"Talks a good game about freedom when out of power, but once he’s in – bam! Everyone's enslaved in the human-flourishing mines."

Open Thread 70.5

This is the twice-weekly hidden open thread. Post about anything you want, ask random questions, whatever.

I’m experimenting with a policy of keeping alternating Sundays’ Open Thread (ie the X.5 thread) culture-war-free. Please avoid any discussion of race, gender, tribalism, free speech, object-level politics, et cetera out of this thread (you can discuss the policy itself if you want). You can discuss culture war topics on the next thread starting Wednesday.

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

407 Responses to Open Thread 70.5

  1. Machina ex Deus says:

    Frist! OK, here’s something I’ve been wondering about for a while:

    I’m pretty sure there are at least three or four SSC readers who don’t have depression. Who are you, and what’s it like? (Feel free to use a burner account if you’re uncomfortable with people knowing you’re sane.)

    Edit: In particular, do you still have problems with motivation and getting stuff done (akrasia)? Do you find yourself discouraged at the apparent lack of meaning in your (and other people’s) life? How often are you actually, positively happy, and why?

    It can’t hurt to include your country, since this stuff seems to be viewed differently different places (e.g. in the U.S. I’m depressed, but I’d make a pretty upbeat Dane, and a downright manic Finn).

    • I don’t have depression. On the whole, I am happy with the present state of my life. I don’t get as much done as I should, but I do get a reasonable amount done.

      • Paul Brinkley says:

        This is roughly me, as well. I took a fairly big career hit in the middle of my life, but that was a gamble – if it paid off, it would have paid off tremendously. This was very sad, and felt very soul-grinding for years, but the cause was concrete and external, and within the normal probability window as far as I could tell.

    • vollinian says:

      Woah so the majority of SSC readers have depression?

      • Paul Brinkley says:

        From the parent: there are at least three or four SSC readers who don’t have depression.

      • Corey says:

        I think the overall US rate is something like 1/3, and it tends to increase with intelligence, so I wouldn’t find it surprising that a group of smarter-than-average folks could have a >50% depression rate.

        • birdboy2000 says:

          Also Scott’s written a fair bit on the subject, which draws in people who care about that. Also sedentary lifestyles are a risk factor and guess who spends all day on the internet.

    • hoghoghoghoghog says:

      I have many problems with motivation, am often discouraged, and am rarely positively happy. However I can usually say why I am not motivated, why I am discouraged, and what it is that would make me positively happy.

    • Philosophisticat says:

      I am not at all depressed, but I have massive problems with akrasia. I don’t think my life or others’ lack any apparent meaning. I am positively happy when I am watching a great movie or appreciating some other form of art, when a class I am teaching goes well, when I have a good paper idea, when other things go well for me personally or professionally, when I am having an interesting conversation, or when I am playing a fun game.

      I’m in Singapore but I’m from the U.S.

    • Scott Alexander says:

      I don’t think I have depression.

      There are certain things where I think I have motivation problems. I’ll leave packages on my table for months because I can’t muster up the minute or so it would take to open them and figure out where to put what’s inside. I got ticketed because I put off worrying about my car registration so long (a five minute process to solve) that it expired. So far I’ve put off a 10-minute phone call I need to make for two weeks. Things like that.

      Luckily, I have a job where it’s really clear how many hours a day I need to be there, and there’s such a bright-line Schelling fence that I never feel tempted to shirk it. I have a lot of habits (like washing my clothes on a certain day, and going grocery shopping on a certain day) that it would be really weird for me to miss them and motivation doesn’t factor into it. I really enjoy blogging, it doesn’t really take motivation, and it gives me my daily dose of feeling appreciated and worthwhile.

      (even so, the difference between normal-me and on-caffeine-me is like night and day)

      I get the really strong impression, both from my patients and from my friends, that there’s a certain category of thing which is really unnatural and which (almost) nobody is good at. This includes everything around grad school, where you have to finish a vague project over a certain number of years but there are no fixed hours, and a lot of programming tasks insofar as it’s not just “write X lines of code per hour” and you actually have to do intense mental work which is hard to measure or make linear progress on. People who have to do a lot of this always end up thinking they have ADHD or depression or something, and for all I know maybe they do.

      Contrast this with something like farming (or like my job!) where you show up at a certain time, there’s obvious stuff to do in a nice methodical linear way, and you do it. I think this is easy for people. The other thing is hard, and either produces mental illness or (more likely) makes people so miserable and unmotivated that they assume they’re mentally ill.

      This is just a personal theory and I don’t think mainstream psychiatry has an opinion on this.

      • Aapje says:

        a lot of programming tasks insofar as it’s not just “write X lines of code per hour” and you actually have to do intense mental work which is hard to measure or make linear progress on. People who have to do a lot of this always end up thinking they have ADHD or depression or something, and for all I know maybe they do.

        This is not my impression, although I’m not sure that’s because I merely disagree with the nature of the job or also with the idea that jobs of that type cause (feelings of) ADHD or depression.

        Even highly fluid tasks like debugging can often be made non-linear, in fact, I attribute my skill at it to changing the problem of ‘shit don’t work’ to a much more linear problem:

        – ‘shit working’ requires A -> B -> C -> D -> E
        – how can I verify that a portion of this chain is working?
        – once I identify that C is not working, learn that C is actually C1 -> C2 -> C3 -> C4
        – how can I verify that a portion of this chain is working?
        – once I identify that C2 is not working, …

        This cycle ends when the cause becomes apparent, although in practice it can be hard to test the parts of the chain. Proper modern programming technique is to make this relatively easy.

        It becomes bad when you have bugs that occur semi-randomly or heisenbugs (where debug statements or other methods to test parts of the chain changes the behavior of the system). These are most common when dealing with multi-threading aka concurrent execution.

        From my layman perspective, the more difficult parts of healthcare seem to involve semi-random bugs and heisenbugs, as well. The placebo effect is basically a heisenbug, for example.

        • John Nerst says:

          heisenbugs

          This is a brilliant word.

        • shakeddown says:

          What about tasks like building a program from scratch? Wouldn’t that have the same challenges as, say, writing a thesis?

          • beleester says:

            Usually a program has some sort of design (hopefully, an actual written one so that you and the client both agree on what you’re doing). And the design can usually be broken into linear tasks – “We need feature X, Y, and Z before we ship. Feature X requires functions A and B to implement. Function A and B require common system D. So we write D, then A and B, then X…”

            Anecdotally, it’s far easier for debugging to stall my productivity. I stall out when I hit a point where I say “I have no idea how to proceed, and I don’t have the energy to find out.” In debugging, that point is all the time, because it’s someone else’s code and I don’t know how any of it works without putting in the mental effort to find out (or the mental effort to call up the person who knows and ask). In new development, it’s my own code, I know how it works, and I generally know what needs to be done. So the mental effort required to do new development is much less.

          • Cypren says:

            Agreed with beleester on this point. Writing new code is probably the single most joyful task in programming; it’s why most of us got into this field in the first place, and it’s akin to an artist with a blank canvas: unlimited possibility and creative potential laid out in front of you.

            I think the main difference between writing code and writing a dissertation is that the former has objective standards it can be measured against; the latter usually does not. A dissertation is being subjectively judged by a committee of humans, and creating a successful one is often a matter of “playing the man, not the ball”: tailoring it to the subjective demands and desires of the defense committee and your faculty advisor. Programming, in contrast, typically results in measurable and quantifiable output that can be objectively assessed.

            The worst task, by far, is modifying or debugging someone else’s code. It’s probably not really appreciated by most non-programmers that code is literally concrete expression of abstract thought. When reading someone else’s code, you are forced to model their thought process and assumptions to understand it; you must literally think like the original programmer for a short period of time and in a limited context.

            It turns out that humans are really bad at this. It’s frustrating, difficult and at times rage-inducing. Young programmers tend to handle it by complaining to their friends and coworkers that whoever wrote the code they’re working on was clearly a complete idiot. Older programmers are more likely to take it in stride and assume the person had reasons for doing what they did, but it doesn’t make it any more fun.

          • The Nybbler says:

            The worst task, by far, is modifying or debugging someone else’s code.

            Not for all of us. I mean, it’s still intensely difficult and frustrating, but getting to the bottom of and fixing that nasty Heisenbug in someone else’s crufty old code can be incredibly rewarding.

          • skef says:

            Agreed with beleester on this point. Writing new code is probably the single most joyful task in programming; it’s why most of us got into this field in the first place, and it’s akin to an artist with a blank canvas: unlimited possibility and creative potential laid out in front of you.

            Programming, in contrast, typically results in measurable and quantifiable output that can be objectively assessed.

            In my experience, and especially on projects that pose engineering challenges, the more joy a programmer feels at a blank slate, the longer it will take the code to converge at the end through longer hair-pulling debugging sessions. Most of those programmers also see those hours as inevitable, so they’re rarely objectively assessed by either the people who go through them or management. And at the end of that debugging process, let one of those programmers come back to the code after two years and they’ll see it as a crufty mess too.

            That is, anxiety over writing new code has a closer link to code quality than you’re making out here, unless all the engineering issues are trivial.

      • houseboatonstyxb says:

        @ Scott
        Contrast this with something like farming (or like my job!) where you show up at a certain time, there’s obvious stuff to do in a nice methodical linear way, and you do it.

        Farming farming farm Er, I suggest using a different example.

      • Nornagest says:

        I am a software engineer in a difficult subfield, and I think the linear, methodical parts of the job actually make me a lot more miserable than the hard, abstract ones. If I’m doing intellectual work, I need stuff to chew on; if all I have to do is setting up lab space or knocking out the stupid and obvious sort of bug, I quickly get bored and unmotivated. There is a sweet spot, though: if I feel like I’m not making any progress on something for an extended period of time (especially easy to do on something like Aapje’s heisenbugs), I get frustrated. But as long as I’m making at least a little progress, then the harder the better.

        I actually really like manual labor too, but it is a different kind of liking.

      • Brad says:

        Since others are pushing back against the software claim, I’m going to jump in and say that how you describe it pretty well describes my experience.

        It’s generally pretty wide open what to work on next and there’s a big disconnect between what is difficult to accomplish and how accomplishing it is perceived, by not just my boss but also by customers and the general public. These two things help to make it difficult to be consistently productive.

      • Machina ex Deus says:

        @Scott:
        OK, I think I got the lingo wrong. I meant to include everyone diagnosed with depression at some point.

        How do you know leaving stuff on your table isn’t a (minor) lingering symptom of depression? Even now that I’m doing well, I still put things off (e.g. disputing a parking ticket, then paying it once it was too late to appeal, then calling the collections agency they sent it to…). It might be linked to that whole gotta-save-energy compulsion. And I’m also late to work a lot (twice a week or so).

        Part of why I asked was so I can tell if I’m normal yet. When I started on my first medication, I thought, “Woah: so this is what it’s like to be normal!” Years later, when I started on my second, I thought: “Double-Woah! So this is what it’s like to be normal (maybe)!”

        I think I have a lot of opportunities that would not be hard to pursue, but which I’m not pursuing.

    • Aapje says:

      I’m not depressed. I had one episode of strong unhappiness during high school when my mental state mismatched strongly with my environment for some time, before I adapted. I am unsure of whether this counts as depression (I thought of suicide in the abstract a little, but not as a realistic option).

      I have an issue where the things I want to do are often not the things I am supposed to be doing. AFAIK this is different from depression where one really doesn’t want to do much. I want to do a lot, just not the ‘right’ things, per se. I think that many/most/all people suffer from the issue that their brains tell them to do A, rather than B, where society/boss/partner demands B.

      I think that there is no meaning to life but to seek fleeting moments of satisfaction. I find satisfaction in stupid things and slightly less stupid things. I’m uncertain whether this is happiness or lack of unhappiness. I’m not sure if there is a difference.

      I’m Dutch.

      It can’t hurt to include your country, since this stuff seems to be viewed differently different places (e.g. in the U.S. I’m depressed, but I’d make a pretty upbeat Dane, and a downright manic Finn).

      I find this confusing, as Danes consistently self-report higher happiness than the US across surveys, no matter what self-reported measure of happiness you use (life satisfaction, positive experiences, whether they think they are thriving). So if you are depressed in the US, you ought to be even less happy when compared to the average Dane.

    • Fossegrimen says:

      I don’t talk about this often as people tend to find it annoying, but hey, you asked….

      I wake up every morning filled with enthusiasm at the prospect of a brand new glorious day. When my first startup went spectacularly bust, I woke up on a parch bench filled with the same enthusiasm for a few months. When my son died, I had maybe two days of feeling less-than glorious, but I don’t think the needle ever tipped over to ‘unhappy’.

      If we define sane as well-calibrated, I’m probably as sick as the severely depressed people, but I don’t really want it fixed.

      When it comes to getting things done, I run two small startups and a consulting gig and still have a few minutes left over for SSC, exercise or a couple hobbies. The consulting gig is precisely the kind of unspecified work that Scott talks about above and I love every minute.

      I do have a tendency to overcommit and then work 140-hour weeks until I catch up and also to take unnecessary risks (both financially and personal) because I tend to drastically underestimate downsides. I don’t see this as problems, but then I don’t count many things as problematic so whatever.

      Until I learned to quiet down, people found me to be rather annoying. Now I manage to come off as unflappable most of the time.

      • herbert herberson says:

        That’s actually really facinating–both the experiences themselves and the way in which you (naturally enough) never talk about them. I wonder how many famous/successful people out there are opposite-depressed on the downlow.

    • Eltargrim says:

      I don’t think I’m depressed.

      I’m in one of those fuzzy jobs Scott described, and I can definitely find myself struggling with motivation and keeping regular hours, because everything is very abstracted from reward and consequence. I’ve adopted a few motivational strategies (pomodoro, Seinfeld habit-building) that are having varying levels of success on various parts of my life (good at getting me to work on time, less good at managing my willpower re: losing weight).

      During a particularly stressful time in my undergraduate degree I had a some intrusive thoughts along the lines of “I should get hit by a car. If I get hit by a car I don’t need to write exams.” I’m no stranger to intrusive thoughts, but this was different in being less abstract, and it felt much closer to being realized. I did not put myself in a position to get hit by a car, and proceeded to strategically let an assignment slip to get more sleep. It worked out well in the end.

      My life isn’t all sunshine and rainbows, but things that depress my mood tend to have a clear cause/effect relationship, and it’s rarely a persistent thing.

      I can always do the things I enjoy doing, but sometimes I’m struck by paralysis of choice. I understand that this is very different from lacking the motivation to do anything, whether you enjoy it or not.

    • I don’t think I have a depression. For the last 6 years or so I was very happy, had friends, well working relationship, generally easy life of an undergrad student who can pass most courses with very low effort.

      My life satisfaction took a large drop over the last 1.5 year to the point where I sort of considered suicide even though I knew this would have been a disproportionate reaction to my problems. Two major things changed.

      I started CS PhD which does put me under a lot more stress. The uneasiness is mostly due to uncertainty of what will happen next, whether my ideas are worth pursuing and whether I am putting enough work into it. These feelings are not very frequent and I can usually deal with them by thinking about the problem rationally. I can force my self to another couple strong pushes that put me in the green zone where I can feel safe again.

      Another thing that had a far more devastating effect was an attempt at polyamory with my long-term partner. I always knew I would be fine managing love for a couple of people. I also expected sharing my girlfriend with others will be somewhat distressing. What I found very overwhelming was the pain of not being able to find a partner that would be up to my standards, while my girlfriend was meeting couple guys at the same time and was getting lots of attention on dating websites, parties, etc. Whenever she would be meeting someone and I would be staying at home (because I couldn’t find anyone) I’d feel completely devoid of social status and unable to do useful work for next couple of days. Initially, I thought I will be able to overpower this feelings with reason and not be jealous. I don’t think this is true and now I’m backpedalling from polyamory.

      I’m fairly confident I’d be happy with my girlfriend meeting other people if I was able to find someone. If it’s just me paying all the costs and she’s getting all the rewards I feel too bad about the arrangement to let it continue.

      (We were continually discussing changes in our relationship and, no worries, my girlfriend tries to help me as well as she can.)

      So I guess even if you’re life is mostly ok, it may be easy to turn it into misery if something important but difficult notice is taken away. In my case it was loss of status that I considered diminished when she would kiss another guy at a party we went to. It’s not exactly tied to my feeling of self-worth. I knew that I can’t find any new partner because my acceptance window is extremely narrow, not because I worthless or something.

    • rlms says:

      I’m fairly sure I’m not depressed. I do have akrasia-type problem, am not discouraged by lack of meaning (I had an existential crisis about that around the age of 7, but that seems to have got it out of my system), and am frequently happy (such as when playing an instrument, having good social interactions, or achieving something). I’m from and living in the UK.

    • Edward Scizorhands says:

      I’m not depressed. The medicine sees to that.

    • Wrong Species says:

      How does someone know when they are no longer depressed? I have never been a happy person but I’m not down on myself like I used to be.

    • Randy M says:

      I live in California, USA.
      I’m not depressed. I’m employed, in good health and enjoy being around my family. That’s a pretty rare trifecta, so I’d be rather ungrateful if I didn’t appreciate my position.

      I don’t worry as much as I probably should. I mean, I’m able to put off deadlines or overlook impending troubles with no more than a vague sense of unease at times, deferring the pain of, say, not turning in a paper or having to do work at the last minute, or having a career that isn’t going anywhere with no real skills or ambition. I blame the fact that it’s worked for me so far (see above) and I have low expectations perspective.

      Being not depressed doesn’t mean I’m filled with motivation to do productive things. I show up for work on time and try to do a good job, but it’s not where my attention naturally goes. And as for my hobbies, I have a dispiriting track record of quickly abandoned novels and scraps of games planned out

      In general I’d say I’m too easily satisfied with life, which is great for being happy and so on, less so for having the motivation for improving my position and getting security.

      As for meaning, I get that through my family and religion. Every once in a while I can force a perspective flip where I think “I’m just one unremarkable little bug on a rock chasing the most stimulation for the least effort for a few years, then oblivion, what’s the point?” Most of the time I feel happy and worthwhile, though.

    • Deiseach says:

      I’m pretty sure there are at least three or four SSC readers who don’t have depression.

      That many? 🙂

      If we’re counting in all the formally diagnosed ADHD/on the Autism Spectrum/Other/not formally diagnosed but strongly suspect, how many of us are mundanes? There are actual happy non-medicated productive people spending time on here?

      • andrewflicker says:

        I’m generally happy, take caffeine on work-mornings and alcohol when I feel like it offduty, but no other mind-altering drugs or any prescribed medicines whatsoever. I’m productive, working more-than-full-time in ecommerce and going to school. So… yes, there do exist “mundanes” like myself around here.

      • There are actual happy non-medicated productive people spending time on here?

        Yes.

        Unless you count blood pressure medication and the like.

      • Nornagest says:

        I am not medicated, other than occasional NSAIDs for exercise-related overindulgence, ethanol for entertainment, and the caffeine dependency common to most engineers. I was diagnosed with ADHD as a child but I’m pretty sure it was a misdiagnosis.

    • J Mann says:

      US respondent: I don’t think I have depression.

      Akrasia: Big time. I struggle with lifehacker/beeminder style interventions, and should probably medicate, but don’t. I make a lot of lists.

      Meaning: I think people can find meaning in their lives, and that that’s inspiring. When all else fails, the meaning in my life is that I can cultivate and experience enjoyment and satisfaction, whether it’s touring Rome, petting a cat, seeing my kid be happy, or finishing a project.

      I think finding meaning in life is a real and substantial challenge, but surmountable and worthwhile for most people. It doesn’t come for free – you’ve got to tend the garden regularly, but the fruits are worthwhile.

    • JayT says:

      I definitely don’t have depression. I’m pretty well satisfied with my life and I always have a general feeling that things will work out well enough. I would say that I am positively happy the vast majority of my life. Times that I am sad or angry tend to happen in short bursts, and I quickly go back to my baseline.

      I definitely do struggle with motivation though, and I think I probably have some combo of ADHD and social anxiety, but it’s not bad enough to really worry me. My life is going well enough, and as I said, I tend to be quite happy, so I don’t see a need to be formally diagnosed or to take medication.

      I’ve never really been concerned about a meaning to life. Making a lasting impression on the world has never really been a driving factor to me, and I get plenty of satisfaction just out of living my life.

      I am in the US.

    • AnonYEmous says:

      Who are you

      It’s ya boi

      what’s it like?

      6/10

      Edit: In particular, do you still have problems with motivation and getting stuff done (akrasia)?

      Yes. In fact, this comment is a direct expression of it. Now I must go do things.

      Do you find yourself discouraged at the apparent lack of meaning in your (and other people’s) life?

      Meh. It’s the kind of thing that gives me pause if I pay attention to it, so I don’t. Besides, my future is technologically based. As are all of yours if you live long enough. So who knows?

      How often are you actually, positively happy, and why?

      not sure, but I always celebrate the occasion by smiling and saying “flex!”. Well, recently. Uh, most recently when I got a 24/25 on my Chemistry quiz which I really didn’t study for that much. Oh, also whenever I hear Donald Trump talk. Don’t ask.

    • Mr Mind says:

      I think my genetic markup makes me pretty optimistic all the time. It’s very rare that I’m super-unhappy. I’m generally, throughout the day, calm and content.
      Unhappiness does happen, but even very big things affects me for a while and then I return to baseline. On a scale from 1 to 10, I’d say a 7 most of the time.
      I do meditate regularly which is able to float me at 7.5 / 8 for up to an hour later.
      I do regularly suffer from akrasia, mostly on stuff like inbox (both paper and electronic), but the lack of meaning in my life is more of a puzzle than a misery: in the end, it’s very clear to me that I can make from it whatever I want, but I’m unable to decide exactly what.
      I’m from Italy, if that means anything.

    • powerfuller says:

      I used to be very depressed, and I can still be prone to it, but I’ve managed more or less stave it off by sort of consciously trying to make myself manic instead; I spend a lot of time pumping myself up to embrace my life as the only life for me, the only life I want. Stop being sad and start getting MAD! If life is suffering, then be a motherfucking sadomasochist! I have to dig ditches all day? Fuck YES I LOVE ditches.

      It’s been more effective than I would have thought possible, and I’m definitely a more capable and positive person on a day-to-day basis. On the other hand, it seems to have atrophied a lot of my softer emotions (for lack of a better term). My appreciation of music is one-tenth what it was when I sadder, and I’ve lost a lot of interest in trying to form really deep connections with other people (in the sense of understanding and being understood). At this point I just strive to be useful to other people when I can, and I go around trying to PUMP people UP(!) and make them laugh. Life is full of trade offs, I guess.

      I’m from the USA! USA! USA!

  2. hoghoghoghoghog says:

    I will describe a form of argument. I’d like to know (a) does it have a name and (ii) is it good? Call it argument by meta-analysis. You are arguing “by meta-analysis” if, instead of directly addressing the question, you start with the most popular or status quo position, analyze for bias the process that led to the most popular or status quo position, and correct for that bias. Examples:

    3) I believe that financial regulation has both costs and benefits. I am not equipped to do an actual cost/benefit analysis. Instead I reason as follows: “financial interests have an outsize influence on the political process. Therefore the status quo is most likely too little regulation. Therefore I support more regulation.”

    IV) I want to know the probability that global warming is mostly cause by humans. I don’t know enough to think about the science directly. Instead I reason as follows: “at this point many careers are motivated by the belief that humans can do something about science change. Therefore the scientific establishment is biased towards believing that humans are the main drivers of climate change. Therefore I believe that humans are the main cause of global warming with only 80% confidence.”

    Pros: This method of argument is very cheap – you only need to know about politics to do it. It also starts from the establishment position, which is likely to have already integrated many perspectives, interests and expert knowledge.

    Cons: Nothing new will ever be discovered by arguing in this mode [*]. It is only a few steps away from ad hominem argument. It gives sophists weird incentives: every sophist who isn’t perfectly aligned with the status quo is incentivized to discredit the political/scientific/whatever decision-making process.

    [*] unless you are actually trying to learn about decision-making processes, in which case you should do so systematically instead of doing so as a byproduct of studying some controversial question.

    • shakeddown says:

      Of the two methods people make decisions on issues they don’t know much about (the other being tribal identification), this seems the better one.

      Also, that indexing scheme is incredibly painful.

      • hoghoghoghoghog says:

        Unfortunately it is compatible with the tribal identification method, since everyone believes that something has gone wrong with the decision-making process of the other tribe, so any decision-making process which incorporates the other tribe is necessarily biased (and in a direction that justifies your ‘correcting’ towards your tribe’s viewpoint).

        • shakeddown says:

          Yeah, I think you’re right. After rethinking, this seems like a decent method *except* on issues that are heavily tribal/political.

        • Mr. Breakfast says:

          Even without applying a correction you describe, tribalism enters through the thinker’s assumptions:

          3) I believe that financial regulation has both costs and benefits. I am not equipped to do an actual cost/benefit analysis. Instead I reason as follows: “financial interests have an outsize influence on the political process. Therefore the status quo is most likely too little regulation. Therefore I support more regulation.”

          From this I gather that you are somewhere over on the left half of the spectrum, assuming as you do that established firms will resist regulation. As a right-anarchist of sorts, I generally assume that established firms favor more and more complicated regulations, especially ones which are scale-insensitive. They do this in order to raise barriers to entry to their industry and to increase minimum overhead costs which drive out smaller competitors.

          Perhaps explicitly stating chains of this sort of reasoning could be more useful as a way to map the assumption sets comprising tribal worldviews than it is in reaching actual conclusions.

          • hoghoghoghoghog says:

            For the record I disagree with both of my examples, and I suspect that this whole mode of argument is really, really bad but I can’t put my finger on why.

            Your second paragraph is an interesting point. This mode is so vulnerable to tribalism that it might best be used to detect tribes.

          • Mr. Breakfast says:

            Or to detect one’s own tribalism; I found your example actually jarring and bizarre the first time I read it. Then I remembered the inference I had forgotten I was making.

    • One problem with the approach is that it requires you to correctly identify both the bias and the direction of its effect. Thus in your first example, you are taking it for granted that regulation is against the interest of financial interests. Regulation can be, sometimes is, used by an industry for its own benefit–the capture theory of regulation. So your argument might imply that there is too much regulation rather than too little.

      • Art Vandelay says:

        I suspect that a lot of what is referred to as “deregulation” of the financial sector is actually a form of regulation in favour of big financial interests. Quite possibly a similar thing happens when governments intervene to create or encourage “free trade” or “free markets”.

    • Philosophisticat says:

      I don’t know the name of this form of argument, but there is nothing wrong with it in general (particular applications aside, as David Friedman notes above). However, it is worth noting that this is a kind of argument whose evidential force diminishes as you get better informed – the more directly bearing information you have, the less impact the meta-consideration has. In particular, this kind of argument is forceful when all you have to go on is other peoples’ opinions.

      • hoghoghoghoghog says:

        Yeah, this form of argument doesn’t take ground reality into account at all. In example (3), It will keep telling you to regulate more as long as banks exist, regardless of what the current level of regulation is. It is probably bad if you keep using it when you become better informed, or if you use it instead of informing yourself.

    • seladore says:

      There’s also the problem that this method (argument from political bias?) seems to ignore the fact that many positions are the status quo for very good reasons, rather than just resulting from bias.

      Contra your point IV, for example:

      I want to know the probability of, say, the germ theory of disease being true. I don’t know enough to think about the science directly. Instead I reason as follows: it would be career suicide for medical professionals to question the germ theory of disease. It’s fair to say that the career of every medical professional is dependent on accepting it. Therefore the medical establishment is strongly biased towards accepting the germ theory of disease, and I should downgrade my belief in it accordingly.

      It seems that your method would lead to a general pattern of downgrading belief in widely-supported, ‘status quo’ positions. And this seems counter-productive; in the majority of cases, the widely-supported position is widely-supported for very good reasons.

      • It seems that your method would lead to a general pattern of downgrading belief in widely-supported, ‘status quo’ positions. And this seems counter-productive; in the majority of cases, the widely-supported position is widely-supported for very good reasons.

        You are downgrading your belief, but downgrading it from a very high level. You started with a probability of .999, based on how likely it was that a hundred people would make the same mistake. After seeing the argument, you lowered it to .99.

        For a real example, consider the probability that everyone has the number of chromosomes in a human cell wrong.

      • Mr. Breakfast says:

        It seems that your method would lead to a general pattern of downgrading belief in widely-supported, ‘status quo’ positions. And this seems counter-productive; in the majority of cases, the widely-supported position is widely-supported for very good reasons.

        I would argue that this downgrading may be a positive in moderation. Looking back through the SSC archives, something felt off about how common are science failures?:

        Suppose there are about fifty scientific fields approximately as important as genetics or psychiatry or psychology. And suppose within the past century, each of them had room for about five paradigms as important as psychoanalysis or behaviorism or Lysenkoism.

        That would mean there are about 250 possibilities for science failure, of which three were actually science failures – for a failure rate of 1.2%.

        My problem is: At what level of wrong is it right to be right about a scientific theory being wrong?

        If Scott concedes that a “typical” scientific field goes through 5 paradigms per century (elsewhere in that post, he basically disqualified scientific paradigms prior to the 20th century, so this really meant 5 paradigms total for all time in his model), doesn’t that mean that at least 4/5 scientific paradigms ever were in some sense wrong? Does a scientific paradigm have to be (in retrospect) laughably wrong in order to be a “science failure”?

        Maybe Scott’s 1.2% estimate is a good approximation of the likelihood that the currently ascendant theory in a given field is horribly corrupt and misleading, but does it reflect the chance that the current consensus is just in some manner unreliable in it’s conclusions?

        So going back to your example, if a layperson understanding that professional experts:
        1) Always required some sort of theoretical framework within which to work and
        2) Progress in a given discipline benefits from the concentration of researchers working within a shared framework/paradigm and
        3) Most of these consensus theories will, over time, be replaced as more is learned

        Might this not reasonably lead that layperson to discount their own confidence in the ultimate correctness of a field’s currently reigning paradigm to a level below that implied my a mere poll of experts’ allegiance in the paradigm?

        So specialists stipulate a theoretical framework in order to get on with their work, and the rest of us remain slightly skeptical of the framework and it’s currently best model so as to minimize our long-run error across all fields in which we are not experts. What is wrong with that?

    • Aapje says:

      @hoghoghoghoghog

      I think that it is perfectly fine way to argue, with the caveat that the main failure mode is cherry picking. Your IV is a good example of this, IMHO. You completely ignore the question of why global warming became a major field of study with so few detractors. You can just as easily argue that phrenology researchers are biased to believing that skull measurements are predictive or that cold fusion researchers are biased to believing that cold fusion is real. Yet both fields failed to (ultimately) be accepted by the scientific establishment.

      You can also make a strong argument the other way: it is highly beneficial to the human race to not be hampered in their activities, so you’d expect that there would be a bias towards disbelieving pro-global warming. So then you’d want to correct the confidence levels of scientists upwards, rather than downwards.

      In general, all logical arguments are imperfect one way or another and merely allow for hypotheses that need to be checked with actual measurements/experiments.

    • S_J says:

      Totally aside from your main thread of discussion:

      I really enjoy the numbering scheme you used for the successive points in your post.

      It is (1) intriguing, (B) rule-bending, and (iii) slightly confusing but in a way that causes me to re-read, thus increasing my desire to clearly understand your sequence of thought.

    • S_J says:

      About point (3):

      if financial interests have an outsized influence on the political process, by what means do you hope the political process that leads to regulation will actually reduce this influence?

      • hoghoghoghoghog says:

        I don’t think there’s a serious paradox here: at any moment there are lots of people pulling against whatever direction they see the bias in, which should be enough to counteract first order bias. (Of course then there is second-order bias: “the common man takes comfort in a belief that big financial institutions are against him rather than his own failures, therefore he is biased to believe the government is biased towards high finance, therefore I favor less regulation.”)

    • Wrong Species says:

      It’s a terrible argument for one reason: everyone is biased. Anyone who doesn’t think they are isn’t enlightened, they just lack self-awareness. So for the global warming example, there are both people who have a vested interest in mitigating CO2 and those who don’t. Are we going to quantify their bias? How would we do so? And even if we could, that still doesn’t answer the question about what we should do about global warming. Anyone who uses this as an argument in a debate is no better than you’re typical Facebook sloganeer.

      • hoghoghoghoghog says:

        It might conceivably be easier to quantify bias than to quantify a climate model, depending on your skillset.

    • J Mann says:

      I’m pretty skeptical that ontological contrarianism will produce much of value.

      I. It depend on you determining the direction and amount of bias in the system. Since we’ve granted as a premise that we don’t understand the subject matter well enough to just test the opposing cases, I’m not sure how we can be confident in our assessment of the direction and degree of bias.

      Take financial regulation: are the regulations biased towards (a) under-regulation; (b) regulations that protect existing and larger firms against competition from new entrants; (c) regulations proposed by motivated activists who don’t understand the details of the market, and therefore do things that sound good but cause more net harm; (d) responses to specific incidents that again, do more harm than good? Probably some of all of the above, so if you see a new regulation proposed, should you support it because of (a) or oppose it because of (b)-(d)? If you’re not confident in your ability to understand how the regulation will affect the market, it’s hard to be confident in your assessment of which bias passing or rejecting this regulation would address.

      Second, it’s pretty hard to come up with a good contrarian filter. Take gambling or investment – if you could come up with a good bias at horse races or stocks, you could consistently make money over time by betting against it, but when people study contrarian investment strategies, they don’t seem to do any better than anything else. (Markets may correct for bias faster than otherwise.)

      c) On the other hand, if you’re aware of a bias, it may feed into your confidence level. We intentionally bias the criminal justice system against a finding of guilt. As such, if someone is acquitted, that’s not as strong evidence that they’re actually innocent than if they were found not guilty by a similar but less biased process. (On the other hand, since the process is biased, a finding of guilt is stronger evidence of guilt than in our biased process.)

      If we presuppose that we’re unable to assess the actual evidence, however, it’s hard to say what to do with this knowledge. Should we treat someone charged and acquitted as if they are more likely to be guilty? Maybe, but that seems unfair without at least considering the evidence.

      Finally, I’m not a fan of the results of this method in everyday life. If I don’t understand advanced statistics, I can’t really understand a fight between dueling research academics over global warming or charter school performance or whatever. If I substitute “find the biased side” as my heuristic, then you end up with vicious fights about bias. Can I rely on facts reported in the NYT? On Fox News? Did researcher so and so accept a grant from Tom Steyer? From the Koch Brothers? Pretty soon you’re in a nasty fight about motivations, and you’re not closer to the truth.

      • hoghoghoghoghog says:

        Many good points here, but indeed your last paragraph is [a better description than I could come up with of] what got me thinking about this. This method is attractive, common and maybe even accurate, but it’s very likely to poison the waters in just the way you describe.

    • nimim.k.m. says:

      The fundamental idea feels sound. Others have already mentioned various pitfalls.

      My main gripe is, however, how did you came up with number 80% in (IV)?

      • hoghoghoghoghog says:

        The literal answer is “90% as establishment figure, minus 10% because that is the smallest nonzero number that is a multiple of 10%.” In general if you wanted to do this you do need to be able to both pick out bias and guess how much that bias effects the establishment figure, which does seem hard.

    • Aftagley says:

      I don’t think this actually reduces the complexity of arguing, it just shifts the burden of research away from the topic and towards the identification of biases. I know your examples are deliberate simplifications, and you aren’t trying to hold them up representative cases of this mode of arguing, but lets look at your financial regulation one anyway.

      The bias you identify is potentially valid, but so is “It’s easy and expedient for politicians to demonize ‘wall street’ which leads to unnecessary amounts of regulation” and that makes you arrive at the entirely opposite conclusion. To be logical in this, you have to weigh every bias against each other and that’s got to be at least as difficult as just learning about the issue.

  3. ThirteenthLetter says:

    So I was listening to a podcast where the concept “the true dystopia we’re heading for isn’t 1984, it’s Brave New World” was discussed: people will be made passive with internet porn, sex robots, universal basic income, legal weed, Game of Thrones marathons, et cetera, while the government does what it pleases. And it’s the sort of thing that certainly sounds plausible. But thinking about it, I’m not so sure.

    Fundamentally, the Brave New World theory suffers from the same problem as M*ldbug’s theory about authoritarianism. To recap, he suggests that a sufficiently powerful authoritarian government wouldn’t have to restrict what its citizens were allowed to say, because it wouldn’t have to care what its citizens had to say (because the citizens aren’t strong enough to harm it in any way.) This is plausible on one level, but it fails to account for an aspect of human nature: namely, the authoritarian government is made of people who will get upset if the peasants are disrespecting them, and will crack down even though it’s not necessary.

    Similarly, the Brave New World government might not need to pay any attention to its citizens, because its citizens are too busy spending their UBI on Game of Thrones marathons. But again, human nature gets in the way. Modern social justice, which is just the latest variation of the fundamental urge that humans have to get all up in each other’s business and tell them what to do, suggests this is unstable, because the people who make up the Brave New World government are going to get upset at what some of the soma-tized humans do. Are they buying sexbots? Oh my god that’s so gross and problematic and objectifying, they have to be stopped. And we’re back to the crackdown.

    So ultimately, we end up back with the boot stomping on a human face, although only until the human face gets pissed off enough and there’s a destructive revolution, but still. Non-brutal authoritarianism, whether M*ldbuggian or Huxleyan, is a fundamentally unstable state due to the human nature of the rulers. Plausible?

    • shakeddown says:

      This seems like a good counterargument, but I’d expect this world to be closer to what we have now than violent revolution.

    • cassander says:

      This is plausible on one level, but it fails to account for an aspect of human nature: namely, the authoritarian government is made of people who will get upset if the peasants are disrespecting them, and will crack down even though it’s not necessary.

      The moldbuggian argument, formally, is that governments are autocratic, but owned by shareholders, who will object if profits go down because their god-emperor CEO is abusing the peasants for not venerating him sufficiently.

      • ThirteenthLetter says:

        That just moves the lump in the carpet around — now it’s the shareholders who will get upset at the peasants for not properly venerating them, and if profits go down due to all the repression that’s just a further sign of how rotten the peasants are and that they need to be brought into line.

        (In a greatly simplified fashion, that’s how a lot of colonialist regimes fell in the real world. This also suggests that outside agitators would have a relatively easy time pushing the Moldbuggian/Huxleyan states off their axis.)

        The only way either of these states could survive is if they were run by someone or something that genuinely didn’t care [what the peasants thought of them]/[what sort of gross problematic objectifying things the citizens were doing with their sexbots] — and that has to extend to all the levels of enforcement and bureaucracy. You kind of have to go with robots or aliens at that point.

        • aNeopuritan says:

          Name 3 examples, by different colonizers.

        • cassander says:

          >now it’s the shareholders who will get upset at the peasants for not properly venerating them

          the shareholders didn’t buy the stock to get veneration, they bought it to make money. And the vast majority didn’t actually buy the company in question, they bought some fund or other because it had high returns. They don’t know what’s inside it and don’t care. When CEOs get kicked out for political incorrectness, it’s almost never the shareholders forcing the guy out.

          In a greatly simplified fashion, that’s how a lot of colonialist regimes fell in the real world.

          When colonies were run by the great trading companies, they tended to act in exactly the fashion I describe. They only got moralizing once governments took over them more directly.

          The only way either of these states could survive is if they were run by someone or something that genuinely didn’t care [what the peasants thought of them]/[what sort of gross problematic objectifying things the citizens were doing with their sexbots] — and that has to extend to all the levels of enforcement and bureaucracy. You kind of have to go with robots or aliens at that point.

          Or a corporation that cares about the bottom line.

          • rlms says:

            “Or a corporation that cares about the bottom line.”
            Corporations are made of and controlled by people. A corollary to “people don’t become dispassionate rational decision makers just because you say they’re part of a government” is that the same applies if they’re part of a company.

          • cassander says:

            @rlms says:

            >Corporations are made of and controlled by people. A corollary to “people don’t become dispassionate rational decision makers just because you say they’re part of a government” is that the same applies if they’re part of a company.

            The difference is that companies that lose money face far stronger and more direct feedback than government. The average civil servant is no more or less rational a decision maker than the average corporate employee, the difference is that the latter’s company is far more directly and immediately punished if they make bad decisions.

          • rlms says:

            How strong is the feedback given to the average employee of a big company? I agree that a sole proprietor of a market stall gets better feedback than a civil servant, but changes in a supermarket shelf stacker’s company’s fortunes only affect them if it gets really close to bankruptcy. You could argue that CEOs experience more feedback than politicians (since many politicians have safe seats and only experience feedback if they really mess up, whereas CEOs are sensitive to small changes in company profit). But I don’t think the latter part is actually true; CEO pay seems pretty uncorrelated with company performance (I think you agreed with this in another thread).

          • John Schilling says:

            The difference is that companies that lose money face far stronger and more direct feedback than government.

            Cite, please? Particularly if you are talking about feedback that reaches the people who actually make the decisions. I’d wager that CEOs as a class enjoy greater incumbency than Presidents and Prime Ministers, that corporate VPs are less likely to be fired for the underperformance of their departments than cabinet secretaries. But this should be amenable to analysis, at least for publicly-traded corporations.

          • cassander says:

            @rlms and John

            You’re both assuming I’m arguing that the feedback is directed upon individual employees, who then change their behavior. That’s not it at all. the feedback is directed at the company as an institution. The best metaphor for it is to imagine companies as boats circling a whirlpool in a fog. They can’t be exactly sure which way to sail, but they’ll all getting drawn in slowly, so the only thing they can do is to strike out in one direction or another. Many will sink, i.e. sell the wrong product to the wrong people at the wrong price, but the few that manage to sail in the right direction (produce the results people want) will survive. The model works even if you assume the behavior of individual companies and their employees is completely random, something that can’t be said of centralized government.

          • John Schilling says:

            It’s the people who make the decisions, so if the feedback doesn’t reach them, what good is it? Some corporations will survive, some will not, but it will be the same sort of people making the same sort of decisions with the same incentives at all of them. Often literally the same people, now that we’ve given up on the norm of working for one corporation for one’s entire career.

          • cassander says:

            @John Schilling says:

            It’s the people who make the decisions, so if the feedback doesn’t reach them, what good is it? Some corporations will survive, some will not, but it will be the same sort of people making the same sort of decisions with the same incentives at all of them.

            Yep, and the ones that fail will keep getting knocked down, and the ones that succeed will continue to stay up. If you assume tastes and institutions never change, then eventually everyone ends up in a company that’s doing something people want. Of course, that assumption is nonsense, but that just means that the result is a constant striving for an equilibrium than never comes. the churn continues forever.

        • Tekhno says:

          The only way either of these states could survive is if they were run by someone or something that genuinely didn’t care [what the peasants thought of them]/[what sort of gross problematic objectifying things the citizens were doing with their sexbots] — and that has to extend to all the levels of enforcement and bureaucracy.

          So you need to put a libertarian in charge of the authoritarian state?

          • cassander says:

            as has been said, “Come, join the libertarian conspiracy! We’re going to take over the world…..then leave everyone alone!”

      • dndnrsn says:

        Does this match up with the reality of corporations? Are CEOs punished for making decisions that benefit the CEO materially, or flatter them, at the expenses of the corporations?

        • cassander says:

          CEOs are rarely punished for anything. But a company that consistently chooses CEOs that self indulge will rapidly be out-competed by companies that don’t. Markets are darwinian.

          • dndnrsn says:

            How does this explain the rise in CEO pay, which seems to happen regardless of good times or bad?

          • cassander says:

            the cost of CEO pay is often essentially trivial compared to the size of the company as a whole so selection pressure on that one issue is weak.

          • Chalid says:

            CEOs that self indulge will rapidly be out-competed

            citation needed on “rapidly”

          • dndnrsn says:

            It would seem to contradict the idea that CEOs who don’t indulge themselves would see their companies out-compete those who do, though.

      • Dr Dealgood says:

        One interesting question that Moldbug and the Death Eaters never seemed to adequately address is that rule by joint stock companies isn’t an untested experiment. There were actually a number of such companies in recent history.

        The British East India Company and Cecil Rhodes’ British South Africa Company, the Dutch East India Company and the infamous Belgian International Association of of the Congo all performed the function of states while ultimately being for-profit corporations responsible to shareholders. While only the last on the list was nominally independent, in practice all of them had substantial armed forces and operated largely independently of their national governments.

        Rather than arguing about abstracts, why not examine the quality of government in those areas? Did company rule provide better governance than direct rule by the suzerain state or subsequent postcolonial governments?

    • Machina ex Deus says:

      Imagine 80,000 pages piled on a human face, forever.

      (That’s about 4 meters, or 13 feet.)

    • cthor says:

      I don’t think SJW’s are as powerful as you think they are. The entire culture wars thing is a working class conflict.

      Essentially, your counter-point is that the ruling class will be stupid, petty, insecure, or malevolent. When compared to the baseline population, they’re typically none of these things. The thing powerful people are most likely to do is whatever they think will maintain their power, and needlessly harassing successfully soma-tised serfs is the opposite of that.

    • Jiro says:

      You could argue that China in real life is much of the way towards being such a government. Not completely–trying to do a large scale protest will get you in trouble–but the censorship is leaky and people just don’t care enough to go around it, it’s officially Communist but has in practice given up ideology and it doesn’t really have either Western religion or social justice to create moral busybodies.

    • onyomi says:

      The Brave New World dystopia still seems most likely to me because I tend to think that politicians and businesses both will mostly leave you alone so long as you are giving them your votes/dollars. That is, they don’t like to offend “clients.”

      The dystopia I see as most likely, given current trends, is one in which it is incredibly hard to break into the ranks of the productive/decision-making classes, because everything is so heavily regulated to favor existing major players, but it is correspondingly much less materially uncomfortable to be a dependent non-entity: you keep voting for the politicians who send you the UBI which you spend at McDonalds, Amazon, and the other major designated providers of food, clothing, shelter, and entertainment. They use the money to donate to the politicians and lobby for maintaining/increasing the regulation. Both established politicians and companies work out an equilibrium where they have little to fear from upstart competition due to the steady streams of votes/money they have locked in so securely.

      Busybodies can fit pretty neatly into this equation because they are the “Baptists” to politicians’ and corporations’ “bootleggers.” In fact, they may be necessary, because they are the ones who can be counted on to believe the sky will fall if new Draconian regulations aren’t kept in place and, ideally, continually added to. On some level, it may not even matter that much what they busy themselves with, so long as it can provide a pretext to regulate something people like, which, in turn, provides the opportunity for regulatory capture and continued vote buying.

      • Tekhno says:

        The Brave New World dystopia still seems most likely to me because I tend to think that politicians and businesses both will mostly leave you alone so long as you are giving them your votes/dollars. That is, they don’t like to offend “clients.”

        When there are very large entities commanding a large market share (for reasons of gov regulation or otherwise), rather than niche sections, the clients they are competing for essentially describe an average ideal of humanity. This means that they will leave the average person alone, but will crack down hard on the odd ones out who might do things to offend the average person/clients.

        This is probably why social media is becoming so concerned about sanitizing and cleaning up offensive power-users that might drive “normies” into the arms of some competitor. They mostly maintain their large share through the network effect, but the technical barriers to entry are low, so they can command a near monopoly for a long time and then suddenly collapse to be replaced as the “next big thing” (myspace usurped livejournal, facebook usurped myspace).

        In the pursuit of the average, they tend to want to create an environment that is bland and inoffensive in style. Youtube originally had the ability to completely custom the art for your channel, but then they brought in the “one channel” design and paired it back to a banner at the top, and mostly homogenized channel style. That’s possibly a small example of this.

        So, if you expand that idea into a totalitarian form of governance, it would be Brave New World for the majority/average, 1984 for the eccentrics that break with the programming.

    • Dr Dealgood says:

      Eh?

      I’m not sure I buy the concept of a Left Singularity, and equally unsure about the idea that you can actually “wake the sleeping giant” in the sense of pissing people off enough to spontaneously rebel. When you look at historical revolts or repression in detail it doesn’t give the impression of impersonal historical-material forces so much as men with outsize influence making poor decisions.

      That said, we seem to already be in a Brave New World scenario for the most part. The parts which are contradicted by what we understand of human nature (perfect behavioral conditioning, stable world government, prosperous command economy, etc) haven’t happened for clear reasons. But in terms of overall philosophy and aesthetics we’re pretty damn close.

    • rlms says:

      Related: what actually makes Brave New World a dystopia? I found it interesting when I read it that the World Controller’s speech in favour of his society was quite persuasive. The main dystopian elements I remember were cultural stagnation, and genetic engineering to keep some people from achieving their “natural” potential. But imagine a real society that doesn’t use genetic engineering that way. Without government surveillance, I can’t see total cultural stagnation happening; if past art isn’t censored, some smart and curious people will inevitably look at it and create their own art based on it. I feel like trading an amount of personal freedom and engagement in “worthwhile” culture in favour of wealth and stability might be worth it.

      • Dr Dealgood says:

        It depends on what you see as the purpose of life.

        If life is all about pleasure and self-satisfaction, BNW is a utopia. All of your base desires for sex, food and games are satisfied. And the society works very hard to ensure that people avoid desires like ambition and curiosity which cannot be satisfied.

        If life is about something greater, though, BNW is a dystopia. All of the meaningless pleasures it heaps on you are distractions to keep you inside the gilded cage. There is no self-overcoming or real growth, no ability to improve your position or to discover new truths about the world. It’s a locked cell with a low ceiling and no windows.

        (This is where people go “You monster! Nobody is starving or dying of malaria in the BNW! Maslow’s hierarchy!” Which would be an arguable point, except that it’s comparing fiction to reality. BNW explicitly tamps down on scientific progress as much as possible and runs a top-down command economy. If any real society ran on Huxley’s blueprint it would make the USSR look like Disneyland.)

      • The Nybbler says:

        There’s your mass-produced deliberately stunted Epsilons, for one thing; it’s more than just a few people. Less is said about the Gammas and Deltas. One conceit of the book is that most people in every caste think their caste has it the best, but I think it’s fair for the reader to disagree.

        Replace the three lower castes with (non-general-AI) robots and you have a tougher question.

        • Jiro says:

          That’s probably what rims was referring to by genetic engineering.

          Brave New World doesn’t actually have genetic engineering, because of when it was written, but that’s basically the sort of thing that would have been done by genetic engineering if it had been written any later, so I can see where the misunderstanding comes from.

      • Deiseach says:

        genetic engineering to keep some people from achieving their “natural” potential

        Not just “some”: five stratified castes, with conditioning and in-vitro environmental controls to make sure the Gammas to Epsilons are stunted intellectually and physically; post-decanting (no natural births, it’s all in-vitro and incubators, with mass cloning of the Gamma to Epsilon castes) conditioning by sleep learning methods and Pavlovian conditioning with electric shocks to make sure each caste learns on an imprinted level what is their place in society; 70% of females decanted are sterile to ensure no uncontrolled reproduction; people who are talented can’t use their talents if it conflicts with what is deemed necessary for the job; exile for those who can’t or won’t conform:

        Still leaning against the incubators he gave them, while the pencils scurried illegibly across the pages, a brief description of the modern fertilizing process; spoke first, of course, of its surgical introduction- “the operation undergone voluntarily for the good of Society, not to mention the fact that it carries a bonus amounting to six months’ salary”; continued with some account of the technique for preserving the excised ovary alive and actively developing; passed on to a consideration of optimum temperature, salinity, viscosity; referred to the liquor in which the detached and ripened eggs were kept; and, leading his charges to the work tables, actually showed them how this liquor was drawn off from the test-tubes; how it was let out drop by drop onto the specially warmed slides of the microscopes; how the eggs which it contained were inspected for abnormalities, counted and transferred to a porous receptacle; how (and he now took them to watch the operation) this receptacle was immersed in a warm bouillon containing free-swimming spermatozoa – at a minimum concentration of one hundred thousand per cubic centimetre, he insisted; and how, after ten minutes, the container was lifted out of the liquor and its contents re-examined; how, if any of the eggs remained unfertilized, it was again immersed, and, if necessary, yet again; how the fertilized ova went back to the incubators; where the Alphas and Betas remained until definitely bottled; while the Gammas, Deltas and Epsilons were brought out again, after only thirty-six hours, to undergo Bokanovsky’s Process.

        Even the Alphas are a product of the hatcheries and conditioning; they have the greatest degree of freedom within the society, but they are still bounded by laws and rules:

        Straight from the horse’s mouth. It was a rare privilege. The D. H. C. for Central London always made a point of personally conducting his new students round the various departments.

        “Just to give you a general idea,” he would explain to them. For of course some sort of general idea they must have, if they were to do their work intelligently – though as little of one, if they were to be good and happy members of society, as possible. For particulars, as every one knows, make for virture and happiness; generalities are intellectually necessary evils. Not philosophers but fretsawyers and stamp collectors compose the backbone of society.

        “To-morrow,” he would add, smiling at them with a slightly menacing geniality, “you’ll be settling down to serious work. You won’t have time for generalities. Meanwhile…” Meanwhile, it was a privilege. Straight from the horse’s mouth into the notebook. The boys scribbled like mad.

      • nimim.k.m. says:

        That the future will be more-or-less like it, and most people living in it will consider it non-dystopic in general and improvement over the historical conditions, in particular.

        After all, if you ignore the state that actually bothers with eugenics, blatant indoctrination and exiling non-conformists, it was quite scary thing for me to read: it felt surprisingly accurate account of the mental state of the society we live in. Just replace world controllers with Moloch.

        Or what Dr Dealgood said.

      • sohois says:

        If you were to rank possible future world states, Brave New World would clearly be in the top quartile; it’s a hell of a lot better than various apocalyptic scenarios, it outranks authoritarian nightmares such as its contemporary 1984, and I would argue that it is a better world than our own. I’d imagine a lot would disagree with that last point, but that might be because the majority of readers in this blog do have it better than what BNW might provide. For the billions that suffer from poverty or death from disease or other hardship, BNW represents an obvious improvement and so on average BNW is an improvement compared to today.

        It’s essentially just a mediocre wireheading society, but at the same time with some freedoms extended to those such as Helmholtz who don’t wish to wirehead themselves with soma.

        Of course, as others have pointed out it is so much easier to do better. The Nybblers suggestion of replacing lower castes with robots would undo a great deal of the larger problems in a snap.

    • dndnrsn says:

      You don’t even have to go to “the rulers are human, and will get pissy about the behaviour of the commoners.” There’s never been a state with enough authority to say “I don’t care what you do because you can’t possibly harm me, so just pony up the gold on time” and it’s unlikely that’s ever going to be the case. Fnargl is a thought experiment. I can’t remember if Voldemort extrapolates from the thought experiment.

      I also think you’re wrong to say that the social justice types are the rulers. They’re not. If they were seriously going after the rich and powerful in a threatening way, they’d get slapped down hard.

    • Deiseach says:

      (T)he people who make up the Brave New World government are going to get upset at what some of the soma-tized humans do. Are they buying sexbots? Oh my god that’s so gross and problematic and objectifying, they have to be stopped.

      Aside from the fact that the World Government wants people to buy and consume stuff, that this is what society is built upon –

      Patiently the D.H.C. explained. If the children were made to scream at the sight of a rose, that was on grounds of high economic policy. Not so very long ago (a century or thereabouts), Gammas, Deltas, even Epsilons, had been conditioned to like flowers – flowers in particular and wild nature in general. The idea was to make them want to be going out into the country at every available opportunity, and so compel them to consume transport.

      “And didn’t they consume transport?” asked the student.

      “Quite a lot,” the D.H.C. replied. “But nothing else.”

      Primroses and landscapes, he pointed out, have one grave defect: they are gratuitous. A love of nature keeps no factories busy. It was decided to abolish the love of nature, at any rate among the lower classes; to abolish the love of nature, but not the tendency to consume transport. For of course it was essential that they should keep on going to the country, even though they hated it. The problem was to find an economically sounder reason for consuming transport than a mere affection for primroses and landscapes. It was duly found.

      “We condition the masses to hate the country,” concluded the Director. “But simultaneously we condition them to love all country sports. At the same time, we see to it that all country sports shall entail the use of elaborate apparatus. So that they consume manufactured articles as well as transport. Hence those electric shocks.”

      “I see,” said the student, and was silent, lost in admiration.

      As I recall, “Brave New World” took care to inoculate against such attitudes; anyone who wanted to interfere with others’ pursuit of (approved) pleasures was subject to disapproval (and maybe re-conditioning; I can’t remember).

      Those who insisted on swimming against the tide of society were exiled to islands (Iceland and the Falklands are mentioned) and there are reservations where the Savages (tribal cultures) live traditional lifestyles overseen by wardens and treated as holiday destinations for the citizens of the World Society. Someone who wanted to ban sexbots would at first be a novelty and a sensation (as John, with his attitudes towards sex and romance in the novel is considered) but then, if troublesome, dealt with.

      • beleester says:

        Not related to the OP, but I feel like this doesn’t make any economic sense. The economy exists to provide people with things they want. If they want trips to the country, more of the economy will end up devoted to trips. If they want sports, more of the economy will end up devoted to sports. But the market as a whole works the same either way.

        In fact, wouldn’t it be better, economically, to make people want flowers instead of sporting goods? If you can keep people happy for the cost of a flower, then you can close up the sporting goods factory and use those resources for something else. Maybe build a car factory, so that people can go to the country more easily.

        This feels like a society-wide example of the Broken Window Fallacy – making people consume sports equipment instead of fast cars isn’t going to drive the economy faster, any more than making people spend money on window repairs would.

        • Nyx says:

          I believe it’s expounded upon that the BNW is so abundant and efficient then they could easily reduce the amount of hours that everyone worked, but experiments showed that the workers just spent their free time blissed out on Soma and weren’t essentially happier or more stable for it. Most of the work in BNW is make-work to keep people busy (just as the EGD classes are deliberately handicapped).

          • Dissonant Cognizance says:

            The book does cover that, but doesn’t then address why they need those people in the first place. Their society has complete control over supply, demand, and labor, so if there’s less work to do than people to do it, why not just decant fewer factory workers?

            I assume there’s some bureaucratic inertia that values maximizing population regardless of caste, in keeping with the general underlying horror of a complete economy that serves no one.

    • J Mann says:

      That’s the counterargument to BNW – that even if the citizens are watching virtual porn and taking drugs, they’ll still want more resources than they are receiving, so you get conflict, and eventually the system isn’t sustainable.

      IMHO, Brave New World works if people believe that changing the system would result in ending the gravy train. Not sure how that will work out in real life.

      • Matt M says:

        if people believe that changing the system would result in ending the gravy train

        How is this NOT a reasonably accurate description of the modern welfare state?

        • J Mann says:

          I think it’s undetermined whether the system breaks because people vote themselves more benefits than the system can pay (US) and/or riot for them (Europe).

    • beleester says:

      the people who make up the Brave New World government are going to get upset at what some of the soma-tized humans do. Are they buying sexbots? Oh my god that’s so gross and problematic and objectifying, they have to be stopped. And we’re back to the crackdown.

      It seems like The Powers That Be in this scenario don’t actually know what they want to do with their power. Like, if you don’t want people to buy sexbots, then why did you set up a society that uses cheap sexbots as a means of keeping the masses passive?

      This scenario only makes sense if the government is pushing the Nefarious Distractions on the populace with the one hand, and trying to crack down on the use of Nefarious Distractions with the other. If they both want the populace to ignore what they do and pay attention to what they do. Or in other words, if the government is basically at war with itself already. Yes, if such a government existed it would tear itself apart, but I’m not sure what chain of events could create such a situation in the first place. Maybe TPTB took on some new hires, and they didn’t get the memo about how the secret social control system is supposed to stay secret?

      (Side note: The government isn’t immune to culture, so I’d expect TPTB to have at least a few sexbot-lovers in their ranks. What’s the point of giving the masses all this awesome entertainment if you can’t enjoy it yourself?)

    • Tekhno says:

      @ThirteenthLetter

      So ultimately, we end up back with the boot stomping on a human face, although only until the human face gets pissed off enough and there’s a destructive revolution, hyper-advanced killbots exterminate the rioting but powerless humans, but still.

      Let’s be honest.

      • ThirteenthLetter says:

        Perhaps if we put the killbots in charge of running the state? Then we’d finally have rulers who don’t care what people think of them (as long as the killbots don’t have fully general AI, that is.)

    • AnonYEmous says:

      Non-brutal authoritarianism, whether M*ldbuggian or Huxleyan, is a fundamentally unstable state due to the human nature of the rulers. Plausible?

      basically the subject of an essay I wrote about 1984

      • sohois says:

        North Korea has essentially been a stable Orwellian society for more than 50 years now. They don’t seem to face any danger of a popular uprising, the main issue the country faces to survive are its disastrous economic policies. Would seem to contradict this suggestion of fundamental instability.

        • 1soru1 says:

          I don’t think NK counts as non-brutal.

          • sohois says:

            Completely skipped over the non-brutal part for some reason.

            You’re right, it doesn’t count in this case.

          • AnonYEmous says:

            Well, aside from Soru’s point – my point was that 1984 was built on the idea of a spartan upper class, as in they don’t debauch themselves or anything. In others words, they hold all the power and don’t do anything with it. How long can that last?

          • Nornagest says:

            How spartan were the upper classes in 1984? We don’t see a lot of them, and what we do see is better than what the middle-class protagonists were dealing with, though not particularly lavish.

          • LHN says:

            Pretty spartan. They get real coffee and real tea and “good food and good tobacco”, and have “two or three” servants (possibly POW slaves? one Winston sees “might have been… Chinese”) and guards. But they still live in a “huge block of flats” (albeit with “incredibly rapid lifts” by Winston’s standards), not mansions or even townhouses. Aside from the servants (and helicopters), there’s nothing they have that a middle class American doesn’t have more of, and this is the top two percent of Oceania’s population. The Soviet nomenklatura were AFAIK likewise much better off.

            According to Goldstein’s book (which, admittedly, is suspect since it’s itself state propaganda; Winston and Julia never get an actual outside view), “It is deliberate policy to keep even the favoured groups somewhere near the brink of hardship, because a general state of scarcity increases the importance of small privileges and thus magnifies the distinction between one group and another. By the standards of the early twentieth century, even a member of the Inner Party lives an austere laborious kind of life.”

          • AnonYEmous says:

            Oh right. I guess that was from his book and not necessarily reality.

            shrug

            but the other thing is that, even beyond amenities, their sheer totalitarian power is only used in ways that perpetuates it, and that’s all. That bothers me. I don’t think that would be real.

          • John Schilling says:

            But they still live in a “huge block of flats” (albeit with “incredibly rapid lifts” by Winston’s standards), not mansions or even townhouses.

            “Flats” like this, perhaps? Whether One Hyde Park qualifies as a “huge block” is debatable, but in British real-estate speak “flat” does not convey spartan utilitarianism; it’s the generic term for any living arrangement where you don’t own or rent the entire building.

          • LHN says:

            What Goldstein’s book says at least matches what Winston sees when visiting O’Brien, and the sorts of black market goods Julia’s able to snag. (E.g., chocolate that sounds good and that Winston compares lyrically to the usual Outer Party ration, but which isn’t obviously better than you or I could get at a drugstore; or real sugar rather than saccharin.)

            Granted, it turns out that all of Winston’s interactions were about getting the goods on him and arresting him. (Which… I know that a crazy-high fraction of, e.g., East Germans were employed by the Stasi. But still, tasking O’Brien and Carrington and whoever was responsible for monitoring the telescreen all to spend months playing cat and mouse with Winston and Julia seems like a pretty resource intensive process for rooting out crimethink.)

            That said, there’s never anything that really contradicts the Goldstein book’s assessment of society– except insofar as it implies that the system is permanently self-reinforcing. (The international war can never end, it explains, and the Party can’t be overthrown internally.) Since the appendix on the principles of Newspeak was written after the fall of Oceania[1], Goldstein is clearly missing something.

            Possibly the Inner Party losing that fervor necessary to focus on nothing but power for its own sake, and instead attempting to engage in “limited” openness and restructuring that would allow them to retain power while actually benefiting from their authority. Which eventually gets away from them.

            [1] In 1984 they were working on the tenth edition of the Newspeak dictionary, and it says the eleventh was the final one, while the system is always described in the past tense, and the final transition to Newspeak as something that “would have” happened by 2050.

    • Theres nothing plausible about MMs scenario. Sciencel fictional god emperors can achieve dominancr through mysterious powers, but real life authoritarians can only achieve it through cobstant vigilence and a cumbersome state apparatus, which is ulinately inefficient compared to a free society. …conpare west germany to east germany.

  4. bean says:

    Part 2 of Battleship Armor. Part 1 is here.
    Now to details of Iowa’s protective scheme, to put all of this into more concrete form. Iowa’s belt is 12.1” of US Class A armor on 30 lb STS, sloped at 19 degrees inboard (bottom-in, opposite of a tank). The slope means that it’s lighter than an equivalently protective belt of vertical armor, but it is mounted inside the hull, which means that the damaged waterline of the ship is somewhat narrower than the undamaged waterline. This is bad for damaged stability, and the next class (the never-built Montanas) would have returned to external armor. The lower strake of the Iowa’s armor (cross-section) tapers down to 1.625” and extends down to the bottom of the ship. It is intended to serve as protection against underwater shell hits, which were a major concern at the time. It was never tested in combat. It also forms the third bulkhead of the torpedo defense system (a subject which deserves a post in its own right). The transverse armored bulkheads that bound the forward and aft ends of the design are 11.3” to 8.5” on Iowa and New Jersey, and were thickened to 14.5” on the other ships. It was a known weak point in the design, which the Navy wanted to correct after the treaty lapsed at the start of WW2. However, the armor for the first two ships had already been ordered, and the delay to replace it would have been unacceptable.

    The deck protection is in three main layers. The main deck is 60 lb STS, and is called the bomb deck. The idea is to explode light-case HE bombs high in the ship, and start fuse action in AP bombs, hopefully exploding them before they get deep into the vitals. The main armor deck is 4.75” of Class B armor on 50 lb STS, intended to keep out shells and bombs. Directly below it is an mesh of empty compartments two or three feet deep, with a splinter deck of 25 lb STS forming their base to catch any fragments knocked off the main armor deck.

    The whole system was very similar to that of the South Dakota, with an immune zone of 20,400 to 26,700 yards against the 2700 lb shell from a 16”/45 gun and a zone of 17,600 to 31,200 yards against the same gun with a 2,240 lb shell.

    The barbettes are 17.3” to 11.6” above the armored deck (they’re Class A, and the thinner armor is used when the barbette structure is screened by other parts of the ship), as are the sides of the conning tower (Class B). The turret faceplates are 17.0” of Class B on a 2.7” backing. The conning tower and turrets all have 7.25” roof plates. The rest of the turret armor is Class A, with 12” back plates and 9.5” sides.

    But how did all of this come to pass? Originally, armor was very simple. Put thick iron plates on a ship’s side, and hope it’s enough to keep out enemy fire. But guns kept getting bigger, and it proved impossible to put enough armor on a ship’s entire side. So the next stage was to concentrate the guns and armor in a central battery, and leave the ends unarmored, except for a thin armored deck at or slightly below the waterline, and possibly a short waterline belt. The purpose of this was to preserve buoyancy in the ends of the ship and the short ranges of the time guaranteed that shells would not be falling steeply onto the decks. Turrets began to appear on warships, using an armor scheme similar to the central battery ships. The problem was that turrets were very heavy, and had to be mounted low in the ship. This lead to an alternative mounting for guns, the barbette. It was essentially an armored tube, which lead from a low-mounted armored deck up to the main deck. The gun was mounted within it, and retracted into the barbette for loading. It was then raised out and fired, with only the aimers coming above deck, where they were protected by armored hoods. However, the rise of the pre-dreadnought with quick-firing guns made this less tenable, and an armored hood was placed over the whole gun installation, which quickly (and confusingly) became known as a turret. Extensive light belts were installed against the QF guns, to protect the ship’s own QF guns and buoyancy.
    Some ships had 20” or thicker belts at this point, although rapid developments in armor metallurgy rendered this unnecessary. Early ironclads had wrought iron armor, as cast iron was too brittle to be usable in the thicknesses available aboard ships. In the 1870s, steel finally advanced to the point where it was a viable competitor to wrought iron, although it was still brittle. This lead to compound armor, a steel faceplate welded on to a cast-iron back. The British used it extensively, while the French (the world leaders in metallurgy) used pure steel. This lasted until the invention of nickel steel in 1889, which made the steel tough enough to be used on its own. This was soon improved by Harveyizing, which placed a thin (~1”) ‘cemented’ layer on the face of the plate by adding carbon and heat-treating. This layer would shatter shells, although the AP cap was soon developed to defeat it, essentially a sacrificial piece of metal which would break the face and be broken, allowing the undamaged shell to hit the back of the plate. Soon after, Krupp invented the face-hardening system described last time, which relied only on heat treatment. Most WWII face-hardened plates were cemented as well as being face-hardened (the Japanese were the main exception), although post-war tests concluded that the cementing did not actually make the plate any better against contemporary shells.
    All you ever wanted to know about armor metallurgy and penetration

    • cassander says:

      I’ve always had a problem with the apparent thickness of sloped armor. Alright, I get that sloping the armor increases apparent thickness. But it also makes the armored area narrower in the vertical. At 15 degrees, your 10′ wide belt only covers about 8′ vertically. If you want it cover the vertical area, you need to make your strake wider, which means more weight, which means angling hasn’t saved you anything. There are other reasons for angling, increasing the effective angle of impact makes shots more likely to glance off, but the apparent thickness argument has always struck me as dodgy accounting.

      • bean says:

        I’m with you on this one, but I think it’s the fault of people who are trying to explain things they don’t understand, because of the fact that everything to do with armor is horribly complicated. On a tank, you’d get reduced deck armor, and thinner plates are easier to manufacture. But that’s definitely not the whole story. The slope gives you extra benefits beyond the added apparent thickness. I can’t say exactly what that is. The link at the bottom of the post is the best source I know of more details on this. Beware, though. It’s not an easy read. Even the author admits that. (I got to meet him once. He was a bit amused to find someone very excited to meet him.)

        • cassander says:

          Good, I’m glad I’m not the only one that finds this problematic. Everyone, it seems, uses the apparent thickness argument, and it’s always struck me as cheating. When you think you see something that no one else does, there are only two possibilities, and the answer usually isn’t “I’m a genius and everyone else is stupid.”

          • bean says:

            I looked into this more, and was able to provide some mathematical backup. Iowa’s belt was is about 3% more effective than a matched-weight vertical plate at 0 deg angle of fall. At 18 degrees (towards the edge of the IZ), it’s 16% more effective, although some of that is because it’s apparently thicker, and when that’s factored out, it seems the result is still 3%.

            Tank plates don’t get apparently thicker with increased fall angle (the opposite happens, in fact) but a plate at 45 degrees is 19% more effective than a matched-weight vertical one at 0 degrees angle of fall.

            (All percentages are in terms of ballistic limit velocity, calculated using a dimensionless version of the US Navy empirical formula.)

          • cassander says:

            right, but you aren’t making the belt apparently thicker, you’re making it actually thicker.

            Imagine a belt 1′ thick and 10 feet tall. It has a cross sectional area of 10 feet. then you angle it at 45 degrees. It still has a cross sectional area of 10 feet, but now it only covers 5 feet in the vertical, so the horizontal thickness (as in, thickness if you’re parallel to the ground) is 2 feet. But you haven’t done any trickery, you have taken a fixed amount of metal and squashed into half the half the vertical distance, so of course you get twice the horizontal thickness. The effect is identical to having a vertical belt half as tall and twice as thick. If you imagine each vertical foot as a separate strake of armor, one stacked on the other, you could just slide the plates on top of one another and change nothing.

            Now, I realize that angles matter, and belt that increases the angle of impact is better than than one that isn’t, but that’s a different argument than the apparent thickness argument that I often see put. which argues that horizontal thickness of the armor is apparently increase. No, it’s actually increased. You’re flat out using more armor. per square inch of armored surface.

          • bean says:

            @cassander

            right, but you aren’t making the belt apparently thicker, you’re making it actually thicker.

            No, I’m not. I specifically increased the thickness of the vertical belts in my comparison to compensate for this, and found that the slope still provided a benefit, sometimes a substantial one. I agree that the conventional reason given for sloped armor can be dismissed with ‘that just makes the armor actually thicker to the shell, and you could smash the sloped plate into a vertical plate with the same benefit’. I was able to show that slopes provide benefits independent of that effect.
            Specifically, the 12.1″ plate at 19 deg was up against a 12.797″ vertical plate, and a 1″ plate at 45 deg went up against a vertical plate of 1.4142″. At battle ranges, I got better than 15% improvement for the sloped plate in both cases.

          • Nornagest says:

            The Wikipedia page on sloped armor list three benefits for it: apparent thickness, efficiency in enclosing a volume, and deflection effects. Any benefit here must be coming from deflection, since apparent thickness doesn’t save us any weight (it might save us money, though) and since sloped armor seems to be less efficient in enclosing battleships’ protected spaces.

            Deflection effects seem to be insignificant if we’re talking about hypervelocity projectiles like explosively formed penetrators, but the shells from a battleship’s main guns would be going much slower than that.

          • bean says:

            @Nornagest
            Correct on all counts. I specifically was measuring deflection effects, holding weight constant. I should point out that battleship-style armor does have an effect I didn’t anticipate of showing the shell more armor at long range than an equal-weight belt, which probably does as much or more for the immune zone than the strict deflection effects do.

            Hypervelocity impacts are a very different kettle of fish from the sort we’re talking about here. I wouldn’t rule them out entirely, though. The slope might make the projectile hit off-center, which absolutely kills penetration in hypervelocity long rods.

            Edit:
            The real problem with steeply-sloped armor isn’t with internal arrangement per se. That’s a pretty typical problem of naval architecture. The problem is that the internal belt narrows the waterline, which reduces stability in a damaged condition, which is what the naval architect really cares about. That’s why they went back to external belts for the Montana, although I think there was still a slope, like in the North Carolina.

          • cassander says:

            @bean

            Ah, I didn’t realize that’s what you meant by matched weight, re-reading it now, though, I probably should have.

  5. Teucer says:

    Hello SSC friends,

    I’m an undergraduate majoring in economics at Midranking Public University. I’m doing reasonably well so far with the 300 level econ material, but I’m not totally sure that I have the mathematical skills to comfortably pursue a course of study at the graduate level, which is what I was vaguely intending to do originally. Though I hear negative things about the legal labor market, I’m considering applying to law school and attempting to enter the legal profession; if anyone has thoughts/advice/useful information of any kind to share about this I’d very much appreciate it. (Also, anything regarding economics as a field, white collar professions, the rat race, etc. is fine.) Thank you all kindly in advance.

    (Incidentally, I’ve heard that the LSAT is very much like a verbal IQ test, so, if it makes any difference to the advice one would give, I scored at the 99th percentile on my SAT verbal, without any studying beforehand. Also, while I don’t know of a comparable way to quantify this, if this is relevant I would say that my social/networking skills are at least fairly weak, in that I have few friends and generally dislike making what I feel to be inane conversation.)

    (And, lest anyone by some chance be worried about this, I’m not going to make any major life decisions solely on the basis of SSC comments; I’ll of course be doing much more research in other avenues before making a decision.)

    • hoghoghoghoghog says:

      Graduate school is very safe. The first 1.5 years you will learn things that are applicable to many other jobs (this is true even if you are studying something stereotypically useless, so long as you are doing it in a good program), so if you need to bail you have not wasted your time. Also they pay you instead of you paying them – of course you have to teach or do brute work in return, but teaching is fun and brute work is just life.

    • Machina ex Deus says:

      Math skill acts like a muscle: you can improve it by using it. Do a bunch of similar mathematical econ exercises in a row; sometimes mathematical principles or insights present themselves as shortcuts, which your mind won’t bother coming up with if it knows there’s just one more of these find-the-area-that-corresponds-to-consumer-surplus problems.* And just like you have to challenge yourself in weightlifting, you have to challenge yourself with math; I didn’t make anywhere near as much progress doing math one day a week for seven weeks as I did doing five days straight of it.

      Since you’re doing OK in 300-level courses, I assume you have enough natural mathematical talent to already understand calculus and basic statistics. If working at it more intensely doesn’t help you advance in whatever mathematics you’re worried about, then maybe you can start thinking about bailing on econ with a bachelor’s.

      Also, talk to your professors a lot. See if they have anything interesting for an undergrad to do (e.g. crunching numbers in R for some paper they’re writing). (This isn’t specific to you: almost no undergraduates take advantage of out-of-class interaction with their professors.)

      Don’t go to law school without first talking to people who’ve gone to law school. You can probably find some of them waiting tables at restaurants near your university.

      (* This would be a good place to point out that I don’t actually understand 300-level economics. Or probably 200-level economics. I blame Dick Wolf for not producing a show about economists.)

    • Two points:

      1. Economics at the moment is much more mathematical than it needs to be. Part of the reason is that most people are working the intensive margin, trying to say something publishable about a problem that smart people have been looking at for a century or so. One way of doing it is to use some new mathematical technique, whether or not it tells you anything useful.

      There is a real opportunity on the extensive margin, using economics to do something new. The old example would be Gary Becker. A current example would be Peter Leeson.

      Of course, you still have the problem of getting through an overly mathematical grad school before you get to do the less mathematical work.

      At only a slight tangent, I like to claim that the one benefit I get from having a doctorate in theoretical physics is that I can do non-mathematical economics without having people think I am afraid of math.

      2. Law schools have been facing a problem of low application rates for several years. On the one hand, that is a result of a lot of people concluding that it wasn’t worth going to law school, and they may well be right. On the other hand, it means that if you have a high LSAT score you should be able to get a substantial subsidy from a school that is trying to feel its roster without lowering its averages too much.

      • Tibor says:

        I agree with the general statement, that using fancy mathematics just to use fancy mathematics is not a good idea (it is actually not even very mathematical 🙂 ). But when I read your Hidden Order, I sometimes wished you used a little more mathematics, basic calculus would more than sufficient. I understand you wanted to write a book anyone could understand, but it would be more concise and hence clearer to me if you used calculus. One could even write mathematics itself entirely in words, the ancient Greeks did it like that, but it makes the arguments much less tractable.

    • ThaadCastle says:

      Lawyer here. The legal labor market is very cyclical and somewhat unstable. It only really makes sense to go to law school (if you are going to have to have loans to cover close to full tuition $150K + 3 years of expenses) if you are going to be able to go into ‘Big Law’ (the firms that pay $150-180K, depending on city, coming out of school).

      These firms really, really care about the rank of the law school and your rank within the law school. For example, probably the top 50% or so of students at schools ranked (by US News) 7 to 10 (Berkeley, Penn, Michigan, UVA) or so got these kind of jobs out of law school. Close to 100% of kids at Yale, Harvard, or Stanford got these jobs (or got jobs of equivalent competitiveness). While it is much much more difficult to get these jobs if you are at a law school ranked event 14 to 20 (UCLA, Vanderbilt, etc.)…you would likely need to be in the top 10-20% of students to get these jobs, and you might be shut out of the very ‘best’ firms.

      The lower ranked the school, the more difficult it gets and the less national your options are, often lower ranked schools are only able to place the top 4 or 5 Students at big law firms, and that is usually restricted to firms in the nearest city…i.e. if you are studying at University of South Carolina, you would have a really hard time getting a job in Texas…no matter your class rank.

      The LSAT is a very different test than the SAT Verbal, but the scores probably do relate closely…getting into ‘top’ law schools is almost completely a numbers game…have a 4.0 (no matter the undergrad school/major) and a 175 on the LSAT and you are almost guaranteed to get into Harvard…the lower the LSAT/the lower the GPA the less likely you are to get into a ‘top school’ (although, within reason, there are lots of things you can do to make yourself more competitive).

      Also the LSAT is at least partially teachable, it is 100% worth taking an LSAT prep course the $1200 the courses cost is a drop in the bucket compared to the future earning potential/potential scholarships you can get if you can increase your score but even 5 to 10 points.

      TLDR: Law school makes sense if you can get into a top 10+ program, the higher ranked the law school the more financial sense it makes…if you can get into Harvard/Yale/Stanford you set in terms of a job. Getting into these schools is based on your LSAT/GPA, if you take the LSAT cold and get in the 160s and have a GPA above 3.8 you can probably get your LSAT high enough to be competitive for the schools that make financial sense.

    • Corey says:

      Second-hand data point: Economists Miles Kimball and Noah Smith are pretty bullish on econ PhDs, and have argued that anyone can be a “math person”, so the level of mathiness you need to pursue econ grad school may be less than you think. (In general I’m a big proponent of outsourcing math grunt work to computers, I loved “Calculus & Mathematica” as an undergrad calc option, for example).

    • James Miller says:

      I have a J.D. and a Phd (econ). Yes, graduate level econ is very mathematical and, unfortunately, you shouldn’t do it unless you have very strong mathematical skills. I did very well on the LSAT, but also much better on the math than verbal sections of the SAT and GRE. I don’t think that the LSAT is close to the verbal sections on the SAT or GRE. Social skills are very important to being a lawyer, although they don’t matter much to doing well in law school. The logical reasoning sections of the LSAT are probably closer to math than verbal IQ tests. Studying for the LSAT probably greatly improved my performance. You might want a strategy of only going to law school if you get into a very good one, and then long-term you could consider trying to become a law professor who uses economics in research. You should look into getting an advanced degree in public policy if you like law and economics and would rather write papers than win clients over lunch.

      • and then long-term you could consider trying to become a law professor who uses economics in research.

        At the moment, law schools are not hiring professors, due to the fact that a reduction in the applicant pool means they are trying to cut costs. That might not be true for the top law schools–I’m in, and observing, a pretty good law school but nowhere near the top level.

        Whether that situation will change and when is unknown.

        My suggestion for someone who wants to follow your suggestion is that he should try to write journal articles as a law student or econ PhD student. That will give him some idea of whether he enjoys doing it and is good at it, and if he succeeds it’s a pretty strong qualification.

        Incidentally, law and econ has to some degree followed econ in becoming mathematical, but the math is generally not as advanced. Some of the most influential work–Coase would be an extreme example–uses no math that a bright college freshman couldn’t follow.

  6. Nancy Lebovitz says:

    Thinking about the NK missile launch and various displays of military capacity nearby from other countries– does that sort of thing ever turn into a war? My impression is that they don’t, but I may have missed something.

    • The Red Foliot says:

      South Korea could easily obliterate North Korea but that would require a commitment of lives and resources. The SK government asks itself, ‘to what end?’ North Korea is poor and in this capitalist age, no one wants it. It makes a nuisance of itself partly to provide internal propaganda for its own citizens and state officials, partly out of caprice, and partly to extort food aid out of their own worst enemies: SK, the UN and USA. An amusing fact is that the food aid comes packaged with propaganda materials denouncing the NKean state, which remain intact even as it gets distributed to the NKean populace, apparently because the NKean state doesn’t bother to remove it. Probably the Great Leaders themselves would have to intervene to get anyone to take the initiative to remove the propaganda.

      • Aapje says:

        An amusing fact is that the food aid comes packaged with propaganda materials denouncing the NKean state, which remain intact even as it gets distributed to the NKean populace

        I have great trouble believing this, given the extent to which N-Korea goes to ensure that visitors do not talk freely with N-Koreans and how upset they got over various propaganda efforts. Do you have evidence?

        • The Red Foliot says:

          Yes, my evidence is that I have viewed some of those “undercover in NK photos” and seen large bags of food emblazoned with US propaganda slogans sitting in plain view in NKean marketplaces which themselves are a mockery of the ‘communist’ state.

          • Aapje says:

            Are the slogans in English? They may figure that a negligible percentage of the commoners can read English.

      • CatCube says:

        I understood that they may have removed written propaganda, but they don’t bother to cover the bags that say “gift of the people of the U.S.” or whatever; the regime just tells the people that they got it from the American Bastards as tribute.

      • cassander says:

        The obliteration isn’t even the problem, it’s dealing with the aftermath. Germany is still suffering the effects and paying the costs of re-unification decades afterwards. East germany had 1/4 the population of the west, while south korea only has twice the population of the north, and hte gulf between east and west germany was far smaller than north vs. south korea. Re-unification will cost trillions of dollars, and the Koreans know that they’ll get stuck with the bill. They are, understandably, not eager to pay it.

        • The Red Foliot says:

          That is true but lamentable. ‘The price of glee in China’ should indicate, most especially to the East Asians, given their proximity to China, that there is a paucity of imagination inherent to financial thinkery. Financial thinkery is where politicians and the masses think only in monetary terms. An example is when it is said that mass immigration is ‘necessary’ because without it there would be less money. No one is capable of wondering, in this age of supreme wealth, why that matters.

          National projects like reunification might be just as pointless, ultimately, as acquiring wealth, but they at least have an aura of grandeur about them which can really electrify one’s brain by their prospect. No such aura exists for the dull accounting of wealth. There are two choices, both probably useless, but one is nevertheless exciting–yet, people choose the dull one out of… dullness. This is why SK fails. And will always fail. Probably Kim Jong Un will choke to death on a hamburger and his successor, as well as the one after, will likewise indulge disgustingly in synthetic pleasures. But the country WILL eventually be reunified, if only after somebody chokes to death without first siring an heir. The reunification will take place; only, it will have no verve.

          • cassander says:

            The re-unification will have to take place….someday. But if I do it today, I might get thrown out of office by angry voters once they get the bill. Best play it safe and let my successor handle it…..

          • An example is when it is said that mass immigration is ‘necessary’ because without it there would be less money

            Who are you imagining saying that? Money can be printed. Do you mean “without it, the present inhabitants of the country will be worse off”?

    • hoghoghoghoghog says:

      The invasion of Iraq was at least partly about Iraq developing new weapon systems, but that’s the only example I can pull off the top of my head. I’m kind of surprised it isn’t more common.

      • Nancy Lebovitz says:

        I’m not talking about invasions that happen as a result of someone being concerned about weapons development. I’m talking about “I wave my aircraft carriers in your general direction” blowing up into something serious and/or being used as a chance to jumpstart an invasion.

        • Civilis says:

          I’d say the Spanish-American War could possibly count as an example of a show-of-force inadvertently escalating into a real war.

          1. Tensions exist between the US and Spain.
          2. The US sends the USS Maine to Spanish Havana to show the flag.
          3. The USS Maine blows up.
          4. Interest groups that want war interpret this as a Spanish attack, and pressure the US into declaring war.

        • 1soru1 says:

          Arguably the current situation on the Korean peninsular is historically unusual enough that a lack of precedent doesn’t say as much as you might think.

          One side has a rational expectation that they could easily achieve total victory, and an equal level of confidence that doing so would be worse than the status quo. The other side knows this, so feels free to provoke them arbitrarily.

          I can’t imagine how that situation could exist before the mid-20C. In the 19C or earlier, either the smaller power would have done everything possible to keep in the good books of the side with military supremacy, or they would have been invaded decades ago.

          • The Nybbler says:

            Does “total victory” include “A smoking glow-in-the-dark hole where the capital city of 10M people used to be?” Even without nukes, as I understand it Seoul is within range of a lot of already-emplaced conventional artillery which couldn’t be eliminated in a single initial strike.

            WWI is a famous case where mobilization led to open hostilities. I don’t know of a case where simple saber-rattling did… but would it be called that in retrospect if it did lead to open hostilities?

          • Cypren says:

            Popular Mechanics published an article a while ago about the somewhat exaggerated concerns that North Korea can “flatten” Seoul and that they’re essentially holding the city hostage.

            The TL;DR version: artillery really isn’t as devastating as popularly believed, and even thousands of batteries take a long time to do more than simple terrorism and harassment of a built-up area. Leveling a city down to rubble requires either a sustained bombing campaign (e.g. Dresden) or nuclear weapons.

          • James Miller says:

            This might explain why North Korea recently demonstrated that it had VX.

          • John Schilling says:

            We’ve always known that North Korea had Sarin and Mustard at least, and we still don’t know whether they can mass-produce VX. Not that this makes much difference.

            Also worth noting that Seoul appears to have subcontracted much of its construction work to the Dwarves, with interconnected subterranean subways, walkways, and malls that reach farther than I could explore and have civil-defense supplies stockpiled at regular intervals.

            Finally, the North Korean government is run by Korean nationalists and Seoul is populated by Koreans. They’ve got more important people to kill than the civilian population of Seoul.

          • James Miller says:

            I get the feeling that if the leader of North Korea was told that he would soon die in an American attack his last wish would be to kill as many Americans, S. Koreans, and Japanese as possible in as painful a way as possible. Of course, this might be just false deterrence signaling.

          • John Schilling says:

            I get the feeling that if the leader of North Korea was told that he would soon die in an American attack his last wish would be to kill as many Americans, S. Koreans, and Japanese as possible in as painful a way as possible

            That’s like saying that if Donald Trump were told he would soon die in a Russian attack, his last wish would be to kill as many Russians, Canadians, and Chinese as possible in as painful a way as possible.

        • Aapje says:

          Russia started mobilizing in 1914. As the Germans thought that war was coming and they thought that their only chance at victory was a quick victory over France before Russia was fully mobilized, this resulted in the Germans attacking France.

    • Salem says:

      One example comes to mind: in 1958, the Iraqi Army was ordered on field maneuvers and military exercises intended to demonstrate the strength of the country’s military capacity. Brigadier Qasim used this as the opportunity to stage a coup.

  7. Montfort says:

    In the SSC tradition, I am seeking some book recommendations:
    1) Books dealing with the history of sub-saharan africa (or some region of it) some time between WWI and today, inclusive.
    2) Books on ancient/classical history that I can pick up and read with only the knowledge of that period gleaned from public school and a few wikipedia articles. Specifically I want to minimize the number of times I read a reference to some event/person/organization that is important but unexplained in the text.

    Bonus points for well-sourced books with footnotes.

    Thanks in advance.

    • hoghoghoghoghog says:

      Not a historian, but I like Harper’s From Shame to Sin. It’s really great if you happen to be interested in the history of ethics. It totally changed my opinion on the advent of Christianity from “essentially random meme takes over a society, like memes do” to “Roman worldview was pretty much broken on its own terms and needed to be refactored.” It is not primarily about sexual ethics, though maybe the author believes that it is.

    • On classical history, I am a fan of Alexander the Great and the Logistics of the Macedonian Army. It’s a history of the campaigns with all the battles left out.

      There is a book called something like “small wars since WWII” that has some stuff on sub-Saharan conflicts, but it isn’t limited to that and I don’t remember the exact title.

    • Fossegrimen says:

      I like “The Faith Of Africa”, lotsa pages, heavy on fact, light on analysis.

    • Anon. says:

      I highly recommend the “Landmark” editions of Herodotus and Thucydides. Extensive notes, maps, and appendices that explain everything.

    • cassander says:

      African Kaiser is a fun tale of the most successful guerrilla campaign in history. Not an amazing book, but an interesting bit of history. Another account of it might be a better choice, I’ve heard good things about, but not read, the great war in africa.

      For classical history, Lords of the Sea is quite good. Donald Kagan’s Peloponnesian War is the standard count and exhaustively foot noted.

    • Montfort says:

      Thank you all for your kind recommendations, I will investigate further and probably order more than I can actually read.

  8. knownastron says:

    Hello fellow SSCers,

    This upcoming year I’m going to be coaching a business case team for my alma mater. I’ll be picking teams this week. In total about 40 students will be chosen, so there’s a lot of candidates to go through.

    Each candidate will do a 5 minute presentation (after working on a case for an hour) and then there will be about a 10 minute interview. I’m hoping to get the most out of the interview by asking good revealing questions. I’m looking for something similar to the famous Sergey Brin question: “I want you to explain to me something complicated that I don’t already know.”

    Has anyone had success or know of good questions to determine someone’s intelligence? What signs or clues should I be looking for? I will also have transcripts, how much should I base on their grades? Is there any red herrings I should be careful of?

    Any advice is appreciated!

    • nimim.k.m. says:

      I will also have transcripts, how much should I base on their grades?

      This sounds like something that would be subject-specific. While attaining a consistent row of good grades usually requires a great deal of work and some amount mental acumen in any respectable school, in theoretical physics the demands are somewhat different than in say, history.

      I’d have a personal preference of using the grades only to look for stuff the applicant has done (and maybe use that information to come up with questions for the interview) and maybe filter out the ones with lots of bad grades and maybe then on a later stage use it for resolving ties. However, that’s just an uninformed opinion, nothing based on experience because I don’t have any experience in anything resembling this.

    • Simply ask them the question “Why are Manhole Covers round”

      Did anyone *ever* believe questions like that added value? Or was/is it all a game of hiring pre-determined candidates and those mandated by politics?

      • Cypren says:

        The theory was, at one time, that asking off-beat questions forced the subject to actually demonstrate analytical process in real-time instead of repeating rehearsed answers. The global higher education system is, in general, very good at turning out reams of candidates who can memorize and regurgitate arbitrary information; it’s exceptionally poor at indicating those who are capable of creating it.

        Of course, this test lost any and all value as soon as the questions became reasonably common knowledge and people simply started studying to the new test and regurgitating canned answers to the questions that were supposed to catch them off-guard. But as with most things, inertia kept it alive well beyond its usefulness.

        While these types of questions originated in tech companies, they’re fairly rare in the hiring process now; most have gotten the memo that the questions are widely disseminated and studied by prospective candidates at this point.

        • houseboatonstyxb says:

          For the applicant, reading such lists could help weed out the companies that can’t make up their own questions.

        • knownastron says:

          I think you’re right that “at one time” these questions were useful when hiring at tech firms but no longer.

          My thinking is for case team selections at a mid-tier Canadian university that it is still that “one time.” This would be the first time we would be implementing something like this. So no candidates would know to prepare for something like this.

        • Why would anybody with a knowledge of psychometric tests and its intersection with so called “crystalized” knowledge resort to such questions? You’re simply decreasing your predictive abilities.

      • Deiseach says:

        Did anyone *ever* believe questions like that added value?

        Fads. Management guru writes book/has hugely successful seminar/generates tons of buzz, every Tom, Dick and Harry wants to implement it in their own company, it gets wildly popular for a couple of years until the next fad, rinse and repeat.

    • WashedOut says:

      I work for an engineering firm. When I was starting out I was asked to explain how I would approach the problem of estimating the number of trees in our state.

      I like this problem (and this question), and would recommend it at least as a talking point to see how well someone can articulate their problem-solving ability.

      • rlms says:

        These kinds of problems are called Fermi estimation problems.

      • knownastron says:

        That’s a great question. A friend got asked “how many pizza shops are there in New York?” in an investment banking interview. I’ll implement something similar in my selection process.

      • shakeddown says:

        “How many cargo ships are there in the world” is also a good one.

      • Is a test of on the spot articulation the best way to find an engineer who has 2 months to think deeply on a topic and write in a structured manner?

        I really think that firms ask questions like this in order to hire pre-favored candidates, or those that they must hire due to certain reasons.

        That is if these questions are asked in a rational duplicitous manner, and people don’t believe their own stuff.

      • skef says:

        I’d be interested to hear what people think what this variety of question tests for.

        It seems to me that during their heyday, the main thing they screened for was a lack of a certain kind of reticence. There are ways of getting more accurate answers to these questions from people with more specific expertise. Some interviewees will think “why are they asking me to bullshit about this?” Meanwhile, the interviewer is just judging conceptual grasp on the problem, and willingness to communicate openly.

        What makes the questions suspiciously cultural-seeming is that you could set them up in a more neutral way. You could, for example, describe a scenario in which the person’s team had to solve the problem in isolation, with certain materials. But this is not generally how the questions were introduced. So a lot of the “test” is of the interviewee’s recognition that question A should be answered in light of question B.

        • Aapje says:

          I think that it tests for pragmatic problem solvers. Engineering firms don’t want people who are too scrupulous, but they want people who get the maximum out of the information that they have.

          • skef says:

            Really? If I needed to estimate the number of trees in my state I would start with a bit of googling, not conceptual analysis from what little I happen to know at the moment. Is the latter pragmatic?

            (Come to think of it, the state of the art probably calls for some test of “googling stack-exchange”, as The Onion put it. I suppose that’s a premise behind take-home limited time programming tasks.)

          • Aapje says:

            The question isn’t intended to mimic a realistic scenario. It seeks to put you in a situation where there is an obvious way to improve the process (googling) that everyone should realize exists, but that you don’t get access to. At that point you can keep complaining about not having access to Google (fail: lack of pragmatism), you can flounder (showing that you have limited problem solving ability) or you can come up with a reasonable solution given the information you have (success: hire).

    • Brad says:

      Remember Cowen’s Second Law: There is a literature on everything.

      This isn’t quite the hiring problem, but it is probably close enough. As something of an aside, whenever I read discussions about hiring and interviewing in my industry (tech) I have the strong sense that we are in the pre-moneyball era. Everyone has strong opinions seemingly based on very little.

      • Machina ex Deus says:

        Contract-to-hire (assuming some people actually get hired) is the most accurate way to evaluate tech candidates. Beyond simply seeing whether they know what they say they know, you can see if what they know is actually useful to your organization, and how fast they can pick up things they don’t know. Also, frankly, to see how well they work with the actual people in your organization: it can be much better to have a team of pretty-good people who work together well as a team than to have a group of excellent people who nevertheless don’t jell.

        (Yes, this is a bit reminiscent of my favorite way of generating accurate estimates: do the work, then tell your boss it will take about that long to do, and by the way, does he have something else he’d like estimated?)

        • Brad says:

          The problem with contract to hire is that the fraction of the potential labor pool willing to consider such an arrangement is small and adversely selected.

          • Machina ex Deus says:

            Are you sure about the adverse selection? The few times I’ve done it, the couple-month contract period had higher take-home pay than the later full-time salary.

            Conversely, if I’d wanted to bail on the company, I could have explained to future interviewers that it was just a contract, so I wouldn’t look like a job-hopper.

          • Brad says:

            You probably aren’t going to get very many people that are currently employed to make a jump for a contract offer. At least in my experience.

  9. Nancy Lebovitz says:

    I’ve been seeing the claim that 1 in 12 trans people get murdered, and I just accepted it without thinking about it. I obviously need a check all numbers reflex, especially if the numbers are shocking.

    I’ve come across some debunking, and this looks like pretty good one, but I’d like to have some more minds on the question.

    • Deiseach says:

      Do they factor in the high rate of trans women involved in sex work? That would seem to me to raise the potential for violence, including murder. As for the high number of POC trans people murdered, this seems to link in as well:

      Fifty percent of black, 34% of Latin@, and 16% of Asian trans people have made a living in underground economies, including sex work, compared to 11% of white trans people, according to Injustice at Every Turn: A Report of the National Transgender Discrimination Survey.

      Reading the linked article, the “1 in 12” figure seems to come originally from one particular person (described as an “expert”). We would really need to see the source data for the “1 in 12” figure and what population it was looking at and what it considered (total of all assaults? included suicide? extrapolated an estimation?) to find out how it was derived.

      And do you classify “transgender person murdered” as part of that figure if it’s “murdered for other reasons e.g. robbery victim”?

      This gives a figure of 21 deaths by violence for transgender people in 2015. A figure of 1.4 million adults identifying as transgender in 2016 comes from here.

      This gives us very broad-brush figures of 21/1.4 million violent deaths. Using this formula:

      A crime rate is calculated by dividing the number of reported crimes by the total population; the result is multiplied by 100,000.

      That gives us 21/1,400,000 * 100,000 = 1.5 per 100,000 trans population.

      Compared to the general population of the USA:

      The US homicide rate in 2014, the most recent year available, was 4.5 per 100,000.

      If trans people were murdered at the same rate as the general population:

      4.5 per 100,000
      1.4 million/100,000 = 14 *4.5 = 63

      We would expect 63 murders.

      I’m open to correction both of the shaky maths and of the assumptions here; if anyone can work out that if 1/250th of the population has a murder rate 1/3rd of the total murder rate, what does that come to if we made them equivalent (i.e. if the trans population were 100% of the general population) and if that comes to “in order for the same murder rate in the general population, 1 in 12 people would need to be murdered”?

      • rlms says:

        There is no reasonable way of getting the 1/12 figure by exaggerating, as that would imply a murder rate of 8333 per 100,000. The most likely origins I can think of are either that it was completely made up, or that it is a vaguely related figure (the proportion of trans people who report being assaulted, or having attempted suicide seem like plausible to me).

        But disregarding that figure, and assuming trans people are a random sample of the population with regards to chance for random murder, combining the 1.5/100,000 and 4.5/100,000 figures gives a figure of trans people being 33% more likely to be murdered than the population at large, which is worrying enough in my opinion.

        • Protagoras says:

          Vaguely recalling some alarming but surely equally incorrect statistics about prostitutes, and the plausible reasons for them, I shall shamelessly speculate. One potential original source (assuming it isn’t just made up, which is not always a safe assumption, of course), is that there was some group of trans people under investigation in some study or other, and of those in that group who died during the study period, 1 in 12 died by murder. This could certainly happen if the group had the right characteristics; if, for example, nobody in the group was particularly old, so they were at much lower than normal risk for the most common causes of death, and perhaps if the group were also in some way biased towards those more likely to be murdered (lower SES, maybe sex workers, or whatever). And then somebody confusedly (or deliberately misleadingly, but while liars are of course common, so are people shockingly incompetent with statistics) reported the death rate of participants in that study group as if it were the death rate for trans people generally.

        • Deiseach says:

          It definitely is a worrying figure, given the tiny percentage of the population who are out as trans. If we take it that a disproportionate number of trans people engage in sex work, that does up the chances of violence and may help account for part of the increased murder risk.

          Now somebody needs to work out the risk for sex workers of being murdered vis-à-vis the general population, and I am not volunteering myself okay maybe I am.

          Right, I can only easily find a 2004 study concentrating on sex workers in Colorado:

          In comparison with the general population, the standardized mortality ratio (SMR), adjusted for age and race, …for death by homicide among active prostitutes was 17.7 (95% CI: 6.2, 29.3).

          I will admit this is a lousy sample to use, but since it’s all I’ve got – so a homicide of 17.7 per 100,000 for sex workers as against 2004 homicide rate of 5.5 per 100,000 in the general population. That’s something like three times the risk of murder for sex workers, which – if we’re taking it that a disproportionate number of trans people end up in sex work – would certainly bump up their risks of being murdered.

          And to further add fuel to the fire, POC have a higher risk of death by homicide, so add in all the POC trans people doing sex work and you’re getting a high-risk population for death by violence:

          Black people have consistently accounted for close to half the country’s homicide victims, making up more than 50 percent of the broader pool of those killed overall every year since 2010. The number of black victims increased 15 percent in 2015 over 2014.

          Of the 13,455 cases from last year in which the FBI listed a victim’s racial information, 7,039 victims – or 52.3 percent – were black. That compares with 5,854 cases – or 43.5 percent – in which the victim was white, an increase of about 8 percent from last year.

          It’s a disparity that becomes more pronounced in the context of population, as 2015 Census estimates suggest that whites account for 77.1 percent of the overall U.S. population of roughly 321 million, while blacks comprise 13.3 percent.

          • John Schilling says:

            I will admit this is a lousy sample to use, but since it’s all I’ve got – so a homicide of 17.7 per 100,000 for sex workers as against 2004 homicide rate of 5.5 per 100,000 in the general population. That’s something like three times the risk of murder for sex workers,

            But that’s going to be dominated (I think, time for more math…) by cisgendered sex workers. MTF transgender sex workers face all the risks of their cis colleagues, plus the big additional risk that comes from saying(*), “You totally have Teh Gay, yes you do, don’t you think I’m sexy, yes you think I’m sexy!” to a cis-hetero male whose mindset and hormone balance are as far on the No I Am Not Gay Dammit side as they are ever going to be.

            As a way to court brutal murder, that’s right up there with e.g. not paying your drug dealer because drugs are illegal and the contract is thus invalid. How much this increases the death rate over that of cis-female sex workers I do not know, but when I see lists of all the transgender murders in year X or city Y an awful lot of the anecdotes fit the pattern of an MTF sex worker whose client suddenly got very angry for reasons never explained in the story.

            * If not literally saying it aloud, implying it at the equivalent of 100+ dB.

      • Deiseach says:

        Okay, I did some more dodgy maths:

        If 0.004% of the population has 0.33% the murder rate, this works out to a rate 82.5 times that of the general population, which gives us 371.25/100,000 as the murder rate.

        This still does not get me “1 in 12”, though; it gets me “1 in 270”.

        The only way I can imagine it is that the original figure was calculated back when the US had a higher murder rate (1980 seems to be the record with a murder rate of just slightly over 10) or something, but I have no idea where the original figures were obtained or what they were. That would give me “1 in 120” on an “82.5 times the general population murder rate” and misplace a decimal point, there’s your “1 in 12”.

      • Eltargrim says:

        And do you classify “transgender person murdered” as part of that figure if it’s “murdered for other reasons e.g. robbery victim”?

        I’d say that including all trans victims of homicide is fair, even if their transness wasn’t explicitly the motivation. Tracking the latter is still valuable as a metric for explicit hate crimes, but if the trans murder rate is higher than the base rate it’s troubling regardless of the reason.

        A comment on this blog post provides a plausible explanation of the numbers behind 1 in 12. I’ll quote the important parts below:

        1. The important thing to keep in mind is that in 1999 (and arguably until maybe 5 years ago) the most commonly regurgitated estimate of transgender incidence was 1 in 15,000. [Reported as 1/10,000 mtf, 1/30,000 ftm–15,000 being the combined total].
        2. Dividing total late 90s US population by 15,000 produces an estimated US trans population of about 18,000 individuals.
        3. Next she must have come up with an annual number of reported murders approx. 19. [This figure is not unreasonable. Transgender Day of Remembrance data show 14 in 1999. Perhaps Brown had a better source, or maybe multiplied some prior year total by 150%–on the assumption of underreporting].
        3. Multiplying the one year figure by average life expectancy. Something like 19 x 77=1463… which is used as the lifetime homicide mortality.
        4. 1463 x 12 = 17,556. VOILA! That’s close to the total population of 18,000. QED–1 in 12 lifetime trans homicide mortality.

        Note that the major error in this calculation would be the rate of transgenderism in the population. 1 in 15 000 is about 7 in 100 000. That’s close to the rate of medical intervention sought for transition (according to Wikipedia, 9 in 100 000), but substantially below current rates of self-identification as being trans (again from Wiki, ~355 in 100 000).

        I think Hanlon’s razor can safely be applied here. There was a reasonable estimate made for the numerator, a reasonable (albeit wrong) estimate for the denominator, and while I’m reasonably sure that the above isn’t the right way to calculate lifetime risk of murder, when I apply it to the US population as a whole I’m not that far off from FBI numbers.

        tl;dr: probably a well-intentioned back of the envelope calculation that got way more legs than it should have, based upon a very restrictive definition of trans.

        That being said, if the murders that are being tracked are solely of trans people who have transitioned, we return to a figure that is uncomfortably close to 1 in 12. I’d want to run the proper numbers in that case.

        • Deiseach says:

          Looking at that article, there is one heck of a lot of guesstimation going on. The figure of “1 in 12” is based on one person’s derivation, the original of which is nowhere to be found anymore, or those who had copies can’t find it because they packed it away somewhere:

          Kay used to have a paper online which went through her derivation, but it is no longer available (as far as I can see). I may have printed a copy back then, but I have thousands of pages of old articles and they are not currently organized in any manner whatsoever. I’ve moved twice since they were even poorly organized.

          Then they fudge matters even more by mixing comparisons: they ignore the annual murder rate (because it would come out to a lot less even at the elevated risk of being a black transwoman sex worker) and work up a lifetime average by assuming that reported murders are only 10% of actual murders etc etc etc:

          But the comparison made between “the average American” and transpeople has some flaws. The “average American” figure is computed by dividing the number of murders in a year by the number of Americans, so that is really the chance of being murdered in a particular year. The transgender number is lifetime expectancy…the likelihood that the cause of death of a transperson is murder. Or, more likely, the likelihood that the cause of death of a transwoman is murder. Transmen do not die of murder nearly as often as do transwomen. And if the transwoman is a woman of color, the likelihood would increase. The likelihood for a transwoman of color who is a sex-worker approaches certainty.

          Yeah, well, it’s not any too great for cis black women sex workers either, or cis black young men.

          And then they take global figures because USA figures alone wouldn’t be high enough and work in a lot of assume this, that and the other:

          Last year (11/20/2008-11/20/2009), the worldwide compilation of transgender murders from the two lists I have as sources, the American list maintained by Ethan St. Pierre, and the more comprehensive worldwide list maintained by TGEU, there were 166 reported murders of transgender individuals.

          Estimates of the transgender population range from 1/10,000 (MTF) to 1/30,000 (FTM) – let’s estimate 1/17,000 overall.

          6 billion people in the world – That gives us around 350,000 trans people worldwide.

          Reported murders may be only the tip of the iceberg – one rule of thumb is to use 10% – making the 2009 TDOR statistic the equivalent of possibly 1,660 murders in one year. This results in an annual murder statistic of about 0.44%. If we estimate that 20% of worldwide murders are actually reported, it’s more like 0.22% per year.

          Life span for trans individuals skews short. Let’s estimate 50 years. If 1/500 is killed each year, that comes out to about 1 in 10 that can expect to be murdered during their lifetime.

          I can’t figure out her numbers but even if we accept her conclusion – “1 in 10 that can expect to be murdered during their lifetime” – that’s incorrect. It’s not “1 in 10 trans people will be murdered”, it’s “1 in 10 chance that the cause of death will be murder” which is a different matter entirely.

          The annual murder rate, as per the formula I quoted, would be:

          166/350,000 * 100,000 = 474 per 100,000 which is really high given that the global murder rate in 2012 was 6.2 per 100,000. Somebody’s figures are screwy somewhere!

          • Eltargrim says:

            I think you may have intended to reply to Nancy downstream. Overall I agree that the DailyKos article has it’s flaws, especially when they bring in global numbers.

          • keranih says:

            Reported murders may be only the tip of the iceberg – one rule of thumb is to use 10% – making the 2009 TDOR statistic the equivalent of possibly 1,660 murders in one year.

            WTF? Only one in 10 murders is actually reported? Where is this from?

          • Nornagest says:

            A 1:10 ratio for reported to actual murders in the US would be insane. If we were talking about rapes or robberies, that would be one thing, but there are going to be questions asked about a corpse found with holes in it, and murder isn’t a crime you have to press charges for.

            Globally, though, it’s more plausible — there are plenty of places in the world where most people spend most of their lives away from the eyes of government. And I’d expect murder rates to be higher in those kinds of places and among those kinds of populations. But for the same reasons, it’s highly questionable to make these kinds of extrapolations. We’re assuming that the excess deaths in the populations we know about are proportionally mirrored in the populations we don’t, and that’s a terrible assumption — we don’t know anything about these populations because of cultural and material differences (viz. they’re dirt-poor and often live in conflict areas), so the cultural and material underpinnings of violence there are going to be different, too.

          • Protagoras says:

            I thought it might be globally. I also considered that it might mean that only 1 in 10 murders of trans people were reported as murders of trans people. In either case, 1 in 10 still sounds awfully low to me, but I could be wrong (while on the other hand I am quite confident, like others in this thread, that the overwhelming majority of murders in the U.S. are reported as murders).

          • John Schilling says:

            Globally, though, [10% of murders being reported] is more plausible — there are plenty of places in the world where most people spend most of their lives away from the eyes of government.

            About a decade ago, I found myself looking into the various discordant claims regarding how many people had been killed during the Iraq war, and found that either A: the Iraqi government was capturing better than 90% of the actual homicides in its records or B: the excess were the subject of a conspiracy so pervasive that if you knock on someone’s door and ask “have any of your relatives been killed in the past year?” everybody in the village will deny it. And I don’t think hypothesis B is really credible.

            If the 2004-2005 Iraqi government can capture 90% of the homicides occurring in a literal war zone, I don’t believe there’s anywhere on Earth that 90% of homicides go unreported. That’s pretty much the #1 priority of even the most dysfunctional bureaucracies on the planet – compile a list of People Who Exist In Our Jurisdiction, and keep track of the changes.

            The true list may, in some countries, be classified.

    • Nancy Lebovitz says:

      Someone found a link for an earlier source for the statistic and put it in the comments to the patheos post. I do not recommend reading the comments.

      https://web.archive.org/web/20100312212415/http://www.hrc.org/issues/1508.htm

      http://www.dailykos.com/story/2012/11/16/1162442/-Danger-Will-Robinson-Danger

      This may be the original paper.

      http://www.sbequality.org/Transgender_Basics.rtf

      Here’s the section about hate crimes:

      “Hate violence. Transgender people are often targeted for hate violence based on their non-conformity with gender norms and/or their perceived sexual orientation. Hate crimes against transgender people tend to be particularly violent. For example, one expert estimates that transgender individuals living in America today have a one in 12 chance of being murdered.[1] In contrast, the average person has about a one in 18,000 chance of being murdered.[2]

      In 2002, community activists commemorated the lives of 27 murdered transgender people in that year.[3]
      However, many crimes against transgender people are not reported because of widespread doubts that state and local authorities will treat them with respect or investigate the crimes.

      Even in cases where the crime is reported, police response is often inadequate. In the hate crime on which the film “Boys Don’t Cry” was based, 21-year-old Brandon Teena was raped and later killed by two friends after they discovered he was biologically female. Teena had been living as a male and preparing for gender-reassignment surgery when he moved to Falls City, Neb., and befriended John Lotter and Tom Nissen. Upon discovering that Teena was biologically female, Lotter and Nissen became enraged and raped and beat him. Teena reported the crime to the police, but Richardson County Sheriff Richard Laux, who referred to Teena as “it,” did not allow his deputies to arrest the two men. Five days latter, on Christmas Day 1993, Lotter and Nissen found Teena in a farmhouse where he was staying with a friend, Lisa Lambert. They shot and stabbed him to death, then killed Lambert – in front of her 9-month-old son – and Philip DeVine, another friend in the home. JoAnn Brandon, Teena’s mother, filed a civil suit against Laux, claiming that he was negligent in failing to arrest Lotter and Nissen immediately after the rape. The court found that the county was at least partially responsible for Teena’s death and characterized Laux’s behavior as “extreme and outrageous.” [4] (Omaha World-Herald, April 21, 2001; The Associated Press, Oct. 5, 2001; The New York Times, April 21, 2001; Chicago Tribune, April 21, 2001.)

      In another apparent anti-transgender hate crime, Hugo Cesar “Bibi” Barajas was found dead in 2002 from multiple gunshot wounds near a Houston club that serves gay, lesbian, bisexual and transgender people. Barajas, who was anatomically male, was dressed as a woman at the time of the murder and was found in a halter top, blue jeans, silver high heels with straps and a woman’s wig. According to police, she was also wearing makeup. At the time of this writing, police were investigating the murder as a possible hate crime, but had no suspects, motives or witnesses. Since the murder, the Texas Gender Advocacy & Information Network and other national groups called for the state hate crimes law to be amended to include coverage for gender identity- or gender non-conformity-bias motivated crimes. The groups cited six other murders of transgender women in the Houston area since 1999.[5]

      Transgender people are frequently mistrustful of local law enforcement authorities because they often lack training and understanding of transgender people. This lack of understanding illustrates the need for a federal backstop for state and local authorities, particularly in cases where the local law enforcement authorities exhibit intolerance or fail to investigate or prosecute cases of transgender hate crimes.

      Sixty-eight percent of Americans believe that we need laws to protect against anti-transgender hate crimes, according to a poll commissioned by the Human Rights Campaign Foundation in 2002.”

      • John Schilling says:

        12 chance of being murdered.[1] In contrast, the average person has about a one in 18,000 chance of being murdered.[2]

        The average American person has about a one in 18,000 chance of being murdered per year, which comes to about one in 240 over a lifetime.

        I am skeptical of 1 in 12 for transgendered persons generally, but that would “only” be twenty times the normal incidence. I could buy 1 in 12 for MTF transgender sex workers specifically, but I’d want better support than I can find from these links.

        And I’d want anyone studying the issue to try and distinguish between people being killed for choosing a high-risk career and people being killed because they hold or express a non-traditional gender identity, because the solutions to those two problems are likely to be very different.

        • Nancy Lebovitz says:

          Add poc to the risk factors, apparently– and it may not be poc in general but black in particular.

          I’ve seen the point made that it’s harder for transgender people who look odd to find jobs in the mainstream economy, so they’re more likely than they’d otherwise be to do sex work.

          They may also be more likely than otherwise to be thrown out of their families as teenagers– this limits their job choices.

          • John Schilling says:

            I’ve seen the point made that it’s harder for transgender people who look odd to find jobs in the mainstream economy, so they’re more likely than they’d otherwise be to do sex work.

            I would think that “looking odd” would be a bigger handicap in the sex trade than elsewhere in the economy. My own hypothesis goes in the other direction, that for someone insecure about being perceived as a Real Woman(tm), a job where cis-hetero men actually pay them for sex provides a higher degree of validation than anyone’s polite affirmations.

        • Deiseach says:

          I think that derivation then phrases it wrongly – it’s not “1 in 12 trans people will be murdered”, it’s “the chance that death is through murder for a trans person is 1 in 12” and that’s different.

          Calculating lifetime mortality rates of homicide is complicated, because you’re adding up the probabilities of (a) being alive at an age cohort (b) chances of murder in that age cohort and I don’t want to even imagine trying to do it for the trans population.

          I think the general message we can take is that: for trans women who are black/non-white and working as sex workers, the likelihood of death by homicide is extremely high due to the associated risk factors. For the US trans population in general, the annual murder rate is 1.5 per 100,000 which is not bad by comparison with the general population (4.5 per 100,000) but which is also very high for the sub-population.

          I don’t think “murdered but not for specifically being trans” should be added in, as that would be the equivalent of, say, husband and wife murdered in burglary attempt by robbers and counting the wife’s death in “homicide rate for women” when you’re trying to show that women are victims of misogyny and so you want a rate for “women killed for being women”.

          • Nancy Lebovitz says:

            I still think it’s reasonble to include any murder that’s related to more poverty than would otherwise be expected for that person, but that would be hard to compute.

          • Skivverus says:

            …it’s not “1 in 12 trans people will be murdered”, it’s “the chance that death is through murder for a trans person is 1 in 12” and that’s different.

            Dunno, those mostly read like two ways of saying the same thing to me. Or at least, given the whole “100% mortality rate of the human condition” bit. Will absolutely agree that they have different emotional valence, though, and that this is relevant.

          • random832 says:

            I mean, comparing it to a baseline rate already factors out any question of whether any specific murder is attributable to being part of a demographic group.

            I mean, if the “homicide rate for women” is higher than the rate for men, then do you really need a “rate for women killed for being women” to show that misogyny is a problem?

          • Gobbobobble says:

            @random

            Maybe, maybe not. Could be that men (hypothetically) survive homicide attempts more often.

          • lvlln says:

            I mean, if the “homicide rate for women” is higher than the rate for men, then do you really need a “rate for women killed for being women” to show that misogyny is a problem?

            Obviously yes? The final homicide rate alone can’t actually tell us anything about the mechanism by which that rate is produced. There are many reasons why women and men might have different homicide rates that don’t depend on bigotry, such as personal risk preferences or employment options. I mean, if the homicide rate for men were higher than for women, would this show that misandry is a problem?

          • random832 says:

            There are many reasons why women and men might have different homicide rates that don’t depend on bigotry, such as personal risk preferences or employment options.

            Well, to some extent I’m not sure it’s not fair to describe differences in employment options that lead to more people dying as misogyny. Risk preferences are culturally conditioned. But I assumed such things would be controlled for in the hypothetical study being discussed.

            I mean, if the homicide rate for men were higher than for women, would this show that misandry is a problem?

            Why not?

          • lvlln says:

            There are many reasons why women and men might have different homicide rates that don’t depend on bigotry, such as personal risk preferences or employment options.

            Well, to some extent I’m not sure it’s not fair to describe differences in employment options that lead to more people dying as misogyny. Risk preferences are culturally conditioned. But I assumed such things would be controlled for in the hypothetical study being discussed.

            Then you’re just begging the question and equating misogyny (or other similar types of bigotry) with whatever cultural, social, economic, etc. forces that may result in different outcomes, with respect to homicide for women in this case. If that’s how you define “misogyny,” then OF COURSE higher homicide rates for women would show that “misogyny” is a problem, but that also tells us nothing, because you’ve radically redefined “misogyny” away from what it’s generally understood to mean by most people.

            It is perfectly possible that different genders have different employment options without a hatred of women (or men, for that matter) being a factor, and since it would be miraculous for any given 2 jobs to have the exact same risk for homicide victimization, it is perfectly possible that different homicide rates are not the result of a hatred of women. Same goes for risk preferences – cultural conditioning can affect people without resorting to hatred of women (or men), and any sort of affect can influence the homicide rate.

            I mean, if the homicide rate for men were higher than for women, would this show that misandry is a problem?

            Why not?

            I was curious, so I looked it up, and Wiki actually that this is the case in most of the world and also in the USA. This alone doesn’t tell me that misandry is a serious issue in the USA, because it’s perfectly plausible that men, through mechanisms that involve no hatred of men, are more incentivized towards jobs that put them at higher risk of homicide, or have fewer jobs open to them that put them at lower risk of homicide, for instance. Again, that’s unless you redefine “hating men” to describe any system which happens to result in them being victimized by homicide more.

          • dndnrsn says:

            Who says it’s about jobs? Men are far more likely to have extracurricular lifestyle choices that raise the homicide risk. I wouldn’t be surprised if the homicide risk is far higher among men with no or sporadic legal employment.

          • lvlln says:

            Who says it’s about jobs? Men are far more likely to have extracurricular lifestyle choices that raise the homicide risk. I wouldn’t be surprised if the homicide risk is far higher among men with no or sporadic legal employment.

            No one’s saying it’s about jobs. Jobs are just one of many possible mechanisms which could explain the discrepancy and which involve no misandry. As long as even one such mechanism exists, a discrepancy in the homicide rates cannot imply that misandry is a problem. That doesn’t mean that jobs are a factor, much less the factor. My guess is that it’s many factors, including genetic, social, political, economic ones.

          • Nancy Lebovitz says:

            “I mean, if the homicide rate for men were higher than for women, would this show that misandry is a problem?”

            Maybe. I think talk about what deals men and women get from the world is affected by the desire to look like a pure vctim who’s being oppressed by the other side.

            However, much of the damage done to men is by other men, and much of the damage done to women is by other women. (I’m inclined to think that men do most of the violence to both men and women, but there’s a tremendous amound of emotional abuse all around.)

            It’s certainly possibly to blame intra-gender abuse on the other gender– for example a claim that wars happen because “women love a man in a uniform”. I’m not making this up– I heard this claim in person.

            Sorry, I don’t seem to have a concluding point, so I guess this is just an observation of something which makes this a difficult topic.

          • Aapje says:

            @Nancy Lebovitz

            It’s certainly possibly to blame intra-gender abuse on the other gender– for example a claim that wars happen because “women love a man in a uniform”. I’m not making this up– I heard this claim in person

            Well, women were sent out into the streets in Britain during WW1 to shame men into signing up for the war 🙂

            But I think that it’s far more accurate to state that societal status is not gendered in that both men and women value and reward behavior that is high status. A man who displays controlled violence demonstrates ability to protect, which is generally desired by women and in turn, men who are desired by women have more status among men. But controlled violence is not just desired by women, but by the collective as well (especially the gender that is tasked with protection mostly), so it is also high status independent of women’s desires. If women would vanish, groups of men would still need to be able to protect themselves from other groups. However, because women do exist, this high status in society in turn makes those men more attractive to women.

            Ultimately, status is far from as simplistic as people (feminists and anti/non-feminists) tend to portray it, with feedback loops and indirect effects that makes it silly to blame one gender exclusively.

          • random832 says:

            @Nancy Lebovitz

            Maybe. I think talk about what deals men and women get from the world is affected by the desire to look like a pure vctim who’s being oppressed by the other side.

            However, much of the damage done to men is by other men, and much of the damage done to women is by other women.

            I’m not sure how this is a problem for the use of the word “misandry”, which I understand to mean “sexism against men, as distinct from sexism against women” (and vice versa for “misogyny”).

          • Nancy Lebovitz says:

            The White Feather Campaign wasn’t about typical female desires. It was orgnaized by a man and only happened one war.

            Also, it doesn’t exactly prove that women love a man in uniform, just that they’re willing to dump on men not in uniform.

            I’ll note that women are also attracted to musicians and to (successful?) atheltes.

            I’ve only read about half of the article.

            In the hopes of cheering at least some people up, a song about a woman with an unusual preference.

          • Artificirius says:

            With respect to the White Feather campaign, the sentiment is most assuredly far older than the First World War. ‘Come back with your shield or on it’ springs to mind.

            On an unrelated note it is so exaperating to see claims of ‘women were utterly powerless!’ juxtaposed with instances of women wielding very real, very definite power over the men around them with such myopic dishonesty. At least this article tries to claim it was only newfound power. (Balderdash.)

          • Nancy Lebovitz says:

            Artificirius, “come back with your shield or on it” is weak evidence– it’s from Plutarch writing about Sparta. In other words, it’s not a first person account, and it’s about an unusually militaristic culture. Furthermore, it’s from mothers to sons, so it isn’t actually about mating success.

            War is close to a human universal, and a lot of the soldiers aren’t volunteers. There is no reason to think that women being attracted to soldiers is a primary cause of wars.

          • Artificirius says:

            I’d heard/read that it was something said by mothers or wives to the soldiers as they were given their shields. Regardless, I wasn’t trying to defend the argument that women are the primary cause of wars. I had missed that in the quoted text, and was just disagreeing with the notion that female disapproval of cowardice and related push towards enlistment was limited to the First World War. Such behaviour is also mentioned in recounts of the American Civil and Revolutionary wars, for instance.

          • Aapje says:

            @Nancy Lebovitz

            At a minimum the White Feather campaign shows that women had this kind of power over men at the time.

            You have to keep in mind that perception of war drastically changed due to the meat-grinder that was WW 1. Before WW 1, war was a chance for a relatively small percentage of the population to prove their manliness. They may have suffered, but their home nations generally faced very little. Furthermore, the soldiers would of course boast after coming home or would feel ashamed to have lost and keep quiet. So there was little perception of the nastiness of war beyond ‘a few people die/get wounded and it can bog down and cost too much money.’

            WW 1 introduced total war. A change from glorious short battles with cavalry sweeping in, drums playing, honor among men, yadda yadda, to people stuck in the mud getting shelled for hours upon end, suffocating from poison gas, in rags, running hopelessly into machine gun fire, etc. The facade of glory could no longer be held up.

            Furthermore, photographs showed people the things that were once unknown and the suffering was so bad that men starting bailing on their masculine norms and actually started speaking out honestly. Women served as medics on a large scale for the first time and saw the worst injuries up close (people who in previous wars would generally never make it back home). People began to realize that death and fairly basic physical injury is not the worst that war can give. It can destroy minds, subject people to suffering that they actually prefer death. Furthermore, it can cause death and injuries on a scale that is devastating emotionally and to society far beyond if a limited tragedy happens; where most families are not affected and those who are generally have 1 or 2 deaths to mourn. In WW 1, you had parents who lost 4, 5, 6 sons in the war. For morale reasons, sons from certain areas would often serve in the same unit, so they would feel among friends. Yet this meant that if their unit was decimated, entire streets lost most of their sons in 1 go.

            It is a mistake to take the feelings that developed on the home front due to this and judge them to have existed forever.

  10. albertborrow says:

    I was reading Krixwell’s liveblog of Worm the other day, and I stumbled across this reaction (paraphrased):

    I thought it was racist to describe the skin color of black people based on food…

    This is in response to Wildbow describing one of his characters as having “chocolate skin”. I know that both Tumblr is a big fan of the SocJus sphere, but I’ve never seen a single quote that sums up my problems with it as well as this one does. First gripe: food? Food is racist? I can understand the sort of “paternalistic” or “patronizing” aspect of describing someone as a delicacy, but that doesn’t seem like the kind of racism we should be combating, even if it was intentional. Besides, this is hardly limited to just black people – if I had a dime for every time a white person had “cream-colored skin” or was “pasty” I would be pretty rich. (I understand that “pasty” has different connotations when used as a description, but to me it still conjures the image) Second, this is not the context in which the word was actually originally offensive, as far as I can tell. Despite this, people still say things like:

    I belong to different groups which talk about diversity in writing. One thing I have heard that suprised me is that using words such as “chocolate” and “caramel” can be considered demeaning. . .

    I can understand this if you’re just combating cliche, but that isn’t the context these are presented in. It’s that the phrase itself is offensive in some marginal way, like a taboo word or an ancient slur. Moreover, it’s not fair to expect disproportionate originality from the author – if something is too much of a cliche on those grounds, then you should be rejecting a large swath of other idioms. The wonderful thing about cliche is that you don’t notice it unless you’re trained to, and so it flies over the reader’s head when the writer gives us a “sharp jolt” or a “cold look”.

    And sometimes original prose can even be overtly racist. Could you imagine the uproar if someone wrote:

    She was like a trash-bag: dark in complexion and rotten to the core.

    Or:

    His skin was like an LCD screen before it had turned on.

    The point is that if you have nits to pick over an author’s work, actually put forward the effort to pick nits, rather than waving a fly-swatter at their face. Calling phrases like that racist is an overreaction and oversimplification, even if you shuck them under the category of “mildly racist”.

    • Dr Dealgood says:

      This seems like one of those things which depends a lot on context and on the particular person you’re talking to.

      For example, one of my friends has expressed a strong dislike of people describing her in food terms (e.g. chocolate) on the grounds that it’s gross and dehumanizing. On the other hand, nearly without exception*, black girls I’ve slept with have used food-color adjectives when referring to themselves. Especially when they find out I take my coffee black (god damn it’s such an unoriginal joke).

      So IME, it’s not really the kind of description you’d necessarily break out with a coworker or a friend but fine for talking dirty. If my notion here is right then you’d expect to see less criticism of characters being described that way in smutty scenes than in normal description.

      *PROTIP: Dominican women really don’t like it when you imply that they’re black.

      • Corey says:

        it’s such an unoriginal joke

        I think there are plenty of good twists out there; I like my coffee like my (wo)men:

        – Strong and bitter
        – Finely ground and stored in the freezer
        – Taken from a third-world country and sold to me for $8 a pound

        etc.

        • Dr Dealgood says:

          I am fond of “Columbian and tied in a sack” myself.

          Maybe that should be my solution: respond to lazy jokes with legitimately funny retorts. I feel a bit embarrassed not to have thought of that. If I end up in that situation again I’ll try it out.

        • Le Maistre Chat says:

          – Drunk in the morning.
          – Served by an entitled bearded dude.
          – Coming from a $6,000 machine.

        • Cerebral Paul Z. says:

          Probably a joke about K-cups to be made here as well.

        • Whatever Happened To Anonymous says:

          Out of a forest cat’s ass?

          … Work in progress, I guess.

        • lvlln says:

          One of my favorites was “hot, and with really really big tits,” which I read at Fark many years ago.

      • The Nybbler says:

        (god damn it’s such an unoriginal joke)

        “…and covered with short brown fuzzy hair”

        • Nornagest says:

          I kinda like the “softened to death” one, assuming it’s not taking itself too seriously. Sounds like something Terry Pratchett would have written.

      • Le Maistre Chat says:

        god damn it’s such an unoriginal joke

        Do you like women the way you like coffee, filled with milk and whipped? =)

        (It probably says something terrible about me that I’m having this much fun with it.)

    • Civilis says:

      Here we have a practical problem: there’s no color that you can use to describe people’s skin color without it being interpretable as dehumanizing, because any frame of reference color is by nature going to come from something other than human. It’s the reverse of the ‘flesh color’ problem, where it was problematic to call one color ‘flesh’ because humans come in so many different skin colors.

      If you want to describe a color in words, you are going to need something in the real world people can visualize. For most colors, that’s going to be a relatively common substance. Take a look at the names of colors (https://en.wikipedia.org/wiki/List_of_colors_by_shade). Most of the more vivid shades are named after plants (often either flowers or fruit) or valuable minerals (precious metals or gems).

      Brown shades, being less vivid, are harder to describe, but are generally pigments (often themselves named after places in Europe where they were synthesized), plants or plant products, or animals… all of which provide potential problems when used to describe people. Per Wikipedia: “Words for the color brown around the world often come from foods or beverages; in the eastern Mediterranean, the word for brown often comes from the color of coffee; In Turkish, the word for brown is kahve rengi; in Greek, kafé, in Macedonian, kafeyev. In Southeast Asia, the color name often comes from chocolate: coklat in Malay; tsokolate in Filipino. In Japan, the word chairo means the color of tea.

      [Edited to add: I didn’t even think of the potential culture war tangent when I posted this. It’s just “how do you describe colors in words?” becomes an interesting question, and “where did the colors get their names?” becomes its own wiki-walk almost as interesting as discussing warships, at least for me.]

      • Le Maistre Chat says:

        That’s interesting, because it implies that Greek, Turkish and South Slavic had no word for brown until the late Middle Ages, and in Southeast Asia not until the 16th century.
        This leads me to bring up color in ancient Greek, where the sea was wine-color, the sky bronze, earth black…

        • random832 says:

          This article briefly mentions Japanese:

          Ancient Japanese used the same word for blue and green (青 Ao)

          One interesting artifact of this is that Japanese traffic lights are described as, and occasionally actually, blue. (A google image search suggests that the usual color is still a color that I would call green, with only some outliers being blue, but a bluer shade of green than I’m used to in the US)

          • lvlln says:

            Don’t know if it’s for similar reasons, but the same phenomenon happens in Korean as well. “푸른색” refers to both green and blue in different contexts (generally blue, but green for plants), and I’ve heard some Koreans refer to the green traffic lights as blue.

          • Evan Þ says:

            That might actually be a good idea, given red-green colorblindness? It’s actually less common in Japanese than Europeans (4% against 8%), but still somewhat significant.

          • random832 says:

            The phenomenon is notable enough to have its own Wikipedia article, which also mentions, among other languages, Greek.

            Incidentally, there are some doubts that the sky was considered to be bronze in color, or the sea wine red.)

          • shakeddown says:

            Conversely, Hebrew has separate words for Blue and Light Blue. (in common usage. Technically English has all sorts of words like teal, but I’m discounting those).

            …actually, does Japanese have any somewhat-obscure words for the particular blue or green shades of bleen?

          • random832 says:

            Modern Japanese has 緑 midori, which means green in most contexts. Wikipedia has a largish list of traditional color names, but no idea how common those are in modern usage.

          • onyomi says:

            Yes, but you do have to say “midori-iro,” as opposed to just “ao.”

            Chinese, Japanese, and, I think, Korean, have a division between the few “true” colors, usually black, white, red, blue/green, and yellow/ochre (related to “five element” theory), which can be standalone adjectives, and words which mean, essentially, “(orange, tea, coffee…)-colored.”

            In Japanese: kuro, shiro, aka, ao (no standalone ki-iro, seemingly), but midori-iro, cha-iro, bara-iro, murasaki-iro…

            Almost none of the words in the linked list are in common usage in Japan today (though people would know what you mean if you said sakura-iro), at least no more than most Americans really know what mauve and taupe mean.

          • Civilis says:

            I’ve actually learned a surprising amount from this thread.

            Conversely, Hebrew has separate words for Blue and Light Blue. (in common usage. Technically English has all sorts of words like teal, but I’m discounting those).

            Thinking about it, indigo is considered distinct enough from blue to have its own place in the traditional ‘ROYGBIV’ rainbow. Interestingly, Wikipedia cites Aristotle and Newton as being behind the division of the rainbow in to seven colors. (I love trivial, non-political topics, as it’s a place where Wikipedia still has a use.)

          • random832 says:

            > Thinking about it, indigo is considered distinct enough from blue to have its own place in the traditional ‘ROYGBIV’ rainbow.

            That’s its own piece of oddness, since, to my understanding “indigo” is actually the blue primary in the additive color system, and the “blue” of the traditional rainbow is more greenish.

            Wikipedia confirms my understanding:

            Later scientists conclude that Newton named the colors differently from current usage.[19][20] According to Gary Waldman, “A careful reading of Newton’s work indicates that the color he called indigo, we would normally call blue; his blue is then what we would name blue-green, cyan or light blue.”[21] If this is true, Newton’s seven spectral colors would have been:

            But of course the point stands that the colors are considered distinct, whether they’re called blue/indigo or cyan/blue.

            —-

            Incidentally, I’ve experimented with refusing to acknowledge orange (i.e. mentally classifying every orange object I see as either red or yellow), maybe I should try doing the same with green.

          • Jaskologist says:

            I’ve experimented with refusing to acknowledge orange

            Any particular goal in mind for this experiment?

          • rahien.din says:

            Civilis, random832,

            To throw some doubt on Newton’s color scheme, he probably just wanted the number of colors to match the notes in the major scale. From K. McLaren. Newton’s indigo. Color Research & Application , Volume:10 , Issue:4 , Page(s):225-229

            Of all the many explanations put forward to account for Newton’s addition of indigo to the five principal colors, red, yellow, green, blue, and violet, the most likely one is that he wished to develop an analogy between colors and the musical scale. This required the addition of two less-prominent colors, one between red and yellow, the other between blue and violet. Orange was, and still is, the appropriate name for the first. Indigo was the appropriate name for the second until the end of the 18th century when it fell into disuse.

            Le Maistre Chat,

            There is a body of work describing the stage-wise acquisition of basic color terms by languages. In short, according to Berlin and Kay, a language goes through five stages:
            1. Light from dark
            2. Light, dark, and red (including yellow).
            3. Choice of A, B, or C
            – 3A. White, red, yellow, black/blue/green
            – 3B. White, red/yellow, green/blue, black
            – 3C. White, red
            4. White, read

          • rahien.din says:

            Somehow I accidentally posted the comment and I can’t edit it (?)

            – 3C. White, red, yellow/green/blue, black
            4. White, red, yellow, and either green/blue and black or green and blue/black.
            5. White, red, yellow, green, blue, black.

            Their work is not uncriticized, but it offers an interesting explanation of things such as “the wine-dark sea.”

          • Douglas Knight says:

            rahien.din, how does that help with Homer? For the sea to be the color of wine, it would have to be a stage 1 language. But ancient Greek has lots of color vocabulary. random832’s last link show that it is at least stage 4.

            On the other hand, the reconstructions of the colors in the link do seem to group into 2 dark colors and 4 bright colors (plus the less certain purple). Maybe the language was jumping from stage 1 to stage 4/5. A century before Berlin and Kay, PM Gladstone proposed that Homer cared a lot more about brightness than hue. The traditional translation as “wine-dark” is pretty much an endorsement of the importance of brightness. But the big problem is whether “wine-looking” even refers to color at all.

          • rahien.din says:

            Douglas Knight,

            Thanks. Personally, I agree that Homer was likely describing the sea’s darkness rather than its hue. If we are to give any credence to Berlin and Kay, then the emphasis on brightness would (as you suggest?) place Homer’s Greek in an early stage.

            Moreover, the post preceding random832’s link gives the following list :

            [Dark colors] Kyaneos covered the area of the spectrum ranging from black towards dark blue, and was also used for what in English would be called ‘black’ hair. Melas covered the range from black to brown to dark orangey-red. Porphyrios referred to a range of deep vivid colours
            [Light colors] Glaukos included light shades of green and blue with relatively low saturation. Chlōros and glaukos divide up pale colours between them.
            Erythros (‘red’) and leukos (‘white’) are just about the only ancient Greek colour terms that mostly line up with their conventional English translations.

            …which seems to support the overall primacy of the bright/dark distinction, with only their term for red designating a particular hue. Sort of a stage 2.5? (I’m thinking out loud here.)

            But, you may be right and/or I may stand corrected. I am doubtless unqualified to resolve any such ambiguity.

            Interestingly, random832’s link describes that both melas (black) and erythros (red) were used to describe wine. If we must presume it describes a hue at all, perhaps “wine-dark” could mean either “wine-black” or “wine-red”?

            Anyway, all in good fun. Fascinating topic.

          • Jaskologist says:

            If Homer was blind, maybe we shouldn’t read too much into his color descriptions.

        • episcience says:

          There is a really good article on colour words, and how different cultures seem to take the same path when coming up with new ones, here:

          If a language has just two color terms, they will be a light and a dark shade – blacks and whites. Add a third color, and it’s going to be red. Add another, and it will be either green or yellow – you need five colors to have both. And when you get to six colors, the green splits into two, and you now have a blue. What we’re seeing here is a deeply trodden road that most languages seem to follow, towards greater visual discernment

          Part two goes into how this can affect perception — so tribes with distinct words for different greens will be better at telling those greens apart than English-speakers.

    • Deiseach says:

      There are food- and plant-based descriptions for white people: peaches and cream complexion, cherry lips, the lily and the rose in her colouring, etc.

      If you’re describing someone who is brown-skinned but not “tanned white person”, then you have a limited range of expressions: food, or metals like bronze, copper; describing someone as “clay-coloured” or “brown as the dirt” might be considered even more offensive!

      Unless we’re going to use paint colour charts, in which case good luck describing your hero as “his hands were pure fiber in the winter, turning to pure soba after a month working outdoors in the summer sun”.

  11. dodrian says:

    Does anyone wear a smartwatch? Would you recommend it?

    I needed a watch for my last job (it would be viewed as unprofessional to keep pulling out my phone to check the time), but had never been able to keep one for more than three weeks or so (lost or broke all of them). I figured that if I bought a watch which was also a toy I’d take better care of it.

    It worked – I loved my Pebble and later upgraded to the Pebble Time, but after over a year and a half of faithful service my bad watch luck finally caught up with me – I’m not sure but I think my watch may be somewhere at the bottom of the pacific ocean.

    Unfortunately, at the end of last year Fitbit bought and unceremoniously axed Pebble. I could pick up some of the remaining stock at very cheap, but run the risk that with no more Pebble software updates a future Android update could just break everything.

    What I loved most about the Pebble were the week+ battery life, and that it was easy to write or modify an app for the watch. And after a few weeks with no watch I find that I really miss it – not just telling the time but being able to see who was calling me or if that text was important without fishing for my phone.

    So, do you wear a smartwatch that’s not a Pebble? Is it any good?

    • Corey says:

      I like the Garmin Vivoactive, its battery also lasts a really long time, assuming you leave the watch’s GPS off (and unless you’re logging exercise without your phone on you, no point in having it on). The one I have is the non-heart-rate-measuring variety, so the Vivoactive HR may have less battery life.

      The only other smartwatch I’ve had is the Basis Peak and I loved that, but they bricked them because of a tendency to occasionally run the CPU in tight loops, leading to burns. (I never experienced that, it was rare but they couldn’t find a way to reliably fix it).

      My use cases, in frequency order:
      – See text messages / calendar notifications from phone
      – Alarm clock (buzzes me awake)
      – Telling time
      – Sleep tracking (how’s my CPAP doing? If I feel low will-to-live, is it just because of sleep deficit?)
      – Tracking steps (dude, too many steps today, take a break and have a burrito)

      • dodrian says:

        Fitness is the one thing I haven’t used mine for, but it appears to be the one thing the market has decided that sells. I’ll take a look though, thanks!

      • random832 says:

        Just out of curiosity, how is it with the GPS on? I’ve noticed that my phone runs really hot (presumably consuming a lot of power) with GPS on, and I’m curious as to whether that’s a problem with the phone or GPS is inherently power-intensive.

        • The Nybbler says:

          This Usenix paper gives a maximum number of 160mW for GPS

          https://www.usenix.org/legacy/event/atc10/tech/full_papers/Carroll.pdf

          That’s a lot by phone standards but not enough to make it run really hot.

          In modules I’ve measured, I’ve seen 200mA of power use (at 3.3V IIRC) before acquisition, about 100mA after; these were prior generations of receiver, however. I’d guess something’s wrong with youts.

        • dodrian says:

          What else are you doing with it? A Maps application is also CPU, display, and network heavy, so it sucks power and heats up.

          If it’s just recording lon & lat it shouldn’t be that power hungry. But it might also be crunching numbers in the background to give you a better location (accelerometers, nearby networks, various derivatives and what have you).

        • Machina ex Deus says:

          Is your battery old? Undervoltage can make a lot of electronics run hot. Years ago, I had a two-hundred-core graphics card that kept overheating, until I got a regulated power supply to plug the host computer into.

        • Corey says:

          I tried an experiment, turned on GPS and left it on from 10:00AM yesterday to 7:00AM today. Used about 3% of battery (it had 97% when I dropped it in the charger at 7) and no noticeable temperature increase. My guess is that in recent firmwares, it only actually uses the GPS when you tell it to log a workout.

          Another data point, from my Android phone: I leave location enabled all the time, with no big impact on battery life or temperature. When LeafSpy logs altitude it doesn’t seem to use / dissipate much power. Pokemon Go, though, is probably dissipating more than 10W, the phone gets hot and still (slowly) loses battery power plugged into a 10W charger.

    • Well... says:

      I don’t wear one. My advice (as with all other technology) is to think very hard about why you want one, whether you really need it, how it might change your life in ways you weren’t asking for, and what all the costs are besides just the money you pay for the watch itself.

  12. Corey says:

    Kevin Drum went and looked at how long FDA approvals take, finding a big downward trend over the last 10 years (and upward trend in the approval rate) and that we’re rather quicker than Europe. What gives?

    Maybe it’s still too long? I also notice cost isn’t included (maybe we approve 15% faster than Europe at 5x the cost or something).

    • Douglas Knight says:

      There is no difference in cost between FDA and EMA. Filing costs are trivial. The cost is overwhelmingly the cost of the trials themselves and the two agencies have the same standards of what should be done in the trial. (And if they didn’t, people would meet both standards, because the markets are too big to pass up.)

      Drum compares times between the two agencies, but only gives approval rates for the FDA. Is this a change in the FDA, or is it a change in the quality of drugs being offered? My guess is that the approval rate per drug did not substantially increase, nor that it was ever very different from EMA, but that the old low approval rate reflected drugs being rejected (CRL) and trying again, while now they get approved in one try. But if you counted such drugs only once, their (old) time to acceptance would be much longer. There is a tradeoff between the two measures, depending on your definitions.

      You can consider approval time and approval rate as costs. Since the average patent life post-approval is 10 years, decreasing approval time by 1 year increases drug value by 10%, which small but a lot larger than, say, the salaries of the people dealing with approval, both at the drug company and at the FDA. Doubling approval rate per drug doubles the value of an unapproved drug, which is big, but I doubt that is what happened, as I said above. Probably this all should be thought of as a larger decrease in approval time than 1 year, so more than a 10% increase in value, but less than doubling.

      If you believe that drugs have the effects that the FDA claims they have, delaying them by a year has a large public health effect, much larger in aggregate than the effect of the few catastrophic approved drugs. You probably shouldn’t believe these claims, but neither is there any evidence that the delays produce new analysis or information, or even a change in what drugs are approved. It is not at all clear to me, even having observed some submissions, why it takes 10 months to do the preregistered analysis. Probably the time could be cut in half with more staff.

      A clear problem with the FDA compared to the EMA is that it is fickle, probably because its review process is more public. Often it will reject a drug because it increases some specific cancer by an amount that is statistically significant, but not controlling for multiple comparisons. Moreover, the amount will not be clinically significant and the FDA will say that cancer is so scary, it doesn’t matter how many lives the drug saves from heart disease, it’s an unacceptable trade-off. Or they will say: let’s wait a few months and hope that one person in the control arm gets the same cancer and we can ignore the problem (statistical malpractice, but probably the FDA knows that the problem was bullshit, but it has to convince the public review board). It could be that both trends are due to reducing this nonsense, which is a clear win. But if it is now comparable to EMA, it is less obvious how to improve things further. (But we don’t know that it is comparable, because Drum doesn’t graph EMA approval rates, let alone compare approvals of the same drug.)

  13. Nancy Lebovitz says:

    A gender-swapped version of excerpts from the presidential debates

    Actors did a meticulous job of imitating the candidates– a woman for Trump and a man for Clinton. There were audience surveys and discussion.

    • dndnrsn says:

      From the link (this is from the interview with Salvatore):

      What did you find most surprising?
      I was particularly struck by the post-performance discussions about effeminacy. People felt that the male version of Clinton was feminine, and that that was bad. As a gay man who worked really hard, especially when I was younger, to erase femininity from my body—for better or worse—I found myself feeling really upset hearing those things. Daryl [the actor playing Jonathan Gordon, the male Clinton] and I have talked about this multiple times since the performances. Never once in rehearsal did we say, “play this more feminine.” So I think it was mostly the smiling piece—so many women have told me that they’re taught to smile through things that are uncomfortable. It’s been really powerful to hear women talk about that, and a learning experience for me. I was surprised by how critical I was seeing [Clinton] on a man’s body, and also by the fact that I didn’t find Trump’s behavior on a woman to be off-putting. I remember turning to Maria at one point in the rehearsals and saying, “I kind of want to have a beer with her!” The majority of my extended family voted for Trump. In some ways, I developed empathy for people who voted for him by doing this project, which is not what I was expecting. I expected it to make me more angry at them, but it gave me an understanding of what they might have heard or experienced when he spoke.

      The Clinton surrogate comes off as effeminate in the chipper smiling, but also in his overall way of speaking. It sounds like there’s some uptalk going on. There’s something in the timbre of his voice. Also, what with Clinton came off as smartest-one-in-the-world wonkishness, comes off with this guy as fussiness.

      The Trump surrogate – the way she speaks is very husky/purring sometimes. It’s like, if this is what a woman from New York sounded like while trying to seduce someone by berating a political opponent, this is what that would sound like. Case in point: “such a nasty man” at 33:33 of the condensed version that’s up on Youtube. In fact, for that, just cut out the “berating a political opponent” bit. It sounds like playful banter with a sex partner.

      • beleester says:

        The full version has been taken down, so I can’t comment on that line. But the clip in the article certainly didn’t sound seductive to my ears. Female!Trump sounded… shrill? Angry and berating? Like, I finally understand why every hit-piece on Clinton included some variation on “shrill liberal protests.”

        The Trump penchant for hyperbole certainly shone through regardless of gender.

        The Clinton surrogate comes off as… artificial. I can’t put my finger on what made his delivery different from Clinton’s, I think you’re right it’s the timbre of his voice, but everything about him screamed “generic politician.”

        I wish they had the original clip from the debate for comparison. (Yes, I could probably find it myself, but it’s late and I can’t be arsed.)

        • dndnrsn says:

          There was a 35 or so minute condensed version on Youtube. Can’t find it now. The Trump surrogate’s voice would go up in pitch periodically, but sometimes would be quite low for a woman.

    • knownastron says:

      Is there only the 2 minute video? Is there a longer version?

      • Nancy Lebovitz says:

        They seem to have taken the longer videos down. I don’t know why, or if the videos will be made available again.

    • Nyx says:

      You have to wonder how much of the reaction to the debates (as initially performed by Man-Trump and Lady-Clinton) was shaped by preexisting perceptions of the candidates. It may not be so much that the genders are swapped, but when you put Clinton’s words in the mouth of a person that you’re not already disposed to like, they come off as phony and unconvincing.

      I think that it’s an interesting idea, though. It might be that women underestimate how willing people are to accept stereo-typically masculine and “assertive” behavior coming from them.

      • cassander says:

        This was precisely my thought. this shows less about perceptions of men and women that it does about perceptions of hillary and trump, with much of the emotional baggage stripped away.

      • Deiseach says:

        The actors copied the mannerisms of Clinton and Trump as well as the words, according to the originators, so it probably does show up the reason Hillary gets accused of being robotic or of trying so hard to appeal to all comers that she does come across as ‘generic politician spouting platitudes’. She does remind me of Maggie Thatcher in this, that both of them were told to work on how they presented themselves in public and Mrs Thatcher worked really hard on this (including learning to lower her voice from its natural range so as to sound less “shrill” and more authoritative).

        Hillary has none of Bill’s charisma and left to her own devices probably would come across as “lecturer addressing a class of not particularly bright students”, so she does have to try harder to give the appealing, confident, ‘trust me I’m solidly in agreement with you’ public presentation, which does mean a certain degree of artifice and this shows up more clearly when it’s a stranger copying her and we can see the performance unaffected by how we feel about Hillary herself.

      • dndnrsn says:

        I think that here the contrast made it. People probably prefer a woman acting masculine over a man acting feminine. Compare the connotations of the word “bitch” when applied to a woman, and when applied to a man. There’s a contempt there in the latter usage that’s not in the former.

        • shakeddown says:

          In some ways, being masculine is (perceived as?) as a tough duty that has to be done. Women who act masculine are stepping up (this can occasionally be negative, in the sense that you may not want unqualified people doing it, but usually isn’t). Men who don’t are refusing to step up and do their duty.

        • Aapje says:

          It also depends on the context. Both men and women have duties that can be shirked and duties that are not their domain. If they forgo the former, the response is usually much harsher than when they step into duties that are ‘not theirs’. Although the latter is often seen as causing the former.

          It’s even more complication because these ‘superweapons’ then get deployed for other causes as well. For example, ‘slut’ can be used regardless of any evidence to target a woman. “Living in Mom’s basement” can be used similarly against men.

          These insults hit at the insecurities of people. Insecurities pushed onto people by gender norms.

  14. Deiseach says:

    I’m not sure if this is culture-warring or not, but re: Trump’s latest pronouncement about Obama wire-tapped him. I’m seeing this attitude towards his insistence it happened: it’s a no-win for him because either it’s false and he sounds crazy, or it’s true and he’s a genuine security threat.

    But remembering that he appears to be cursed with luck (the mockery over the Sweden thing followed by rioting, The Intercept and its article which was then followed by one of its ex-reporters doing exactly what they had denied was being done), suppose he does get this investigated and it turns out that there was no wiretap because no judge would give a warrant (that’s assuming that some agency even looked for one in the first place).

    Doesn’t that then undercut the “Trump is a Russian plant/puppet” narrative? After all, if there isn’t any evidence implicating him being entangled with the Russians before/during the election or no state agency even thought of wiretapping him, then he’s not the huge security risk everyone is going on about.

    What better way to put an end to all the constant speculation and accusations of Russian involvement and talks with Russians and people in his administration having meetings with Russians and undue Russian influence than to have it turn out that this was considered and rejected by state intelligence agencies/not enough evidence to convince a judge to give a warrant was found prior to the election?

    • The Nybbler says:

      The Sweden thing was definitely luck. The false-flag hate crime only partially luck; I think the odds that the threats were fake were better than even (and I think the same about the so-far-unaccounted-for threats). There have been a LOT of hoax and false-flag hate crimes lately. The luck comes in that it was actually an ex-Intercept reporter.

      I don’t think he’s playing n-dimensional chess with the wiretap thing. My best guess to how this plays out is his opponents deny it and insist he’s absolutely crazy to even suggest it, then evidence comes out that doesn’t prove his claim but demonstrates that his opponents denials are far from reality as well.

      After the Snowden revelations and the IRS abuses, anyone rejecting these accusations out-of-hand is putting themselves _way_ out on a limb.

      • random832 says:

        Can someone explain to me the “IRS abuses”? From everything I can find, it seems to have been an attempt to find political organizations using a tax-exempt category that it is illegal for political organizations to have, by using their names to determine whether they were political organizations. And there’s a distinct lack of discussion as to what left-wing political organizations should have been targeted.

        If there was a disproportionate impact on right-wing organizations, I think that requires proof that right-wing organizations aren’t actually disproportionately likely to fraudulently apply for tax-exempt status for it to be actual misconduct.

        • houseboatonstyxb says:

          If there was a disproportionate impact on right-wing organizations, I think that requires proof that right-wing organizations aren’t actually disproportionately likely to fraudulently apply for tax-exempt status for it to be actual misconduct.

          One factor you’d have to consider, is that right-wing organizations have more money to hire lawyers to get their paperwork right in the first place.

          • random832 says:

            Regardless, even calling them “right-wing organizations” seems to be conceding the point that they’re guilty as hell – the only misconduct on the IRS’s part was, then, in not equally targeting their supposed left-wing counterparts.

          • hlynkacg says:

            Well Koskinen admitted that they had been specifically targeting organizations critical of the IRS. I don’t know about you but that sounds like “abuse of power” to me.

        • Douglas Knight says:

          First of all, it was about both 501c3 and 501c4 organizations. Political organizations can be 501c4. The single most famous lobbying organization, the NRA, is a 501c4. This is not a new Citizens United thing! The rule is only that a 501c4 must spend less than half its budget on lobbying, but it can spend the rest on educating its members about politics, thus be 100% political. (There are also 501c3 charities that are 100% political, like part of the League of Women Voters, but they are held to a high standard. of being non-partisan)

          Second, the public doesn’t really know what happened, because the IRS is secret. How many of the groups were 501c3 vs 501c4? I don’t know, though some sources say it was mainly 501c4. Some people say that the IRS flagged “tea party” and should have flagged “occupy.” But some say that it did flag both. Some say that it flagged both, but then approved the occupy organizations, while leaving the hundreds of tea party organizations in limbo for years.

          • Aapje says:

            Basically, people interpreted it based on their biases and we don’t really know the truth.

    • Jiro says:

      “It’s true and he’s a genuine security threat” is an unduly narrow option. What about “it’s true (in that the wiretap actually happened) but it was overreach”?

      • Matt M says:

        Yeah, this.

        People on the right are comparing this to Watergate. Dismissing it by saying “well obviously someone only gets wiretapped if they’re a serious security threat who has clearly committed a serious crime!” requires quite a lot of…. mental agility and flexibility, shall we say.

      • What about “it’s true (in that the wiretap actually happened) but it was overreach”?

        Then Trump wins twice. “it was overreach” implies that there wasn’t evidence of treason and “it actually happened” implies that everyone indignantly denying it, from Obama on down, was lying or at least (“it happened by not at Obama’s explicit orders”) being deliberately misleading.

    • J Mann says:

      – If Trump was correct and Obama ILLEGALLY influenced a wiretap, that would be astonishing and a national scandal on the order of Watergate. But Trump’s own people concede he doesn’t have any evidence other than the public media stories over the last several weeks.

      – My guess is that it’s likely that the reported stories have some basis in truth, and FISA approved wiretapping the Russian bank that sources briefly suspected was engaged in secret email traffic with Trump’s business operations. (IIRC, the traffic spike was that a marketing bot was sending the bank electronic spam for a Trump hotel.) If so, nobody was listening in on Trump phone calls, and I assume once they figured out it was business spam and not espionage, they dropped it.

      • nimim.k.m. says:

        I think there’s a spy thriller novel in about a scenario where those spam emails were Russian state-of-the-art steganographed communication.

    • Controls Freak says:

      On the law:

      FISA requests targeting a Russian bank are almost certainly acceptable, especially if there was no political input. It is possible that someone at a lower level was doing reverse-targeting (which is illegal), but this is not likely. There is likely sufficient foreign intelligence justification to overcome some knowledge that it could capture data related to the Trump campaign.

      So far, pretty much everyone has released what look like careful crafted statements which leave open a bunch of different possibilities. It’s very easy for them to all be technically true with nothing untoward having happened. It’s very easy for them all to be technically true with something untoward having happened. I don’t think we know yet.

      On the politics:

      I don’t like Hyperoperation Tiddlywinks, but man, somehow these things work out for him. There was a rush of people absolving themselves of guilt, and that can lead to interesting statements. For example, the BBC reports:

      In his interview, Mr Clapper also said that no evidence had been found of collusion between the Trump team and the Russian government.

      I get the desire to keep investigation information under wraps (and am not a fan of, for example, the Flynn leaks (though I’m not sad to see him go)), but it is somewhat shameful that these things come out as soon as someone’s self-preservation is at stake.

      On the state of SIGINT:

      I love that people are kinda almost learning about how SIGINT authorities work. The Intercept of all places had an article this weekend where they acknowledged that incidental collection is a thing! They even acknowledged that it was inevitable! They even acknowledged that collection could be targeted at a legitimate target and incidentally collect on a US Person! (They cited the Flynn case approvingly!) I know it was only a brief moment of clarity within what was otherwise an propaganda-filled anti-SIGINT piece (leading up to the 702 reauth battle), but man, it’s weird that it takes a strange Trump tweet to get even that.

      • The Nybbler says:

        I read somewhere (may have even been The Intercept) that since the PATRIOT act, reverse targeting is now legal as long as there’s a substantial foreign intelligence justification (i.e. the US person may be the primary target).

        • Controls Freak says:

          50 U.S. Code § 1881a includes:

          (b) Limitations An acquisition authorized under subsection (a)—

          (2) may not intentionally target a person reasonably believed to be located outside the United States if the purpose of such acquisition is to target a particular, known person reasonably believed to be in the United States;

          There are more complicating factors when it comes to querying the database, but “targets” for collection must be foreign and “reverse targeting” is illegal.

      • Deiseach says:

        I don’t like Hyperoperation Tiddlywinks, but man, somehow these things work out for him.

        That’s what I mean by “cursed with luck”. We’ve had the hyperventilating over “the Russians hacked the election” and all the “Trump is a Russian stooge”, as well as gloating over various members of the administration having to walk away because of contacts with Russians.

        So suppose it turns out that Trump (or business interests associated with him) were considered for investigation/subjects of preliminary investigation and then it turned out there was nothing there? Certainly nothing of “the geese depart at midnight” style collusion with a foreign power and treason?

        At the very least, it may force all the intelligence agency leaks about “we have reason to believe there was contact with the Russians” to explain where they got this evidence leading them to believe this*; if they say nothing, people will assume it was from wiretapping which means that under the former administration intelligence agencies were interfering in the election – y’know, the kind of thing Comey was excoriated for allegedly doing for the Republicans?

        It’s a “damned if you do, damned if you don’t” deal – if they say “no, we never wiretapped Trump!” then the natural question is “why not, if you had all these suspicions about him?” and if they say “yeah we did, kind of, accidentally, but there was nothing to find”, then all this “Russian spy plant stooge traitor” stuff has had the legs cut from under it (except for the True Believers and there’s nothing to be done about them).

        And all this from a guy doing is outrage! tweeting at ungodly hours of the morning because nobody can take his phone away from him 🙂

        *I’m ignoring “British ex-spy told them” because surely they checked it out for themselves and if not, they should have read up on Spycatcher in which a British ex-spook made all kinds of juicy revelations which have been Offically Denied 🙂

        Actually, I’m finding the current “intelligence services versus Trump” brouhaha really reminiscent of Harold Wilson versus MI5:

        Peter Wright claimed that he was confronted by two of his MI5 colleagues and that they said to him: “Wilson’s a bloody menace and it’s about time the public knew the truth”, and “We’ll have him out, this time we’ll have him out”. Wright alleged that there was a plan to leak damaging information about Wilson and that this had been approved by ‘up to thirty officers’. As the 1974 election approached, the plan went, MI5 would leak selective details of the intelligence about Labour leaders, especially Wilson, to ‘sympathetic’ journalists. According to Wright MI5 would use their contacts in the press and the trade unions to spread around the idea that Wilson was considered a security risk. The matter was to be raised in Parliament for ‘maximum effect’. However Wright declined to let them see the files on Wilson and the plan was never carried out but Wright does claim it was a ‘carbon copy’ of the Zinoviev Letter which had helped destabilise the first Labour Government in 1924.

        • J Mann says:

          The way I’ve heard that stated is “sometimes it’s better to be lucky than smart.”

          The most ridiculous example, IMHO, is that after Jeff Sessions omitted two meetings with the Russian ambassador from his answer to Congress, both Claire McCaskill and Nancy Pelosi volunteered that they had never meet with the Russian ambassador, only to be presented with evidence that they had. Both of them had to qualify that they had a private understanding of what constituted a relevant “meeting’, which was pretty much the exact thing that Sessions said.

          Is there any chance that Trump somehow obtained the Spear of Destiny from Rubio, possibly in a Maltese Falcon-like robbery?

          https://slatestarcodex.com/2015/11/16/hardball-questions-for-the-next-debate/

          • Deiseach says:

            The most ridiculous example, IMHO, is that after Jeff Sessions omitted two meetings with the Russian ambassador from his answer to Congress, both Claire McCaskill and Nancy Pelosi volunteered that they had never meet with the Russian ambassador, only to be presented with evidence that they had. Both of them had to qualify that they had a private understanding of what constituted a relevant “meeting’, which was pretty much the exact thing that Sessions said.

            That is less “Spear of Destiny” and more “Sword of Chang” 🙂

            I know little about McCaskill save that she’s the Missouri Democrat incumbent Senator, but I’m always happy to see Nancy “I’m A Theologian And The Bishops Aren’t” Pelosi getting egg on her face.

          • Sandy says:

            It was frankly surprising that McCaskill got up in arms up about this because she’s up for reelection in 2018, and the only reason she won in 2012 is because she was up against Todd “legitimate rape” Akin, whose comments destroyed his funding and alienated suburban Republican voters and married women. A halfway competent opponent could beat her without all that much difficulty, and I’m sure she’s at or near the top of the list of 2018’s targets for the Republicans, so you’d think she’d be doing everything she could to avoid getting into these public flaps.

          • Deiseach says:

            A halfway competent opponent could beat her without all that much difficulty

            From what little I’ve gathered courtesy of a blogger living in the great state of Missouri, this is eminently true and probably the reason she stuck her neck out here: she needs something to keep her name in the public mind as “send her back to the senate” and she jumps on any kind of possible bandwagon that might make her sound like “vote for me, I’m Democrat (so not one of those nasty mean Republicans) but not too Democrat (so not one of those crazy California ‘independence for two year olds’ types)”.

            Presumably Trump-associate-bashing sounded like the thing to attract disgruntled Republicans who aren’t fans of Trump and keep Democrat voters, but as you say, it turned out inconveniently.

          • bean says:

            I know little about McCaskill save that she’s the Missouri Democrat incumbent Senator

            Ah. The Wicked Witch of Missouri. I had the misfortune of having her as one of my senators for a while. Her victory over Akin was only after his epic foot-shooting, as has been noted elsewhere. I’m almost surprised she didn’t suffer backlash over the sheer number of times she paid to put that clip on TV.
            (He was my Congressman before he ran for the Senate, and during that time did not distinguish himself as a nutjob.)

          • Chevalier Mal Fet says:

            As a life-long resident of the great state of Missouri, McCaskill is every bit as deserving of egg on her face as Pelosi. She won election in 2006 on the anti-Bush wave, then survived 2012 by supporting Akin (a thoroughly wretched politician himself) through the three-way Republican primary, an operation in which all-too-many of the state’s Republicans cheerfully cooperated.

            I admire her, grudgingly, for that, but I am also looking forward to 2018.

        • Aftagley says:

          It’s a “damned if you do, damned if you don’t” deal – if they say “no, we never wiretapped Trump!” then the natural question is “why not, if you had all these suspicions about him?” and if they say “yeah we did, kind of, accidentally, but there was nothing to find”, then all this “Russian spy plant stooge traitor” stuff has had the legs cut from under it (except for the True Believers and there’s nothing to be done about them).

          Not trying to be one of the true believers, but I think you’re simplifying this a bit too much. The burden of proof to wiretap a US citizen in the normal course of events is very high. Suspicion and what people could reasonably claim is just coincidence is not grounds for a wiretapping. When a major political figure is involved, the standards are going to be even higher. You would pretty much need proof that you were going to find something before a judge would approve it. It’s possible people made a rational decision that this just wasn’t possible/necessary without fundamentally destroying the existence of any kind of collusion.

  15. Nancy Lebovitz says:

    https://www.thisamericanlife.org/radio-archives/episode/611/vague-and-confused

    Part One (“That’s How I Rule”) is about a micronation of sorts. Available as podcast or transcript.

  16. Nancy Lebovitz says:

    http://jolomo.net/solarsystem/1936.11.html

    Possibly the ssc sort of fun. John Campbell, writing in 1936, talks about what the surface of Mars is like. How much of the article holds up? I’m fond of the bit about taking iron compounds from black to red sucked a lot of the oxygen out of the atmosphere, but I have no idea whether it makes sense quantitatively.

    • Nornagest says:

      Hmm. Well, 1 atmosphere is about 15 psi, which works out to a partial pressure of about 4 psi of oxygen on Earth. If there’s four pounds of oxygen in the atmosphere for every square inch of land area, then you’d have to oxidize a hell of a lot of rock to get rid of it.

      But Mars’ atmosphere is way thinner; about 0.1 psi, most of which is carbon dioxide. At that level it seems plausible that oxidation could have substantial effects on it.

  17. psmith says:

    Haggling: do you do it? (especially for peer-to-peer purchases of consumer durables, cars for instance, in the first world–but feel free to post about other instances.) What’s your favored approach? Any interesting economic/game-theoretic insights?

    • Well... says:

      Yes, I haggle for stuff like cars (from both individuals and dealerships) and houses.

      I might haggle at a farmer’s market-type place if I’m feeling bold and if the price of something seems too high.

      I’ll haggle at a garage sale if the price seems too high though it usually doesn’t, and I’m usually in a charitable mood at garage sales anyway. Same goes for buying stuff on Craigslist.

      I have haggled or tried to haggle on phone bills and insurance premiums, with mixed results. Same goes for customer service calls in general.

      I don’t have any big insights, I just try to remember some basic negotiating tactics, anchoring, etc. The biggest thing for most Americans, I imagine, is to not feel intimidated: it’s just a transaction, not a fight.

    • J Mann says:

      I don’t like haggling in person – it triggers my conflict avoidance. I typically won’t buy in an environment where haggling is required to get the non-sucker’s price. Generally, I prefer to compare prices on the internet and buy from the lowest price reputable seller.

      Exceptions:

      – On cars, I check what people are paying on the forums (mostly Edmunds), then email as many dealerships as I can find and offer a price, then negotiate slowly by e-mail, refusing all invitations to talk by phone or in person.

      – On home buying, I discuss with my wife and broker, and offer something less than asking price to test the waters, then try to negotiate on the details.

      – If I’m at a place that offers free or discount credit (like a furniture store), I ask if the manager will approve a discount if I pay cash.

      – I would like to haggle for cable, phone, and internet, but I don’t.

    • Corey says:

      Utility-style services with high fixed costs are a fruitful place for haggling. SiriusXM is the Platonic ideal: costs them billions of dollars to service the first subscriber and approximately $0 for each additional one. So they have to price-discriminate to make any profit.

      My haggle-fu is pretty weak so I just request to cancel (or actually cancel) then get offered a reasonable price.

    • andrewflicker says:

      Haggling- know what BATNA means, and how to use it in a negotiation. (Google it if you don’t know the term.)

      I haggle on things like cars and small-business services, but I also tip generously and will agree to overpay for certain services as part of a long-term “I cooperate” prisoner’s-dillema-style strategy if I think it likely that a high-trust relationship will pay dividends. (ie, it’s worth overpaying for a great mechanic, if you know he won’t ever cheat you)

      I will occasionally haggle on big-business services, like internet bills, but have largely stopped as the return-on-time-investment has declined. (I don’t have TV, or that one would probably still be worth arguing about)

      I will haggle at the farmer’s market if I’m buying a large quantity or I think they’re wildly overpriced- but I’ll also walk away at the drop of a hat and conspicuously shop at the next stall over. *shrugs*

      Car story, since I like telling it: My last car purchase was used private party, so just person-to-person. I had a budget for around 10k, but valued paying less than that and using the rest to pay down debts if I could. Found a guy that sold the car I wanted (2001 BMW Z3), for something like 7200, but would probably need minor repairs. I negotiated him down to 6800 over the phone, then agreed to pay for an inspection. Inspection returned significant repairs- 4500 worth if I had everything fixed. Negotiated with the mechanic down to 3500 for everything, then told the seller that I wanted 5800 since it needed so many more repairs than he had led me to believe. He resisted, we went back-and-forth, he walked out. I was disappointed, but didn’t budge, because all I was out was an inspection fee, and I could go back on the market for a better deal.

      That night, he called me around 10pm (I was already in bed, reading). He asks tentatively “how much did you offer, again?” I tell him 5800 or nothing, as before, and he says “Well, 5800 is really just too low…” and half-interrupting him I hear a woman yelling from the background “YOU HAD A GUY OFFER 5800 IN CASH AND THAT DAMN THING IS STILL IN MY DRIVEWAY?!” and I start laughing. He comes back on sheepish and says “meet you at the bank tomorrow?”

      Paid him the 5800, drove it a mile to the shop, paid the mechanic 2k for the important fixes, did another 1k of fixes myself at home, and left the rest as “eh, stuff I don’t care about”. Still happily driving it two years later!

    • John Schilling says:

      I do not like haggling for price, but I’ll do it to some extent for houses, cars, airplanes, and salaries because I know it is expected and priced into everyone’s initial offer. If the initial offer is more than 10% off fair market price, I will almost certainly go someplace else.

      I prefer “haggling” over what I am going to get for the price. The airplane comes with a fresh annual inspection from a mechanic I trust; the job comes with a private office, etc. Nothing outrageous, but everything that might plausibly be part of a fair deal between people who are going to keep doing business in the future and is likely to be left out by anyone seeking short-term advantage once they’ve gotten to “yes”.

      If someone insists on an outrageous price, I might try attaching outrageous demands to balance it, particularly if I know they are seriously cost-constrained.

    • knownastron says:

      I think the most important thing about haggling is to not be outcome dependent.

      I’ve noticed in myself that the apprehension to haggling is that the other person gets might get offended and walk away or I “lose” the negotiation and pay original price. But the more I do it, the more I realize a) I’m likely never going to see this person again and b) paying the original price after 5 minutes of negotiating isn’t the worse thing in the world.

      I don’t get too intricate with my haggling. Usually I offer a price that is almost unreasonable and then when they counter offer, keep “meeting in between” until they accept your price or you walk away. I don’t try to persuade or reason with the person. My mom is a master haggler and basically uses the method I outlined. My mom is also very willing to ask for discounts. I believe haggling’s more about willingness to engage and walk away, than it is about psychological tactics or deep understanding of game theory (although that might that help at the margins or high level negotiations).

    • rahien.din says:

      1. Always be willing to walk away. This can’t be emphasized enough. The customer’s only source of power is their physical presence.

      2. Learn as much as you can about the other party before engaging, and use that information to your advantage. For instance, if they are paying taxes on and making repairs to a vacant house a state away they inherited from their parents, their desire to get out from under that burden may make them willing to negotiate.

      3. Engage the other party when they are at a psychological or emotional disadvantage. The classic example is shopping for a car mid-week on a rainy day.

      4. Figure out what would be a good price for the thing you want. My wife and I almost missed out on a great deal on a car because we didn’t realize how good the deal was and tried to play hardball. Luckily we realized our error in time to get the car.

      5. Decide what you want to pay, and always ask for better than what you want. Make an initial offer that is plausible but slightly painful to the seller. If they accept right off, your offer was too high.

    • Schibes says:

      Haggling: do you do it?

      I do it and I HATE it. I bought a 4 year old used car last week and somehow managed to haggle 5.5% off the purchase price but wound up giving it all back and then some through my purchase of an extended warranty for the car since its factory warranty expires next month. I’m currently rationalizing it as my having successfully haggled 3 years of additional warranty coverage from the dealer for “free”.

      One of the reasons I hate haggling is that my father, as a recently retired small business owner who successfully kept his doors open for 40 years, is an expert at haggling over just about anything and everything, and never wastes a single opportunity to tell me how stupid I am for “letting people rip me off left and right all the time”. (Aside from this and a variety of other petty disagreements, we actually get along pretty well most of the time.) So definitely some personal issues going on here behind my hatred of haggling. Even times when I’m “successful” at it I still don’t tell anyone because there’s this voice in my head saying Dad would have held out for an even better deal.

  18. J Mann says:

    Is anyone aware of a fairly neutral attempt to determine what proportion of reported hate crimes over the last N years are real, fake, indeterminable, or apparently crimes of opportunity?

    • Protagoras says:

      I feel like what you’ve describe here would be incomplete without including crimes that are actually hate crimes but not reported as such (a difficult category to identify, to be sure, but the same is true of your categories).

      • J Mann says:

        True, but I’ll take what I can get.

        I’m mostly trying to get some kind of Bayesian foundation for what to make of news stories to the effect of “Swastika found painted on local college doorway” or “Woman writes facebook post about harassment on subway” – you can find people who assume that they’re mostly hoaxes, and people who believe that they are mostly written by Nazis or white supremacists, but I don’t know what to make of it.

        • suntzuanime says:

          I think it depends pretty heavily on the specifics of the incident, and trying to get an overall probability for “reported hate crimes” is a silly idea because there’s so much variation in reported hate crimes and the hoax rates for different types. For example, nobody is going to shoot up a mosque as a hoax, pace Alex Jones, but writing a swastika on a thing is pretty well optimized for causing freakouts without doing any actual damage, meaning it’s a good candidate for a hoax.

          Swastikas are an especially good marker of a hoax in my opinion, because they’re such a salient marker of hateful ideology in people’s minds, but so few people actually associate themselves with that particular brand.

          • dndnrsn says:

            And, what makes something a “hoax”? The guy trying to frame the woman who dumped him for stuff – is that a hoax in the same way that Rachel Dolezal claiming she was the target of racial intimidation that all the evidence indicates she fabricated is?

          • J Mann says:

            @dndnrsn:

            I’ll leave that question to the researchers with their fancy degrees.

            Ultimately, what I think is relevant is whether hate crimes are increasing and who is committing them.

            So if the story is “swastika found on college prayer room door” or “woman posts something on facebook saying ‘someone yelled at me ‘go home – Trump rules’ “, should I update my priors and by what?

            – You could take the radical skeptic view and say “there’s no way to tell what this means until someone is arrested,” which is probably right.

            – Roughly speaking, my personal taxonomy is:

            a) The perpetrator is identified as someone who is likely to have been motivated by the racial animus evident in the original story = probably true. (The swastika or subway harassment were performed by a nazi, or at least by a person of different race from the victim seeking to give racial offense).

            b) The perpetrator is identified as someone who is not likely to have been motivated by the racial animus evident in the initial story. (The swastika was drawn by one of the members of the victim group to raise awareness; videos show that the subway incident didn’t happen; the mosque was shot up by a member who was upset by local issues, etc.)

          • The Nybbler says:

            There’s a number of different types of actions here

            1) The event never happened, and the report was entirely false.

            1a) An event happened but was substantially different than claimed by the target, enough to not constitute a “hate crime”.

            2) The event happened but was caused by or with the connivance of the “victim”.

            3) The event happened but was caused by someone with the opposite ideology of the apparent perpetrator, in order to stoke hate crime hysteria.

            4) The event happened but was caused by someone with the opposite ideology of the apparent perpetrator, for other reasons (e.g. to evade responsibility or to blame someone else)

            5) The event happened and was intended to harm the target but was non-ideological (e.g. simple vandalism that used swastikas or whatever merely to shock, an attack on a minority member for “ordinary” reasons such as a dispute over money)

            6) The event was misreported or misunderstood.

            7) The event happened and was aimed at the target for the reason of prejudice against a group the target was a member of.

            “Hoax”, I think, only applies to 1,1a and 2. I’d call 2-4 “false flag”. Swastikas I expect are usually 5, sometimes 6 (“Trump” with a swastika replacing the T reported as anti-Semitic when intended as anti-Trump, for instance).

  19. Tibor says:

    (continuing the language thread from the previous OT):

    @Creutzer: Interesting. I figured that the similarity in construction in Czech and German is due to the German influence on Czech. The first language of the people who were trying to reconstruct* Czech was German so they tried to construct the new words the same way one would do it in German. But if those similarities are with other Slavic languages as well, then there is something else going on…perhaps the proto-slavic and proto-germanic languages were closer than modern Slavic and Germanic languages? Btw, this is an interesting graph of lexical distance among languages of Europe. I knew that the Finno-Ugric languages were very different (they’re not even Indo-European) but I am surprised how far the Celtic languages are from everything else.

    It also surprises me that Czech is further away from Russian than Croatian and Serbian are. Simply based on geography I’d expect Czech and Russian to be closer than Croatian and Russian. The central position of Slovak among Slavic languages is interesting. I am always surprised how easily Slovaks understand pretty much all Slavic languages. Czech is mutually intelligible with Slovak, but while Slovaks understand Polish with ease, for me it is very difficult. I can read Polish about as well as Dutch (because of my knowledge of German) and Russian about as well as Swedish (I recognize a few words but I don’t follow the meaning).

    In Germanic languages it is interesting how close the northern Germanic languages are and how English and German+Dutch are sort of outliers (similarly to Romanian in Romance languages). There seems to be no central language in Germanic languages unlike in Romance (Italian) and Slavic (Slovak) languages.

    Also, how the hell is Greek related to Dutch (even if only a little) and Irish to Portuguese??

    As for auxiliary verbs – I find it easier to use the “ir+infinitive” construction in Spanish and Portuguese (French has it too so I would guess that all Romance languages do) than the simple future tense (it is also kind of formal, but that’s beside the point). Similarly, it is definitely easy to forget preteritum and use perfect “haber+past participle” construction for past like they tend to do in Spain than to use the rather irregular preteritum. Even in German it is easier to use Perfekt instead of Präteritum (I heard that Swiss German does not even have Präteritum) as is now common practice anyway (save for modal verbs). Both are fairly regular in German but Perfekt more so and the irregularities are kind of regular, whereas in Präteritum they aren’t. Btw, what always puzzled me is that while German is fairly regular, it is a horrible mess when it comes to assigning genders to nouns. Save for a few word endings there is very little system in it. Czech also has three genders but each noun belongs to a category which can be easily recognized by its declination, so even for adopted foreign words the gender is immediately clear. In German, even natives are sometimes not sure – for example some people say “die Nutella” although based on what is written on the packaging the company considers it to be “das Nutella” (which sounds strange to me, I’d also say die).

    *reconstruct is a bit exaggerated, although Czech (was used basically only in the countryside and by the working class in cities and it was full of German words (much more than it is today).

    • Creutzer says:

      I figured that the similarity in construction in Czech and German is due to the German influence on Czech.

      You’re definitely right that loan translations are a thing, especially in legal, administrative, and scientific vocabulary (also, idioms). But beyond that, Slavic languages and German (probably Germanic, I just know very little about the other contemporary Germanic languages) have (kept) the same old Indo-European word formation pattern of prefix-root-suffix. They still have a wide range of prefixes (although they don’t tend to map 1-to-1) and use a lot of metaphor in the prefix-root combinations. Sometimes they hit similar metaphors. In addition, the whole pattern results in many instances of word A being related to word B within, say Russian, in the same way as their translations A’ and B’ are related within German. (Sorry for being lazy and not digging up an example. I just notice them once in a while when I hear them and then forget.)

      Czech is mutually intelligible with Slovak, but while Slovaks understand Polish with ease, for me it is very difficult.

      There is huge individual variation in how good people are at understanding related languages. One ought to be careful with making rankings of closely related languages with respect to comprehensibility.

      In Germanic languages it is interesting how close the northern Germanic languages are and how English and German+Dutch are sort of outliers

      But that aligns perfectly with the historic grouping: North Germanic vs West Germanic.

      Also, how the hell is Greek related to Dutch (even if only a little) and Irish to Portuguese??

      The graph is not complete and I don’t know how the author chose which links to draw. As far as I can see, the thinnest dotted lines ought to be the same as no lines at all, so it doesn’t actually say that Irish is closer to Portuguese than Welsh is.

      In general, I’m wary of reading much into this graph, since it’s only about the lexicon, not about phonology and phonetics, and I don’t even know what measure of lexical distance it uses. The source is, obviously, inaccessible.

      heard that Swiss German does not even have Präteritum

      Probably. Neither do Austrian dialects.

      Btw, what always puzzled me is that while German is fairly regular, it is a horrible mess when it comes to assigning genders to nouns.

      Albanian is the same in this respect. It’s just what happens when you reduce final syllables but insist on keeping genders. French sort of got out of this problem to some extent on account of just the very particular way in which they reduced final syllables.

      • Tibor says:

        You’re right I should have viewed that graph more critically, given that I just found it at a random website online.

        Isn’t English considered a Northern German language though? Of course, English has been heavily altered over time by all the Romance words.

        So did words of a particular gender always end with a given syllable/set of syllables in middle German or before? So instead of “die Milch”, would it be something like “die Milche”?

        It is a bit more complicated in Czech, each gender has a couple of pattern words that all other words conjugate like. A word that ends with a consonant might be masculine or feminine, but if you see it declined (i.e. not in nominative), you will know which word it is declined as and hence its gender. There are 4 of these pattern words for each gender and extra 2 for “unliving” (translating it more or less literally from the Czech term) masculine words (those two are forest and machine). It is a more complicated system than that of Spanish, where the suffix almost always enough to tell the gender (and they only have two genders of course), but at least there is a system to it. In German, it is often guesswork and to make matters worse, the genders are often the same as in Czech, but not always. I kept saying “das Milch” and “die Wasser” for quite a long time and I still find it mildly cringeworthy that such “obviously” feminine words as Whiskey, Brandy or Vodka are masculine in German, since pretty much all alcohol other than beer is (but at least there is some pattern to that 🙂 ).

        • Creutzer says:

          This whole distance measuring business is, of course, a thing in historical and comparative linguistics. This thread here contains a bunch of references which I have not read but which look serious and potentially interesting: http://linguistics.stackexchange.com/questions/1206/do-linguists-measure-the-relation-distance-between-languages-how

          What I couldn’t find in a quick search is serious mutual comprehensibility research that maybe even links up to these computational distance measures.

          Isn’t English considered a Northern German language though? Of course, English has been heavily altered over time by all the Romance words.

          No, English is West Germanic, together with German, Dutch, and everything in-between. North Germanic is everything Scandinavian, East Germanic is Gothic (and hence dead).

          So did words of a particular gender always end with a given syllable/set of syllables in middle German or before? So instead of “die Milch”, would it be something like “die Milche”?

          It’s not true of the particular example “Milch”, but in general, you could infer the declension class and gender of an Old High German word with much greater reliability, even if not always with absolute certainty. Roughly like with Czech.

          How on earth do you decline Whiskey and Brandy as feminine nouns in Czech?!

          • Tibor says:

            Come to think of it, Whiskey and Brandy are not declined at all (they’re the same in all cases. I never realized such words exist in Czech. I suppose it is only the case of some foreign words.

            What’s even stranger, the only such examples I can think of are feminine. Gin is masculine in Czech and it is declined like “hrad” (which means castle) – one of the two “unliving” masculine pattern words (btw, now I realized that while gin is pronounced exactly the same way as džin – which means genie, like the one from Aladdin – gin is declined differently, since gin is unliving and džin is living. And actually, even words that end with -ey will be declined if they’re masculine. The name Bradley is (obviously) masculine and will be declined, although it will be declined differently than either of the four pattern words. I guess that foreign words might still be a bit messy in Czech. The more I think about it, the more I like Spanish where genders are almost always crystal clear 🙂

            I’ll check out that linguistics forum later, thanks!

            EDIT: Thinking about it a bit more, I think you can also decline whiskey similarly to how you would decline žena (woman), although not exactly the same way. I think I’ve heard people decline it that way but I think it is not grammatically correct. With brandy you can’t even do that.

          • Creutzer says:

            Ah, okay. Russian has the same default rule, except it makes indeclensible foreign borrowings neuter. I guess all Slavic languages must have something like that, otherwise what are you going to do if you want to borrow a word that doesn’t fit into any of your declension classes?

            If you actually cared about anything at all, of course, you would have borrowed it as whiský, genitive whiského.

          • Tibor says:

            Whiský would be an adjective though and pronounced with a longer y at the end (like “Whiskeey”). Surnames are sometimes adjectives turned into nouns (e.g. pan Červený / paní Červená – Mr. / Mrs. Red) and then it would be conjugated like that (you could also have Whiská and decline it like the female adjective surname), but it only really works for surnames. Although Bradley is conjugated like that, except that you don’t change the root (it is Bradley and not Bradlý), so I guess it would work like this:
            Whiskey (nominative)
            Whiskeyho (genitive)
            Whiskeymu (dative)
            Whiskeyho (accusative)
            Whiskey (vocative)
            Whiskeym (locative)
            Whiskeym (instrumental)

            This way it would be masculine, like in German, which is what I found strange in the first place (but of course it is just a matter of being used to something). You will still have problem with say Ashley, since that is a female name and that you really cannot decline like a masculine word, so Ashley is not declined, just like whiskey (although unlike whiskey there is a good reason for Ashley to be feminine).

          • Tibor says:

            Actually, you can turn most adjectives to nouns, like in English or German, although some might sound a bit strange. I also realized that the declination is not a special rule for adjectives turned to nouns, it is just how adjectives are declined (I did not realize that adjectives are declined, although it should have been clear, in German they’re declined as well).

            It is really strange to think about the grammar of your native tongue. I mean, sure you learn this at school, but you somehow treat the grammar as something annoying you have to learn at school and you don’t need since you know how to decline in your own language anyway. But ones you get into the details it becomes more complicated and less regular than you think it is.

  20. Tekhno says:

    Lasers! Pew! Some math free (I’m lazy) sci-fi weapon speculation here.

    Blast a hole through my body and the hole needs to be large enough to cause sufficient organ damage and blood loss. A large volume of removed material requires a large amount of energy. Let’s call this a type A attack.

    Make a very very tiny hole through my body and draw that path along, and I will have been bisected. This only requires a small amount of energy. If you remove a cylinder of flesh from my neck, my head falls off regardless of whether the cylinder is one atom tall or billions. Let’s call this type B.

    Bullets are type A. Is it possible to create a laser that can achieve type B? I’m assuming that one barrier is that you are trading off very low energy for very high power, and plasma breakdown of air would be induced impeding the laser (incidentally, this has been done on purpose to create “laser guided lightning”). The other might be the impossibility of making a dynamic lensing system that can actively focus a laser to the precision needed against a moving target, or even focusing a laser like that at all, full stop.

    One of these barriers goes away in space, so could military satellites be slicing other satellites in two with nanoscopic low energy pulses in the future?

    • Skivverus says:

      Maybe, but I imagine it would be cheaper still to aim for type C: “make a tiny hole into – but not all the way through – relatively vulnerable important spot X”. Energy that makes it out the other side of your target is energy not being used to kill your target, after all.
      Of course, all that energy savings might well be spent on figuring out X, but you do kinda have to figure that out anyway to avoid killing the wrong things, albeit maybe not to the same precision.

    • beleester says:

      Lasers aren’t disintegration rays, they just heat up the target. And heat tends to spread. So I don’t think it’s possible to vaporize a particular atom without spilling a bunch of extra heat into the surrounding atoms. Or in other words, if you try to vaporize a microscopic hole through your target, a lot of energy will probably go into searing the flesh and boiling fluids around that hole, with a result sorta like a bullet hole.

      There’s also the issue that type A won’t necessarily take more energy than type B, because how you make that hole matters. A laser makes a hole by burning and vaporizing the material, which takes a buttload of energy. A bullet, on the other hand, simply needs to pierce the target and push material out of the way as it passes, which is a bit easier. According to Wikipedia, a .22 pistol has a mere 160 J of muzzle energy, but it’s still perfectly lethal if your aim is good.

      (If your aim is good enough to cut someone’s head off with your laser cutter while keeping an atom-wide focus, it’s probably also good enough to put a .22 into their eye.)

      • The Nybbler says:

        I think ultrashort pulse lasers can cut materials non-thermally. But I don’t think they can be focused from a distance to cut a line.

    • John Schilling says:

      Lasers aren’t disintegration rays, they just heat up the target. And heat tends to spread. So I don’t think it’s possible to vaporize a particular atom without spilling a bunch of extra heat into the surrounding atoms.

      Not only that, but “vaporize” does not mean “disintegrate”, and lasers don’t remain perfectly collimated as they travel through dense, hot, turbulent gas with entrained particulates. You can probably drill deep-ish holes with a series of pulses that give time for the vapor and particles to clear away but not for the surrounding solid to collapse into the hole, if you can coordinate a hundred to a thousand tightly-focused laser pulses to hit the same spot in rapid succession, but it’s going to take at least as much energy as a bullet for the same level of damage.

      Some analysis, with probably more math than you want, here.

      • Tekhno says:

        Yeah, I read that page a few years ago. My idea is that you could use less energy by making vastly smaller crater trains and just lining them up so they cut through a target. That you probably can’t collimate it is probably the real issue.

    • Urstoff says:

      Those lasers already exist, as seen in the documentary Congo

      • Nornagest says:

        One of my college professors had a gag he’s use during safety demonstrations where he’d walk in front of the lab’s obscenely powerful CO2 laser wearing a white shirt with a black tie. The shirt was left intact (because it reflected more energy); the tie was neatly severed.

        It left an impression.

  21. CatCube says:

    Anybody else playing Breath of the Wild? I made the mistake of picking it up on Saturday, when I know I have to study for the PE in a month and a half.

    On Sunday after I got up, I figured I’d play for about an hour before I made breakfast. Next time I paid attention to the clock, it was 4:20, and I decided that having at least one meal that day was a good idea. It’s been a long time since I did that with a game.

    • Cypren says:

      I’ve been ignoring it the last couple of days and playing Horizon: Zero Dawn instead. But I took my Switch with me to the office today, anticipating I’d have time to kill at some point during the day.

      Whoa, mistake. I think the Switch is staying home tomorrow.

    • Eltargrim says:

      I was planning on accomplishing things this weekend.

      That didn’t happen.

      Instead I climbed all the mountains!

      • CatCube says:

        How’d you manage that?! It took me an hour just to get to the map tower for the third area!

        • Eltargrim says:

          a) I exaggerate. I only just unlocked my last tower just now.

          b) I’ve done basically nothing else for the last few days. Facing a seemingly intractable problem at work has not left me with a desire to take it home with me.

    • Loquat says:

      I’m enjoying the bomb rune way too much. It’s great for fishing, mining, hunting game, opening boxes, killing groups of low-level enemies… this rune does it all!

      • shakeddown says:

        I haven’t got a chance to play yet, but now I’m imagining the bomb rune lets you do dynamite fishing. (If this is wrong, you don’t have to correct me. I’m happy living in the matrix.)

        • Loquat says:

          Oh no, that’s exactly what it does. Normally the fish dart away from you and you have to move fast to catch them, but if you toss in a bomb they float to the surface and can be easily collected.

        • Schibes says:

          I’m happy living in the matrix.

          You can unplug from the Matrix for Breath of the Wild – while it’s certainly no Dwarf Fortress in terms of being a vitrual petri dish that one uses to culture emergent gameplay scenarios, it is a gigantic step in this direction, especially for console gaming. Besides dynamite fishing I highly recommend an evening of running around lighting everything on fire. You will die (a lot) but it may feel a bit cathartic to purge the corrupted, monster-infested world of Hyrule with cleansing flame.

    • Jordan D. says:

      It’s a ridiculously addictive game. They really figured out how to incentivize exploration in a way that makes it kind of irrelevant how you explore. I’ve played an irresponsible amount over the last few days, but there’s still a frankly terrifying amount of map unexplored.

      So far, the best interactions have come from chatting with NPCs staying in the stable waystations who off-handedly mention strange things to explore. For example, without spoiling anything, I’d just finished selling to one of the wandering merchants when he gave me a weird arrow, and told me it came from a laboratory in the far north. After I discovered the arrow could solve the problem I was having with Guardians, I started trekking north. After a few false starts (and a major discovery which became important later), I located the place and, after solving a problem for them, opened up a stock of ancient items which were very expensive but would allow me to climb two more map towers.

      Basically, everything flows into everything, your capabilities increasing so gradually that you never really notice them, until suddenly you ARE the ancient knight of legend.

      • CatCube says:

        Right now, the biggest thing that’s irking me is the breakable weapons. When you find something awesome, it basically sucks up an inventory slot until you come across an enemy that’s worth fighting with it.

        Even if you could *repair* them, that’d be cool. Nothing like getting ambushed by a Chu and breaking your good sword that was on its last legs.

        • beleester says:

          The weapons that you get as a reward for beating the Divine Beasts are repairable (for a high price). And the Master Sword will not break, simply run out of power and recharge after a while.

          There’s also a Hylian Shield that won’t break, but it’s hidden in Hyrule Castle so you won’t get it until the endgame.

  22. Silverlock says:

    Per Ioannidis, most scientific papers are probably wrong. (From paper at http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124 .)

    Key quote:

    Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias.

    It seems that replication studies are becoming more and more critical to the self-correcting ideal of science, but they are lowest on the list of things most scientists want to do, for several reasons: there is little funding available, it is tedious, demanding work, and there is little reputational benefit to be had, especially considering the time and effort that goes into doing it properly.

    So where do we go from here? Is it possible to get increased government funding for replication attempts? Would that even be a good idea? Is there a way to raise the cachet associated with the work? Given the large number of projects going on at any given time, is there a way to determine which attempts might be most beneficial?

    Despite frequent claims that political conservatives are a danger to science, the biggest threat these days seems to come from the scientific community itself, between the situation as described by Ioannidis and anti-fact bent that seems to be growing among some (mainly Social) scientists.

    It does not sound like it, but I am mainly sympathetic towards the folks out there in the trenches; it cannot be fun coming up in today’s “publish or perish” world. I am just not sure what steps need to be — or can be — taken.

    • Paul Brinkley says:

      The obvious solution to me, that a lot of scientists would probably scoff at, is to devote replication work to the findings that appear most likely to be profitable. So if some paper says a solar cell built a certain way will get around the 30% efficiency barrier, and the physics look plausible, put your replicators on that, and go into business making super-efficient cells. If another paper says a hybrid of black beans and kidney beans shows signs of cutting cancer risk by 50% in men, replicate that and sell it to ADM.

      As I often put it: a theory is True Enough if you can make money off of it.

      • Dr Dealgood says:

        A lot of biomedical science is already what’s called ‘translational research,’ where there are immediate clinical applications to a finding. Funding increasingly seems tied to that.

        It doesn’t seem to have helped.

      • CatCube says:

        From reading Derek Lowe’s In the Pipeline, I understand that the first step any drug company takes when an academic lab has identified a possible drug target is to do the research on their own to ensure that the findings replicate.

    • Dr Dealgood says:

      I am just not sure what steps need to be — or can be — taken.

      One recent suggestion for the biomedical sciences has been the idea of ‘Preclinical Trials.’ The idea is that most biomedical labs would conduct exploratory research, with no explicit statistical testing required. But before publication, any positive results would have to be confirmed by a specialized lab using a standardized high-powered study design in an appropriate model organism.

      Some of the ideas there are very good. Insisting on study designs with experimental rigor, and favoring papers with a single high-powered experiment over those with numerous under-powered experiments, would greatly improve the literature. And while confirmatory research by specialized ‘replication core facilities’ might not be practical, it would absolutely help.

      But overall, if implemented it seems more likely to kill biomedical science rather than cure it. It would be essentially impossible to get a PhD outside of bioinformatics, where you could still do publishable research cheaply and quickly enough, and only the (again, mostly computational) large research consortia would be able to generate data. If you like Omics research then I’m sure that sounds great, but there still are a lot of open biological questions which call for more specialized approaches. Pretending that we’re physics will only hurt the field.

    • Aapje says:

      One obvious solution is to change the stupid ‘impact factor’ rankings of scientists’ work, so people get rewarded with grants, jobs and status for replicating, not for conning the system well.

      • Dr Dealgood says:

        Impact factors are an issue, but replacing them with some statistic of replication is just running on a Goodhart’s Law treadmill. Whatever measure you pick will become just as useless as impact factors are now.

        Personally, a blinded approach to funding where the researchers and their affiliations aren’t considered sounds a lot more appealing. It would necessarily create more churn, as you can’t coast on reputation, and also put the focus on the quality of the experiment rather than of the experimenter.

        • Aapje says:

          replacing them with some statistic of replication is just running on a Goodhart’s Law treadmill.

          Yeah, I realized that as I wrote it. Still, it seems better to incentivize replication and risk that people do it badly, than the current situation where this core part of science is disincentivized.

          I’m also a bit unclear how your blinded approach would solve the issue that journals have strong bias towards original work and ‘positive’ outcomes.

          • Dr Dealgood says:

            I’m also a bit unclear how your blinded approach would solve the issue that journals have strong bias towards original work and ‘positive’ outcomes.

            It wouldn’t. High-impact journals would probably continue to chase the most exciting, least likely, results.

            What it might help with is making those high-impact journals less relevant. If your funding isn’t tied to how often you publish in Nature, there is less reason for you to tailor your research to fit Nature’s tastes.

      • nimim.k.m. says:

        I like this idea! Introduce the Blade Runner factor that measures how many replicants high-impact studies not corresponding to reality they have retired with replications.

    • Dr Dealgood says:

      Why does my comment keep getting eaten or kicked out of the reply tree?!?

      I am just not sure what steps need to be — or can be — taken.

      One recent suggestion for the biomedical sciences has been the idea of ‘Preclinical Trials.’ The idea is that most biomedical labs would conduct exploratory research, with no explicit statistical testing required. But before publication, any positive results would have to be confirmed by a specialized lab using a standardized high-powered study design in an appropriate model organism.

      Some of the ideas there are very good. Insisting on study designs with experimental rigor, and favoring papers with a single high-powered experiment over those with numerous under-powered experiments, would greatly improve the literature. And while confirmatory research by specialized ‘replication core facilities’ might not be practical, it would absolutely help.

      But overall, if implemented it seems more likely to kill biomedical science rather than cure it. It would be essentially impossible to get a PhD outside of bioinformatics, where you could still do publishable research cheaply and quickly enough, and only the (again, mostly computational) large research consortia would be able to generate data. If you like Omics research then I’m sure that sounds great, but there still are a lot of open biological questions which call for more specialized approaches. Pretending that we’re physics will only hurt the field.

    • Urstoff says:

      Open data seems like a good first step. It’s not a replication, but at least you can catch statistical tomfoolery (or simple mistakes). Changing the culture of professional science seems much harder, but a first step would be to embrace the Popperian perspective that falsifying a hypothesis is the most important thing you can do in science and is the only way to progress knowledge (Popper may not actually be right, but the sentiment is good).

      • Nancy Lebovitz says:

        I gather that open data is a good start, but you also want the computer programs that were used to work on the data to at least be known. I’m not sure whether open source programs are necessary, though it wouldn’t surprise me if that would help.

        • Cypren says:

          Open source is a requirement in order to accurately assess any computer-processed data. Otherwise, the computer can do whatever manipulation the programmer wants behind the scenes, including swapping the entire data set for a replacement hard-coded into its source.

          Data without the transformation algorithm is worthless.

        • Iain says:

          This depends pretty heavily on the area of research. Given the raw data, there are a lot of papers where the subsequent analysis is straightforward. Given the results claimed in the paper, it will often be easy to double-check the math.

          In theory it is possible to release a closed-source program that deliberately falsifies data analysis. In practice, it seems like it would be career suicide. Consider, for example, what happens if you hard-code the results you want into your “analysis” program, and then somebody tries feeding different data into it. Any paper worth its salt will at least describe the intent of the analysis, and it’s not that hard to black-box debug a program to verify that it does what it was claimed to do.

          • Nancy Lebovitz says:

            I think– I hope– that more or less honest mistakes are much more common than deliberate fraud.

          • AnonYEmous says:

            i can’t stress enough my personal opinion that “publish-or-perish” factors into this and is one of the significant causal factors, representing a middle ground between ‘honest mistakes’ and ‘deliberate fraud’. Though maybe others would consider it to just be ‘deliberate fraud’.

          • Aapje says:

            Yeah, how does ‘let’s not do this because it hurts me/us and we can’t afford to do it (even though it is the right thing to do) rate on honest vs deliberate?

            Pretty much any large scale misbehavior will involve a system with bad incentives and people justifying their collaboration by changing their own morality to match the system.

    • rahien.din says:

      Reminds me of the short address our med school dean gave us on the first day of class. Paraphrasing :

      Good morning and welcome to medical school! You will work very hard over the next four years. Also, 50% of what we are going to teach you is wrong. We don’t know which things are in that 50%. Good luck!

      Then he walked out.

    • I’m wondering if there is some way for the initial researchers to post a bounty on their own heads. I’m imagining a journal which will only accept an article if the authors agree to pay a cash reward to the first person who demonstrates, in the judgement of the journal’s editors, that their article is wrong. There are obviously a lot of problems converting that idea into something workable, but it might be possible.

      It works from two directions–an incentive not to publish an article without being pretty sure it is right and an incentive to identify articles that are wrong.

      • Douglas Knight says:

        If the journal cared about truth, they could make a very small change along those lines: it could put a bounty of publication on such a rebuttal. But, in the real world, the journal’s editors are virtually never convinced by any rebuttals, so the bounty you propose would be very safe, thus worthless. The current system is exactly the opposite: a grantor will defund you if you rebut their other clients.

  23. nimim.k.m. says:

    Background: A while ago I asked for references (and got some recommendations) on how to do statistics and understand how to judge if research is statistically sound. (In particular, I was recommended Judea Pearl’s Causality and I got it from library, but I’m afraid I did not like it much, the notation and maybe the concepts felt too alien). I’ve gone through the usual statistics courses (hypothesis testing and all that) and quite much more advanced stuff on Bayesian inference and modeling, but I’m still often confused when it comes to reading about the applications in the real world.

    (In particular, I don’t think I’ve ever seen Scott posting research that uses Bayesian methodology here, so while it’s popular at my university and ML/AI crowd in general, apparently it’s not yet caught on in the fields that do research with more profound impact on society and political discourse…)

    So that’s the introduction why I’m posting this link. The meat: Today I saw this draft by Hernan and Robins (https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/) trending on HackerNews, and at the first glance it looks like exactly the book what I have been looking for. I thought some of you might find the link useful, too.

    • Douglas Knight says:

      That book (and Pearl’s) is about doing causal inference. That doesn’t seem very helpful to your question of reading existing papers.

      • nimim.k.m. says:

        My initial impression is that it’s the closest thing resembling the ideal guide. After all, understanding and judging papers for just for the sake of reviewing them isn’t my ultimate final end goal: what I’d really like to grok would be how to do it right and how to not to do it wrong, which (hopefully) would mean that I’d be able also to say something (esp. on areas where statistics and statistical inference are common or at least claims of statistical inference are common) with some confidence that I have some idea what I’m talking about (other than explaining what is a p-value and confidence interval or some simple regression modeling). Currently my knowledge of all this feels awkwardly inadequate.

        I’d assume that knowledge how to do inference properly would be helpful for recognizing pitfalls. Currently my impression is I know more or less the shape of things I don’t know, which is a great starting point for learning but nobody will pay me to tell that.

        Also, thus far I like the presentation.

        • Douglas Knight says:

          First of all, you are equivocating between statistical inference and causal inference. It is rare that you wouldn’t want to do causal inference, but it is hard and sometimes you have to settle for, say, computing correlations. But even such simple statistical claims are often false. I doubt that this book will tell you how to judge such claims in papers. Probably it will just take basic statistical inference as a prerequisite black box. But even if it tells you how to do statistical inference, that is pretty far from really communicating how it can go wrong, let alone preparing for adversarial reading.

          Second, this book has its own brand of causal inference. Virtually no papers use this method, so this book won’t help you judge the existing literature. Maybe the right thing to do is to just reject causal inference if it is neither a RCT nor invokes a respected brand name. But maybe the book will help you evaluate “natural experiments.”

          • nimim.k.m. says:

            “Probably”? You are telling me the book is useless without reading it?

            edit. After cooling down a little… to reiterate, hopefully with less snark. I feel that not being able to say much anything on some real paper in the wild is a symptom of larger, underlying hole of missing know-how. The causal inference looks very relevant background knowledge or viewpoint (if not the magic bullet: magic bullets do not exist) on how statistics are often claimed to be used (to infer causal relationships)?

          • nimim.k.m. says:

            Actually, I think this whole argument was stupid and for some reason I feel like I should apologize to someone for having it. The draft is short-ish anyway, only some hundreds of pages on stuff I’m vaguely familiar with and thus far I’m having fun, so maybe I’ll just study it and report afterwards for what purposes I then think it was useful, if at all.

  24. Machina ex Deus says:

    I know I’ll probably get banned for this, but:

    If you put yogurt in one end of a trough of warm milk, and sourdough starter in the other, which would take over?

    • Eric Rall says:

      Probably the yogurt, since it’s selected/adapted to feed on milk, while the sourdough culture is selected/adapted to feed on grain flour. Generally, yogurt cultures are primarily a mix of a couple kinds of lactobacillus (bacteria), while sourdough is a mix of lactobacillus and wild yeast. The most common kinds of lactobacillus in both places are related, but different species are adapted to different conditions and feeding patterns.

      But it depends somewhat on the type of yogurt and the warmth of the milk. Yogurt cultures can be either mesophilic (most active around 85 F) or thermophilic (most active around 110 F), with thermophilic being more common. And sourdough is almost always mesophilic. So if the milk’s only around 90 F and the yogurt is thermophilic, the temperature gives the sourdough a big advantage, possibly enough to offset the yogurt’s advantage of being in its preferred food.

      • Aapje says:

        #OnlyOnSSC

      • Rosemary7391 says:

        That’s interesting. I didn’t realise that sourdough had things other than yeast in it. I’ve currently got kefir on my countertop, which also has yeasts and various sorts of lactobacillus, and feeds on milk.

        • Eric Rall says:

          That’s where the sourness comes from: the wild yeast makes CO2 and ethanol, just like domestic baker’s yeast, and the bacteria make CO2 and lactic acid.

          There’s also a regional delicacy in Appalachia called Salt-Rising Bread, which is leavened by a culture of a fairly nasty food pathogen (C. perfringens). The heat of baking kills the bacteria and cooks off the toxins to safe levels. I haven’t tried it myself (I came across some articles about it while researching sourdough and sauerkraut cultures), but it’s supposed to have a cheeselike flavor.

          I haven’t tried kefir, but I’ve made yogurt and cream cheese using commercial yogurt as a seed culture, and I’ve made sourdough and one batch of sauerkraut. I’m planning on trying to make a batch of fermented pickles next time cucumbers are in season.

    • shakeddown says:

      …dammit, that took me a while. Well played.

    • The Nybbler says:

      Yog-Sotrough!

  25. keranih says:

    The Adventures of Buckaroo Banzi Across The 8th Dimension is now available on amazon prime.

    (I’m still not sure what I’m going to do about this.)

    (Probably stare at the title screen for odd periods of time on and off for a couple weeks. Then order a large pizza and a sixpack of cider and watch it twice back to back, hoping it will make sense by the end.)

    • Machina ex Deus says:

      The Adventures of Buckaroo Banzai Across The 8th Dimension will never make sense: I mean, it has a woman named John, for crying out loud!

      Although I’ve heard worse campaign slogans than “Where are we going? Planet 10! When? Real Soon!

      Also: Peter Weller now holds a PhD. in Italian Renaissance Art History. (Non-honorary.)

      • rmtodd says:

        Not just a PhD, but an occasional gig teaching college-level classes at Syracuse, if I remember right.

        Several years ago I was watching this special on History Channel, “Rome: Engineering an Empire”, all about various engineering projects that the Romans undertook. They had the usual parade of interview bits with historical experts, and one of them was clearly labeled on the screen “Peter Weller, Syracuse University”. I checked and yeah, it was *that* Peter Weller. Apparently he made quite an impression on the people running the channel, because when they went ahead to making an entire 6 or 7 episode “Engineering an Empire” series, they hired Weller as the host, presumably on the grounds that he was a reasonable expert and that he obviously wouldn’t have a problem talking in front of a camera.

        And the thing is, he was an excellent host for the show, and made it a really good show. It was clear from every narration he gave and every interview he carried out with the other historians appearing as guests that he really loved history as a subject and could hardly wait to tell you all about it so you could love it too. I don’t know if they ever hired him for any other programs, but if they did I’d want to see them.