad.

      • Q says:

        OMG, you are right about Strauss.

        Another idea about high pitched sounds – mosquitos make them.

      • Randy M says:

        Baby cries?

      • Phil Goetz says:

        A fair amount of Chinese classical opera is very nasal and sounds like high-pitched whining to me. China also has a unique instrument that sounds beautiful to me, but I have to admit it could be described as like a kazoo pitched an octave higher. The instrument is a plant leaf, blown across like the membrane in a kazoo or harmonica.

      • rehana says:

        One of my high school textbooks (physics or biology, I don’t remember) said it was because the volume level where you can hear it is closer to the level where it becomes painful than for lower sounds. Seems as good a guess as any.

    • Lavendar bubble tea says:

      I don’t really have any idea, but this question made me realize that some of my favorite music is that which does mess with my survival cues. I sometimes listen to a lot of industrial, dark ambient and metal music, all of which actively try to mess with survival cues. This reminds me of an article I found purposing “benign masochism” as an explanation for several behaviors. ( )

      Apparently animals don’t seem to experience the same thrill seeking mannerisms.

    • Nornagest says:

      What is human equivalent of music we would not like, because it interferes with our survival cues ?

      I could make so many jokes here.

      But the serious answer that comes to mind is some artists’ habit of sampling police sirens (e.g. in Judas Priest’s “Breaking the Law”), which can be seriously disconcerting when you’re playing the track in your car.

  17. Regarding the mental rotation paper, it’s not the first time this correlation has been observed. This paper cites other sources suggesting the effect is not particularly stable, though of course it certainly cuts against popular narratives.

    Incidentally, compare the data/graphs in the paper to the language. It’s amazing the verbal gymnastics they go through to ignore obvious conclusions.

  18. Gadren says:

    This is probably obvious to most, but I thought it important to point out that the obesity study shows a difference not just in metabolism but also in food consumption. I bring this up only because I know there are some — particularly in the HAES (Healthy at Every Size) movement — who skip this part, saying that many gain weight despite eating nothing more than what others are eating. I suppose it shows the importance of food measurement / calorie counting rather than similar advice to “eat until/before you feel full” since that signal may not be delivered at the right time for one’s body.

    • zslastman says:

      There are some differences in ‘metabolism’, mainly due to brown fat levels, but in genetics, obesity is generally accepted to be heritable via behavior. Genes that cause chronic heritable obesity are usually involved in the satiety system. Note that behavior encompases not just how much you eat, but how much you move – people’s tendency to get exercise, and to just walk around and fidget during the day, varies widely.

    • Alrenous says:

      Eat until you’re not hungry works fine unless your satiety hormones are being borked.

      Low-fat diets interfere with many people’s satiety. Mine, for example. I think the next major interference is sleep disturbance (including alarm clocks) followed by sedentarism. You can’t expect your body to take care of you if you don’t take care of it.

      • No one special says:

        Satiety being FUBAR is the primary idea behind the Hacker’s Diet. He calls it “the eat watch”. The entire idea behind that book is to replace a broken satiety system with an artificial one using math.

        (The Hacker’s Diet is for those who don’t know it.)

  19. Zorgon says:

    Given that you don’t really do open threads, I’m gonna take the opportunity of this link thread to observe that WordPress has apparently chosen to gave me a stylised swastika as an avatar. Erk.

    • Andy says:

      All of the randomly-generated avatars seem to be generated in the same way as a swastika – generate a square pattern for one quarter of the canvas, then put it in the adjacent space, turned 90 degrees, rinse and repeat until you fill the canvas. I wouldn’t call yours a stylized swastika at all.
      And if it bothers you, imagine it as a stylized Native American “Rolling Log” pattern… which happens to look pretty much like a swastika, only with a totally different cultural context.

    • Armstrong For President 2020 says:

      Crosses circles triskelions and swastikas tend to crop up everywhere for the same reason; they are simple shapes and have radial symmetry, making them very pretty to look at.

      It shouldn’t be surprising if an algorithm for producing simplistic pretty shapes occasionally throws out something cross- or swastika-like.

    • James Babcock says:

      The stylised random images are generated by Gravatar ( If you don’t like yours, you can upload an image file there and it will use that instead.

    • Mary says:

      You can call it a fylfot instead to avoid connotations.

    • Phil Goetz says:

      What’s more embarrassing is, your swastika is backwards.

  20. Dave says:

    > And it ends with all of Odysseus’ sons hooking up
    > with all of Odysseus’ sexual partners, which I guess
    > isn’t especially weird for a Greek myth.

    It was probably Odyssey fanfic that later got assimilated into canon.

    • tiny nerd says:

      IIRC the Epic Cycle was originally all attributed to Homer but modern scholars have decided it wasn’t a single author. However, one could say the whole Odyssey is fanfic, since an “authoritative version” of something that old doesn’t make much sense. I’d be very interested to see some kind of textual analysis aimed at picking out which parts of the Iliad/Odyssey are most likely from a single origin and which were added on later but that’s probably impossible?

  21. zaogao says:

    I really wanted to see a picture of the deep sea horror but realized a google image search for “Black Swallower” is probably NSFW.

  22. alexp says:

    I think the reddit guy was very intelligent, but seems to have a bit too rosy a view of the old Confucian Civil Service. While I believe the entrance exams were mostly above board, I don’t seen any indication that advancement once you got in had any less Patronage, politicking, or personal connection involved than the modern CCP, or any large bureaucracy you can think.

    That’s not to mention the subjectiveness of moral examinations, or the fact that that system was famously resistant to change.

  23. gwern says:

    Less Research Is Needed: an article on how much the author hates the phrase “more research is needed” and how in some cases it can be used to make debate interminable so that the “wrong” side of a controversial question can never be proved.

    No, their article is about how they endorse terrible research practices. I disliked it so much I began fisking it as I went, only to discover when I reached the end that comment are closed! Dammit. So I’ll just post it here.

    These four words, occasionally justified when they appear as the last sentence in a Masters dissertation, are as often to be found as the coda for a mega-trial that consumed the lion’s share of a national research budget, or that of a Cochrane review which began with dozens or even hundreds of primary studies and progressively excluded most of them on the grounds that they were “methodologically flawed”.

    If studies have to be excluded for lacking basic desiderata like randomization, blinding, and having necessary summary data available, that’s an indictment of the standards of practitioners, not of meta-analysts, in employing techniques known to dramatically inflate the rate of false findings (both positive & negative), and detect effects due to placebos, experimenter bias, etc.

    On my first day in (laboratory) research, I was told that if there is a genuine and important phenomenon to be detected, it will become evident after taking no more than six readings from the instrument. If after ten readings, my supervisor warned, your data have not reached statistical significance,

    For someone who sneers at anyone who notes that most studies are ‘methodologically flawed’, you seem awfully eager to advocate no rater blinding, data snooping, and multiple-testing. I hope your lab enjoyed all its false positives. (Or wait, it probably did, since as you point out, there’s no accountability for that…)

    you should [a] ask a different question; [b] design a radically different study; or [c] change the assumptions on which your hypothesis was based.

    This makes sense only under the bizarre assumption that all effects are large, and no one ever cares about small effects.

    The authors conclude that not only is more research needed into the genomic loci putatively linked to coronary artery disease, but that – precisely because the model they developed was so weak – further sets of variables (“genetic, epigenetic, transcriptomic, proteomic, metabolic and intermediate outcome variables”) should be added to it. By adding in more and more sets of variables, the authors suggest, we will progressively and substantially reduce the uncertainty about the multiple and complex gene-environment interactions that lead to coronary artery disease.

    And what exactly is wrong with this? We can estimate, based on techniques like GCTA, that most of the genetics variables must be numerous and of small effect in many things we might want to invetigate. There’s not much one can do except increase sample size and collect more covariates to reduce variation! It’s never going to be easy to detect small effects.

    But when we try to predict what the weather will be next month, the accuracy of our prediction falls to little better than random. Perhaps we should spend huge sums of money on a more sophisticated weather-prediction model, incorporating the tides on the seas of Mars and the flutter of butterflies’ wings? Of course we shouldn’t. Not only would such a hyper-inclusive model fail to improve the accuracy of our predictive modeling, there are good statistical and operational reasons why it could well make it less accurate.

    Organizations like the NOAA do spend huge sums of money on more sophisticated weather-prediction models incorporating enormous datasets from around the world and using incredible amounts of computing power. And they do get much better weather forecasts as a result. Not out 30 days, but out 9 days perhaps.

    Lack of hard evidence to support the original hypothesis gets reframed as evidence that investment efforts need to be redoubled.[2]

    So, when we have a ‘lack of hard evidence’, is deciding the original hypothesis must be bunk really ‘the only logical conclusion that can be drawn from a set of negative, ambiguous, incomplete or contradictory data’?

    Despite the almost complete absence of ‘complex interventions’ for which a clinically as well as statistically significant effect size has been demonstrated and which have proved both transferable and affordable in the real world, the randomized controlled trial of the ‘complex intervention’ (as defined, for example, by the UK Medical Research Council [3]) should remain the gold standard when researching complex psychological, social and organizational influences on health outcomes.

    One man’s modus ponens is another man’s modus tollens. Why do the highest-quality studies, in which everyone is carefully blinded to avoid their own biases, sample size is high enough to overcome natural variation, analyses are prespecified so there is no chance of data snooping (oops, I guess your supervisor doesn’t approve; too bad), treatment is randomized to avoid the infinite universe of confounding factors, show such few effects? Could it be that… the interventions don’t actually work? Or have we discovered in RCTs a potent ‘anti-sciencium’ which mysteriously makes all treatment equal in its presence? That would be useful to research.

    For every paper that concludes “more research is needed”, funding for related studies should immediately cease until researchers can answer a question modeled on this one: “why should we continue to fund Kitty-cat’s attempts at flight”?

    Indeed, one might ask that. But on which side should the burden of proof be? Who’s the cat?

    A more constructive approach would be to advocate more use of value of information & futility analyses.

    • Scott Alexander says:

      “Could it be that… the interventions don’t actually work? Or have we discovered in RCTs a potent ‘anti-sciencium’ which mysteriously makes all treatment equal in its presence? That would be useful to research.”

      I think you’re missing sarcasm in the original and that it is actually agreeing with you.

      • gwern says:

        I don’t think it’s sarcastic at all. It seems quite sincere in its espousal of the biased ineffective approaches which have been discredited by RCTs.

        • Anonymous says:

          It’s not sarcastic, but it is one of four listed fallacies.

          But she’s not arguing against RCTs per se. She’s making the observation that almost all RCTs of “complex interventions” failed to show significant results, and arguing that perhaps we should re-evaluate what we’re doing. She’s arguing that it’s a mistake to try the same old thing, only trying harder this time.

          That doesn’t mean we should go back to non-randomized or non-controlled studies. Those things are important. But maybe we’re missing something. Maybe RCTs are not good enough. Or, maybe “complex interventions” generally don’t work and we should do RCTs on something else.

          She’s arguing that if we fail to fly every time we jump off a cliff, we shouldn’t simply try to jump off a higher cliff. Maybe we need to add something, like wings. Add some wings, and then jump off a cliff. Or maybe we need to reevaluate the goal. The goal of flying is to catch the bird. Maybe we could skip flying and use a net instead. Who knows? All we do know is that what we’re doing right now isn’t very effective.

  24. Anonymous says:

    Because speaking out against the government and going to prison for it is exactly the same thing as taking classified information you have access to at work and defecting with it to a communist country.

    • Steve Johnson says:

      You meant defecting from a communist country.

      • Creutzer says:

        I think he mean “defecting to a former Communist country”, which is of course an irrelevant property in this context.

  25. Armstrong For President 2020 says:

    If these guys have properly interpreted the psychometric data, it seems to me that there is a more fundamental problem with the CCP’s bureacracy than cronyism which testing could potentially help solve.

    As efficient as the CCP’s Organization Department is, especially given how dysfunctionally the nomenklatura system usually works, it has a fairly serious flaw that Confucian test-based systems lacked; it requires leaders to rise up through the ranks one at a time based on performance evaluations at each job. This means that since it is difficult to lead effectively when there is greater than a 19 point IQ difference between leaders and followers, intelligence for anyone in the higher ranks would be effectively capped at about a 124 (assuming that the typical CCP member is of average Han intelligence). This isn’t stupid certainly, about on par for corporate leadership and science PhDs, but it is still a waste of potential for such a bright population.

    By using a formal system of testing like the Confucians or relying on associative mating like Indo-European aristocracies and selecting high-level leadership from within that pool, this problem should be mitigated significantly. Between the growing Chinese interest in psychometrics and the rise of ‘Princelings’ socially we may well see them do both. Either way, it will be interesting to see the results.

    • Scott says:

      Link for the IQ difference claim? That sounds very interesting.

      • Armstrong For President 2020 says:

        It’s from the link at the top of the post; follow it and scroll a few paragraphs down. They also give a citation for the original research their conclusion is based on.

    • gwern says:

      If these guys have properly interpreted the psychometric data, it seems to me that there is a more fundamental problem with the CCP’s bureacracy than cronyism which testing could potentially help solve. seems like a really weird analysis. Leaving aside the lack of any correction for Flynn and my skepticism that they really can’t find more recent data for various elites, its theoretical results seem to contradict the SMPY results, and I’m left a little boggled by the apparent procedure:

      This all ‘hangs together’ statistically. In other words, an IQ of 145 is at the 99.8%’ile of the elites from which these eminent members are selected. This implies that IQ is an important component of success in entering and remaining in these elite professions and that the most eminent among them have higher IQs to a statistically significant degree. However, imbedded in these statistics is a surprise. By dividing the distribution of the elites (126 SD 6.7) by the distribution of the general population (IQ 100 and SD 15) we can statistically infer the relative probability that a person of any given IQ will enter and remain in an intellectually elite profession. Not suprisingly, the probability increases with higher IQ. It does so up to an IQ of 133. It then begins to fall, slowly at first but precipitously at higher IQ levels. By an IQ of 140 it has fallen by 1/3. By an IQ of 150 it has fallen by 97%! There are an estimated 250,000 English speaking people with IQs over 150. They are being nearly entirely excluded from intellectually elite professions

      Can that possibly be right? This feel like it’s drawing on illicit assumptions like symmetry of distributions.

      And the conclusions they draw about leadership don’t seem like they’d apply to many fields (why don’t all the ultra-high-IQ people just go into mathematics or similar technical fields where ‘leadership’ would hardly seem required? do you need leadership to simply get on the science faculty of Cambridge?).

      • Armstrong For President 2020 says:

        I’d be more confident that the 19 point leadership limit is accurate in the 100-145 range than for the extreme upper limits. From my limited knowledge it sounds like a lot of trends in IQ don’t hold after about 3 sd, which would seem to fit the intelligence/height comparison.

        I’d be glad if you could take a closer look at it though; your summaries on LW were always really top notch, and this is something potentially very interesting.

      • Douglas Knight says:

        No, it’s complete nonsense for exactly the reason you say. Have more faith in yourself.

  26. Anonymous55 says:

    Scott, if you are in the New England area on Monday, I just wanted to let you know that Robin Hanson is visiting the rationalist house in Boston. See
    Any chance you can make it?

  27. anon says:

    While I don’t understand quantum mechanics, I’d be interested in seeing whether anyone thinks Bohm can be reconciled with the approach to quantum gravity mentioned in the comments here:

    • Chris Billington says:

      Ordinary quantum mechanics has an element of Dawinian evolution to it, see Quantum Darwinism. It’s basically the idea that interaction with the environment favours particular quantum states — and these are the states we see in practise.

      It was previously a great mystery that although the single particle Schrödinger equation predicted that small spatial wavefunctions would evolve into large, extended ones, we tend to observe particles as well localised in nature. The Darwinian thinking solved this within the context of environment induced decoherence, explaining why localised states tend to survive interaction with the environment better than other states (this is also called ‘einselection’).

      Because this came out of the field of decoherence, it fits well into a many worlds interpretation. But since Bohmian mechanics includes the wavefunction too as the pilot wave, it ought to be subject to quantum Darwinism as well.

  28. Kaminiwa says:

    So what happens when we combine the Chinese goal of efficient governance with the Hong Kong subway AI/Expert System?

  29. Alrenous says:

    Practice makes

    I thought the point of the 10k hour rule was that since a talent with no practice is so different from a talent with practice, a no-talent could hack together a respectable showing by practicing anyway. Sure, practice + talent is always going to outweigh practice + 0, but probably not by a ratio of 3 to 1.

    To pick an example totally at random, someone could out-philosophize the entire establishment by putting an obscene amount of time into improving their skills, despite having a relatively mediocre IQ.


    I think you have the causation backwards. Humans, particularly men, cared about sexual purity but merely asking their women to stay pure wasn’t working, so they made up a story to bolster the case and validate their anger and disappointment. That is, Sophistry is a deliberately evil subculture but also sophistry is a natural thing humans think of doing.

  30. Matthew says:

    On the subject of obesity, this study was making the news a few days ago. I’m curious if others see obvious flaws in it. (I don’t trust my own judgment because it flatters the conclusion flatters my prior. I keep seeing everyone else argue weight is mostly about diet, not exercise, which annoys me because my own experience speaks to the contrary.)

  31. Douglas Knight says:

    It’s not less research, but less replication that is needed.

  32. James says:

    The article on neo-nazi hipsters seems to be using “young person who wears t-shirts” as their definition of hipster.

  33. Flesym says:

    I didn’t really understand your comment about cryptoequity; it reads to me like you perceive some sort of a Pandora’s Box. But I don’t see a problem with cryptoequity. It looks like an intriguing new way of “incorporating” and seeking funding.

  34. Nestor says:

    I wonder if there are people like Tarrare and Charles Domery running around in the modern world, and we just don’t know because it’s relatively trivial to feed their appetites with modern food availability.

    (Could’ve sworn I first heard about Domery from this blog, but a cursory search didn’t find him, but it unearthed Tarrare while I was doing it. Amazing there were 2 people like that in the same time period and close geographical proximity. Must’ve been something in the water.)

  35. Jaskologist says:

    It seems worth pointing out that @SalonDotCom has been suspended. One presumes that it is because it was impossible to tell from the real one.

  36. Matthew says:

    I can’t put this in the open thread because flamebait, but I thought this would probably interest people: The Feminist Leader Who Became A Men’s Rights Activist

    • suntzuanime says:

      Presumably the reason for the “no random flamebait in the open thread” rule is to keep there from being random flamebait in the comments, not to ensure that the random flamebait is in the comments of some other article.

      • Matthew says:

        I would dispute the term “random” in this case. It’s not irrelevant, and there are already gender links in the above post.