Open Thread 70.75

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

Google is saying this site has been hacked. My technical support people say it’s not actually anything bad, but this still might be a good reminder to make sure your password here isn’t the same as your password for anything more important.

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

480 Responses to Open Thread 70.75

  1. Deiseach says:

    Good afternoon! And in this week’s instalment of “Will Liverpool continue their quest to drive the stattos of FiveThirtyEight to the use of strong language and stronger liquor?”, let’s have a look, shall we?


    Win – Liverpool 73%, Burnley 10%
    Draw – 18%

    Since Burnley are a side from the bottom half of the table and we are currently riding high at No. 4, this is going exactly as you’d expect: Burnley scored in the first half, Liverpool haven’t managed to equalise, and it’s half-time.

    Which means there’s another forty-five minutes for Burnley to get a second goal 🙂

    BREAKING NEWS: And Liverpool have just made me eat my words and given hope to the livers of the FiveThirtyEight staff! A literal last-minute equaliser before the half-time whistle, so they go in 1-1 each, meaning anything (including a Liverpool victory) can happen in the second half!

    EDITED: 76 minutes into the game, which means about 14 to go, and FiveThirtyEight are entitled to laugh loudly at me. Liverpool 2 – 1 Burnley. Not impossible that Burnley might equalise, but it’s looking (touch wood) that Liverpool have this one won.

    Their prediction has been vindicated and unless something very strange happens in the next quarter of an hour the result will match what they forecast.

  2. Nancy Lebovitz says:

    Book Review: Albion’s Seed starts has

    But I didn’t know that the Puritan migration to America was basically a eugenicist’s wet dream.

    Much like eg Unitarians today, the Puritans were a religious group that drew disproportionately from the most educated and education-obsessed parts of the English populace. Literacy among immigrants to Massachusetts was twice as high as the English average, and in an age when the vast majority of Europeans were farmers most immigrants to Massachusetts were skilled craftsmen or scholars. And the Puritan “homeland” of East Anglia was a an unusually intellectual place, with strong influences from Dutch and Continental trade; historian Havelock Ellis finds that it “accounts for a much larger proportion of literary, scientific, and intellectual achievement than any other part of England.”

    Furthermore, only the best Puritans were allowed to go to Massachusetts; Fischer writes that “it may have been the only English colony that required some of its immigrants to submit letters of recommendation” and that “those who did not fit in were banished to other colonies and sent back to England”. Puritan “headhunters” went back to England to recruit “godly men” and “honest men” who “must not be of the poorer sort”.

    Are the descendents of Puritans notably intelligent?

  3. Deiseach says:

    Eurovision is coming! (Grand final will be Saturday, 13th May, note it in your calendars now).

    Australia are competing this year. Yes, Australia is now part of Europe, after their “one time only appearance in 2015, okay that went so well they can come back in 2016, wow this is going well so come on 2017!”.

    Best song so far appears to be the <a href="” rel=”nofollow“>Italian one – catchy and got a dancing gorilla, what more do you want?

    Irish entry will go nowhere – it’s a ballad. The Australian one is also a ballad, but they do it right. Our one is just meh.

  4. Tibor says:

    Metric vs. imperial system?

    I suggest that the best solution would be essentially a decimal “imperial” system.

    For a long time, I though that the Americans (and to some extent British) are just silly clinging onto a medieval system of measurement which makes little sense. However, then I looked it up a little and it turns out it is in many ways superiour to the metric system. Essentially, whenever you want to measure something and don’t care about it being exactly precise, the old measurements are just better. When you do care about being precise, you use a measure anyway, so it does not really matter. However, the system should be decimal. For doing calculations it is better than the duodecimal system. True, it is not as easy to divide by four and especially by three, but while that is useful at a medieval marketplace to figure out how much the customer has to pay, nowadays it is not such a big issue.

    Particular examples:

    If someone tells me that some places is 15 km away, I have a vague idea about the distance, but if they tell me it is 1000 paces, I have a lot better intuition. Similarly, the inch (incidentally, in Czech, the words for the inch and for the thumb are the same) is just the width of my thumb, which I handily carry around all the time. These measurements are natural, whereas a meter is the length of a distance from the equator to the pole going through Paris. Unless you are, god forbid, a Frenchman, that means nothing at all. I would fix an inch as the average thumb size and fix the foot to 10 inches (incidentally, my foot is pretty much exactly 10 of my inches, I am very decimal!) Then 10 feet would be a dace, short for double pace (one could probably come up with a more elegant name for that), literally about two paces and then you can have mildaces/miles which would be 1 000 daces. One decimal mile would then be about 3 km.

    Then you have weights. Here, I am not very happy with either system. A kilogram means nothing and a pound is equally useless (also, it is basically just half a kilo). Instead, I propose cup as the basic measure. Cups are already used in US recipes (as well as old European cookbooks which do not try to pretend that it matters whether you use 5% more or less of an ingredient) and it makes cooking faster. So for weight I would use the weight of a cup of (say distilled so that it is really precise enough for when you need to be precise) water as the basic measure and again have it decimalized. The same goes for volume – a cup of water would be the basic measure. Of course, you can still used cubic decimal inches, etc. but for cooking and some other things which mostly have to do with water and other liquids, you’d want something like this).

    Temperature is actually really well done in the metric system. 0 for when water freezes and 100 for when it boils is very useful for everyday use.

    Other units like bells or candela are hard to really imagine in a meaningful approximative way anyway, regardless of how you define them, so those could probably stay the same.

    • powerfuller says:

      I think Fahrenheit is easier for day-to-day use than Celsius for describing the weather, which I guess is how temperature is mostly frequently used. Sure, the freezing-boiling point of water is nice and tidy, but it’s not really that much more useful (for weather) than, say, the meter’s relationship to the distance from the equator to the pole (with the exception of below zero = icy roads). I’m far more interested in how temperature feels to me than how it relates to water, and with Fahrenheit, 0-100 is a pretty good gauge for the average fluctuation in weather over the course of the year for many places. I might feel differently if I lived elsewhere, but where I am, it rarely goes below 0 and it rarely goes above 100, which means I have 100 numbers to describe the weather (though that often gets simplified into low/mid/high decades). With Celsius, the larger different between degrees makes it harder to give a quick, but decent, approximation without using decimals, and the numbers 50-100 are basically wasted.

      I also like the Imperial system because, as you said, it’s mostly based on the body, and “man is the measure of all things.”

      EDIT: grammar, etc.

      • The original Mr. X says:

        I’m far more interested in how temperature feels to me than how it relates to water, and with Fahrenheit, 0-100 is a pretty good gauge for the average fluctuation in weather over the course of the year for many places. I might feel differently if I lived elsewhere, but where I am, it rarely goes below 0 and it rarely goes above 100, which means I have 100 numbers to describe the weather (though that often gets simplified into low/mid/high decades). With Celsius, the larger different between degrees makes it harder to give a quick, but decent, approximation without using decimals, and the numbers 50-100 are basically wasted.

        I think that, in practice, this isn’t really an issue. When you want to describe how the temperature feels, you pretty much never have to get it right to the exact figure; “It’s in the high twenties” or whatever is accurate enough.

        • powerfuller says:

          @ The Original Mr. X

          True, it only really matters when getting into arguments over the thermostat. “Mid 60s” versus “High 60s” seems like an easier and more useful distinction than “18” versus “20,” but that’s what I’m used to. The whole conversation boils down to, “Whatever you’re used to using is easiest,” anyway.

          Next, shall we discuss why baseball is better than soccer? 😉

          • The original Mr. X says:

            Not having a thermostat at home, I’m fortunate enough to avoid these sorts of arguments. :p

          • Tibor says:

            Hmm, the mechanical thermostat on my heater goes from 0 (well, it is really not 0, it is a picture of a snowflake) to 5 😀 It also has colours describing what is cold and what is hot. there is a blue triangle starting with 2 and going down to 3 and then a red triangle which ascends from 3 to 4. Between 1 and 2 there is a picture of the moon.

            I don’t know if you have these in the US, but they’re the most common (mechanical) thermostats in Europe. But when I describe then in words, they really sound quite ridiculous.

            I think the numbers for temperature are abstract anyway and the reason mid or high 60s mean something to you is that you are used to it. However, inches or feet are genuinely easy to relate to.

          • Matt M says:


            We use those for our amplifiers I think.

          • Tibor says:

            “Matt: Well, of course, we also have those, but the thermostats that go to 6 are really expensive.

        • Deiseach says:

          It really does come down to what you’re accustomed to; Irish weather is generally mild and doesn’t go much past the tens to twenties range, so when it’s in the 30s or higher on the Continent you know it’s really hot.

          Telling me the weather is in the “low 60s” Fahrenheit, I don’t know if that’s cold, mild, hot, or what. Until I convert it into Celsius I have no idea and I imagine for people who grew up on Fahrenheit the reverse is true.

        • JayT says:

          I really don’t see what flaws in Fahrenheit Celsius fixes. The fact that water freezes at 0 and boils at 100 (at sea level) just doesn’t really matter in day to day life. Both it and Fahrenheit are obviously very flawed metrics, but like Powerfuller said, at least with Fahrenheit you have 100 units to describe the vast majority of temperatures that you will encounter throughout the year, where Celsius gives you like half that resolution.

          • Artificirius says:

            Is that resolution meaningful at all? Given that we are not really able to meaningfully self measure the temperature, especially when confounded by other conditions internal and external.

          • hlynkacg says:

            We aren’t? I find that I am generally able to gauge temperatures to within +/- 5 degrees Fahrenheit. Is that weird?

          • JayT says:

            I think the resolution is useful, and like hlynkacg I would guess I’m usually able to measure the tempurature fairly accurately. Even if we decide that it isn’t though, I would still say that Fahrenheit is better just because going from 0-100 for most temperatures you will encounter is better than going from -18 to 38. Even if it’s just for aesthetic reasons.

            Also, even if that resolution doesn’t matter much, what do we gain with Celsius where we lose it?

          • johnjohn says:

            The fact that water freezes at 0 and boils at 100 (at sea level) just doesn’t really matter in day to day life

            uuh, it’s very very very relevant that water freezes at 0 for day to day usage

          • johnjohn says:

            For some reason I can’t edit my comment even though I just made it?

            Whether it’ll cross from above freezing to below freezing outside, is probably the most important piece of information a weather forecast can give you.

          • Tibor says:

            It probably depends on where you live, but in most of Europe, there is snow and ice in at least parts of winter and there 0=water freezes is actually very useful.

          • John Schilling says:

            Fahrenheit is better just because going from 0-100 for most temperatures you will encounter is better than going from -18 to 38.

            For you, maybe. What about he guy in a slightly different climate who mostly encounters temperatures of -20 to 40 and isn’t keen on your plan to convert to an archaic system where he has to deal with the -4 to 104 range?

            “The endpoints are nice round numbers if I define the range where the endpoints are nice round numbers” is a bogus metric.

          • Artificirius says:


            If your accuracy is roughly 5 degrees F, then that is about an accuracy of 2 degrees C. If the complaints of resolution are based around not liking decimals (Which can only be taken as slightly meaningful in temperature and weather at that), then you’re still not needing a finer resolution.

            If you were able to A) tell the surrounding temp at a feel within one degree F and B) had an extreme hatred of fractions or decimals, that would sound like an internally consistent reason to use Fahrenheit. As it stands, all it sounds like is just what people are used to, which is not an argument of merits.

            Also, out of curiosity, do you live somewhere with a largely temperate climate with little daily or seasonal variation?

        • Dr Dealgood says:

          Day-to-day use by ordinary people is a nice feature, but ultimately the point of standardized weights and measures is technical. Celsius is the superior unit because virtually all science is done in celsius or kelvin (quick quiz: who even knows what Rankines are without Googling?).

          I like American units, I’m used to them, but converting back and forth between those and SI units is a hassle and introduces errors.

          • hlynkacg says:

            who even knows what Rankines are without Googling?

            I do! But then I’m in a job where I’m looking at pressure, temperature, and viscosity calculations on a regular basis. Speaking of which; I have a calibrated pressure gauge labeled in slugs/ft^2 that I use to annoy metric chauvinists. 😉

          • JayT says:

            “Celsius is the superior unit because virtually all science is done in celsius or kelvin”

            That’s not terribly convincing to me. How many people in the US actually have to deal with worrying about conversions? One or two percent at most? Also, a lot of the time you’re going to have to convert from Celsius to Kelvin anyways, so the chance of error is still there, even if it is slightly lessened.

          • The Nybbler says:

            I knew Rankines (or degrees Rankine, unlike with Kelvins no one objects to calling them that, another advantage), they’re the sensible alternative to Kelvins, absolute temperature units the size of a Fahrenheit degree.

            @hlynkacg: Was that gauge actually intended for use or is it a gag? I’ve heard of slugs but never seen them used except by physics instructors to scare their students metric; pound-mass is much more typical. And using slugs for pressure (when psi, pound-force per square inch, is the basic pressure unit in the same system) would be bizarre.

          • Controls Freak says:

            who even knows what Rankines are without Googling?

            If anyone who does science doesn’t know what a Rankine is, they shouldn’t be doing science. I live in a more abstract world now (where I pretty much never deal with units at all), but I have plenty of memories of fluid properties charts/tables/equations in Rankine.

            What I find deeply disingenuous about the “all science is done in metric” song and dance is that all science is actually done in whatever units happen to be convenient. Sure, a lot of fields branch off from some SI starting point, but that’s hardly where things stop. Literally everyone uses minutes/hours/days/years… and they’re not divided by decimal orders. I’ve met a lot of “metric chauvinists” (as hlynkacg puts it), but only one person who regularly works on practical problems with radians instead of degrees (the built-in spellchecker here doesn’t even recognize “radians”; that should tell you something). If you’ve ever done orbital mechanics, you’re used to canonical units. If you’ve done control theory or acoustics, you’re used to decibels. If you’ve done jet propulsion, lots of things are still in lbf (and specific fuel consumption in lb/(h*lbf)!). If you’ve done fluids experiments, you sure as hell know what a millimeter of mercury is, and you probably have memorized the conversion factors to bar, atm, and Pa. If you’ve done quantum mechanics, you know what an electron volt is. If you’ve done relativity, you don’t even bother with having separate length and time units!

            Understanding how to work with different unit systems is a prerequisite for science, not a hindrance thereof.

          • CatCube says:

            Structural engineers use slugs for seismic. (And one of the scientific fields where you will almost never see metric in day-to-day use; if we have a metric building code anywhere in my office, I couldn’t produce it for you if you put a gun to my head.)

            Pounds-mass are an abomination unto God, since you have to carry g in the denominator throughout the equation. If you just convert all masses to slugs at the start, you can use F=ma without having to worry about that nonsense.

            Use pounds as a unit of force and slugs as a unit of mass, and the math is just as easy in the US customary as in the metric system.

          • hlynkacg says:

            @The Nybbler,
            It’s an older gauge, but it checks out. (still passes Cal.) It was made for testing/calibrating aircraft pito-static systems back in the day and includes slugs on the dial because the NACA Standard Atmosphere table is in imperial units and doing so removes a conversion step in the relevant density and viscosity calculations.

            Sure I could use one of the new digital gauges but sometimes it’s fun to bust out the chunky old analog unit and do that shit by hand. 😉 Besides, the NACA table is still considered a valid calibration source by the FAA and is often cited directly in the manuals of older aircraft so why not use it?

            I also second everything Controls Freak said above.

      • Tibor says:

        I think that 0=ice is actually very useful. And for cooking, 100=boiling is also sometimes good. I guess the 0 becomes less useful when you live somewhere where it rarely ever snows.

    • Artificirius says:

      Then you run into issues of peoples paces, thumbs (inches), feet and whatever else are wildly different. If you standardize said measure, then it ceases to be a useful measure for casual use via peoples actual body parts and you may as well then use something like a fraction of the distance light moves in a vacuum in a second.

      The issue with Imperial is only that it’s divisions are nonsensical and different. Why are there a thousand mils in an inch, but only 12 inches in a foot? Why are there 3 feet in a yard but 1760 yards to a mile? If the mile is particularly wet why is it a bit over 2025 yards instead?

      If someone tells me that some places is 15 km away, I have a vague idea about the distance, but if they tell me it is 1000 paces, I have a lot better intuition.

      Scaling issue. A thousand paces is no large distance. Neither really is 15km, when you get down to it. If someone told you something was 3000 km away, is this impression of distance more or less vague than someone telling you its 4347826 paces away?

      • powerfuller says:

        It’s obvious nobody’s foot is actually a foot long, but for a quick-and-dirty approximation, it usually works out, even with the differences in body size. The standardized foot doesn’t have to suit everybody equally to be useful to most people most of the time. The mile = 1,000 paces was based on the Roman legion, right? In which case, at least for them, that would have been fairly uniform.

        That being said, “Eh, my thumb’s about an inch,” is only marginally easier than, “Eh, my thumb’s about two and a half centimeters.” I wonder if anybody’s ever compared the abilities of metric and imperial system users to estimate distance, weight, etc. It would be hard to control for other factors, but if one group were consistently better at estimating, would that settle the question?

        I feel sorry for Canadians, who have ended up using some convoluted mixture: Celsius for weather but Fahrenheit for cooking; meters for distance but feet for height…

        • Artificirius says:

          Tell me about it. As an electrician in Canada it is maddening. Height or distance can be in metric or imperial, often on the same blueprint.

          And I don’t see why estimating should be a decent measure. Ideally, set up sets of three companies in a variety of industries. One uses purely metric, one uses purely imperial, and the other gets to use the standard North American hodge podge of both. See who fucks up the most in producing their goods/services.

      • The original Mr. X says:

        Why are there 3 feet in a yard

        Isn’t the yard meant to be the average length of an adult’s pace?

      • Eric Rall says:

        Why are there 3 feet in a yard but 1760 yards to a mile?

        This is a consequence of optimizing unit sizes for specific utility rather than for consistent conversion factors. A mile was originally exactly 1000 double-paces in Roman infantry drill, and a yard is a slightly-longer standardization of a single-pace, so there’s a bit less than 2000 yards in a mile.

        There are also several intermediate-sized units that have fallen into disuse except in specific technical contexts (mainly land surveying). There are 5.5 yards in a rod, 4 rods in a chain, 10 chains in a furlong, and 8 furlongs in a mile. Still different, but not so wildly different as the foot:yard vs yard:mile ratios.

        A furlong’s about the size of a city block (exactly the size of a city block, in many cities), and was originally notable as being the standard length of a medieval English plot of farmland (the width being 1 chain, and the area being 1 acre: the 10:1 length to width ratio being due to plowing technology making it more efficient to plow a long, narrow field than a square one).

        If the mile is particularly wet why is it a bit over 2025 yards instead?

        Because a nautical mile is 1 minute of arc (1/60 of a degree) along a great circle of the Earth’s surface, which happens to be slightly longer than a regular mile, so it gets called a variant on “mile” as a hint of about how big it is. The “nautical” part comes from the unit historically being useful in ocean navigation, where you’re measuring locations in terms of degrees of latitude and longitude.

        • Deiseach says:

          Well, this is plainly “make Deiseach feel old” day on here, because Lord God, back in the dim and distant days when I was learning measure in primary school, we learned about miles, furlongs, and “rods, poles or perches” 🙂 And did none of you learn measuring by using the “span” of your hand (from thumb-tip across to the tip of the lúidín), which approximated to six inches?

          It’s all metric nowadays, of course (save that furlongs are still used in horse racing).

          • Eric Rall says:

            I don’t use spans, but I do use paces (right around a yard), cubits (from the elbow to the tip of the middle finger, which is just over half a yard on my arms), palms (width of the hand across the knuckles, excluding the thumb, which is about three inches on my hands), and thumb widths (right around one inch). I’ll convert paces to feet, but I tend to use cubits, etc straight, since the imprecision relative to the standard doesn’t matter if I’m the one measuring the things I’m comparing.

      • CatCube says:

        I use approximate measures based on my body pretty frequently. If I need to measure a rope or cable, the span of my arms is close enough to six feet that I can just count out the number of times I pull the rope taut.

        If I need to figure out the approximate measure of, say, a concrete slab, I can often just use my closed fist with my thumb extended, as that’s pretty close to 6 inches, and many slabs or structural members are multiples of 6 inches. (Obviously, if I need a closer verification, I’ll get out a tape measure, but for quick checks it saves some time.)

        As far as why 1000 mils per inch, that’s because inches used to only be used in dyadic fractions, that is, the denominator is a power of two. You wouldn’t originally use 1/1000th of an inch, you’d use some multiple of 1/1024, but machinists in the 19th century started using even thousandths, and called it the “mil”.

        Land surveyors usually use tenths and hundredths of a foot, and for longer lengths will use survey stations, which are either 100 or 1000 feet, depending on the survey. (If you see a paint marking on a roadway like “2+23.23” that means that the mark is 223.23 feet from the start of the survey; 2 stations + 23.23 feet. If they use 1000 foot stations, it’ll be 0+223.23)

      • Tibor says:

        True, but most of the time you go 15km somewhere rather than 3000 km.

        As others pointed out, of course my thumb is slightly different than your thumb. But if you standardize the inch based on some population average, it will be a reasonable approximation for both of us. The point is not to be precise, for that you still need a measure anyway.

        As for 1 inch vs. 2,5 cm, I find 1 easier to work with than 2,5. Then again, I’ve always sucked at calculation, that’s why I ended up doing maths and not accounting 😛

        The nondecimality of the imperial system is an issue. It has its reasons – it is easier to divide 12 by 3 or 4 than it is for 10, which is why a pound has 12 pence and not 10 and it was similar with other currencies, If you want to cut a piece of cloth – again the same. But while useful in the middle ages and for when you don’t have a calculator or a measure, nowadays this creates more problems than it solves, so a decimal imperial might be a nice compromise. And since 10 vs 12 is not such a big difference, 1 decimal foot still roughly corresponds to an actual person’s foot. So as I said, a decimal imperial system would give you the best of both worlds.

        • The original Mr. X says:

          which is why a pound has 12 pence and not 10

          Actually it’s the shilling that had 12 pence. There were 20 shillings in a pound, making a pound 240 pence.

    • Matt M says:

      If someone tells me that some places is 15 km away, I have a vague idea about the distance, but if they tell me it is 1000 paces

      Wait what?

      I have a pretty good idea of how long it would take me and how far I would get if I walked 1km. I have absolutely no idea how far I would get or how long it would take to walk 1000 paces.

      • Tibor says:

        1000 paces = 2000 steps. You can immediately see what your speed in steps is if you have a wristwatch.

    • Machina ex Deus says:


      So for weight I would use the weight of a cup of (say distilled so that it is really precise enough for when you need to be precise) water as the basic measure and again have it decimalized.

      The weight of a cup of water is already 8 ounces, or exactly half a pound. “A pint’s a pound, the world ’round.”*

      At least for 16-ounce pints: beer is sold in Imperial pints, which are 19.2152 U.S fluid ounces**; if you’re European, you can just think of them as 568.261 milliliters.

      On their own terms, Imperial pints are 20 Imperial fluid ounces, so 12 Imperial pints make a fluid shilling (or possibly a fluid guinea; Deiseach can correct me).

      (* If you’re paying a pound for a pint of beer, I guarantee you’re not in London.)

      (** I’m sure there’s a reason behind that, likely having to do with Medieval kings, relative beverage taxation rates, and the phase of the moon.)

      • random832 says:

        The US fluid ounce is based on a medieval standard gallon (of 128 ounces) which is 231 cubic inches. The Imperial fluid ounce was standardized in 1824 and is 1/160 of a standard gallon which is ten pounds of water measured at a temperature of 60 degrees Fahrenheit. The number of ml per imperial fluid ounce is therefore much closer to the number of grams per ounce (though not exactly, because the metric system’s weight/volume equivalency was based on a lower temperature)

        The wine gallon that became the US gallon was supposedly defined by Queen Anne, so no medieval kings.

    • Deiseach says:

      Instead, I propose cup as the basic measure. Cups are already used in US recipes (as well as old European cookbooks which do not try to pretend that it matters whether you use 5% more or less of an ingredient) and it makes cooking faster. So for weight I would use the weight of a cup of (say distilled so that it is really precise enough for when you need to be precise) water as the basic measure and again have it decimalized. The same goes for volume – a cup of water would be the basic measure.

      What size cup? Tea cup? Coffee cup? Mug? Unless you have a standardised “cup for baking”, and if so then you may as well use a measuring jug and weighing scales. Half a pound of butter is easy enough, there are blocks of butter in that size. Need quarter of a pound? Cut a half-pound in half, that’s enough of an estimation.

      • CatCube says:

        It’s a standardized cup, as used in baking. 8 fluid ounces, or 1/16th of a gallon. “Tablespoon” and “teaspoon” are also standardized measures, of 1/2 and 1/6th fl. oz., respectively.

        1/4 pound of butter is basically 1 cup, or 8 tablespoons. (In the US, 1/4 lb sticks of butter usually have markings on the label so you can measure out tablespoons of butter by cutting)

        • Tibor says:

          I expect that most mugs in the US are actually standard size and so are teaspoons and tablespoons, right? That makes cooking easier, since it turns everything into a measure.

          • CatCube says:

            No. I don’t know if that was true at one time, but an 8 oz cup used in cooking is generally considered smaller than a standard serving these days.

            I actually don’t know what my coffee mugs are, so I just went in the kitchen and filled different ones with a measuring cup. Both of the ones I commonly used were larger than 8 oz, somewhere between 9-11 oz, but both were different.

            One thing your comment about coffee mugs reminded me, though: a “cup” as used as a measure of coffee is 6 fl oz. This was a standard pour of coffee when they standardized the measurement on coffee pots. So when a coffee maker has gradations on the carafe for the number of “cups,” you won’t get nearly that many coffee mugs out of it, because standard portion sizes have risen in the years since.

          • Tibor says:

            @CatCube: That’s a shame. I’d imagine that they’d advertise the “correct” cups as “this is a standard cup size, perfect for cooking!” or something 🙂

            A coffee cup should be something like a cup for an espresso, no? That is tiny, maybe like 3-4 teaspoons. Although, judging by Starbucks and similar franchises, Americans like their coffee large (apparently, everything seems to be larger in the US, based on what people told me).

          • random832 says:

            A coffee cup should be something like a cup for an espresso, no?

            Unmarked “coffee” in the US refers to drip-brewed coffee, not espresso. Drip-brewed (or percolated, the standard before the 1970s) coffee has a lower concentration (of both flavor and caffeine) than espresso, which would be the reason the term “Americano” was coined for watered-down espresso.

          • The Nybbler says:

            Speculating here: A coffee “cup” might be based on the once-ubiquitous diner coffee cup, rounded with a small base made to sit on a saucer, which is 7-8 oz and not filled to the top. The thing most Americans put their coffee in is a 24oz thermos coffee mug, which is cylindrical and considerably larger.

        • Polycarp says:

          @CatCube: “1/4 pound of butter is basically 1 cup, or 8 tablespoons.”

          Correction: 1/4 pound of butter is basically 1/2 cup, or 8 tablespoons.

          A pint’s a pound the world around.

          • The Nybbler says:

            A pint’s a pound the world around.

            * Offer valid in US, Liberia, and Burma only

          • CatCube says:

            What’s really embarrassing about that is I knew 1/2 stick is 1/4 cup, and I multiplied 1/4 by 2 and got 1.

    • Mark V Anderson says:

      I would really like the US to go to metric. This is for two reasons:

      1) I really love the use of decimals. It is so much easier. Note this reason doesn’t hold for temperature.
      2) The rest of the world uses metric, and sometimes I do need to talk to people outside the US. Quite often, actually. This reason does hold for temp.

      I don’t get Tibor’s comments about imperial measures being based on common objects. I have a feel for the size of inch, foot, yard, mile, pound, ounce, gallon, etc., but this has nothing to do with common objects. I could get just as used to meter, gram, and liter. I do not find either Celcius or Farenheit inherently superior, but I’d like to use what the rest of the world uses.

      • Tibor says:

        Well, an inch is roughly the width of your thumb. A foot is, well, the length of your foot and in fact it is about 10 widths of your thumb anyway. Then you can go decimal from there. It is more pronounced with length. Weight is kind of abstract either way, which is why I’d have it based on something useful in cooking, like the weight of a (standardized) cup of water. And everything decimal, since that is actually more useful in the modern world, especially if you intend to use the same measure for precise measurements as well as for everyday imprecise use.

        • Nancy Lebovitz says:

          The width of your thumb? An inch is roughly the length of my thumb.

          How big are people’s thumbs, anyway?

          • Nancy Lebovitz says:

            I meant that the top joint of my thumb is about an inch.

          • hlynkacg says:

            Length? Seriously? How tall are you?

            My thumbs are 3/4ths of an inch (just under 2 cm) wide and 3 inches long. For the record, I’m 6’1″ and 190 lbs.

          • Tibor says:

            Mine is about (using a bendable plastic ruler) 24-25 mm in the widest point. Slightly less than an inch. The length of the last join of my thumb is about 29 mm. Maybe for women the length of the last joint is closer to an inch and for men the width.

          • The Nybbler says:

            The length of the distal part of my left thumb from knuckle to tip is about 1″. Right thumb, 1 1/4″. Proximal joint, 1 11/16″, both thumbs . Width of left thumb at thickest point (near knuckle), 1″. Right thumb, 15/16″. Good thing I’m not king, we’d need different units for left and right.

          • Cerebral Paul Z. says:

            My thumbs are yuge, just like the rest of my hands. Don’t let anyone tell you different!

          • Nancy Lebovitz says:

            I’m 4’11” (150 cm).

            My thumb is 5/8″ (1.6 cm) wide.

        • Montfort says:

          Thumb width appears to be very slightly sexually dimorphic, but generally around 20-23mm.

        • Paul Brinkley says:

          Bunch of commenters waving their thumb lengths at each other, bragging about whose is closer to / farther away from the international standard inch. #onlyOnSSC

    • Cheese says:

      The ‘advantages of the imperial system with respect to commonly dealt with units’ is pretty much entirely ‘I grew up with this or was exposed to it a lot and so i’m in the habit of thinking it in this way’ in my opinion. I’ve also seen a fair bit of ‘it’s easier to say’, which I think broadly amounts to the same thing.

      I mean I have pretty accurate internal references for pretty standard metric amounts like 1kg, 1m, 100m or the like. If you start talking in paces or cups i’ve got no frame of reference and i’m out. Even when we’re talking building materials i’m still far more comfortable talking about ‘mils’ than 4×2 or whatever you guys use.

      If you want to make a completely new decimal system with different references then sure go ahead. But there already is one so that’s a bit of an effort for everyone isn’t it.

      • The Nybbler says:

        You need too many fractions with metric.

        Celcius temperatures are too big; you have to resort to half-degrees to get acceptable precision.

        There’s no unit akin to the fluid ounce; thus you get wine in fractions of deciliters, and soda in “.333L” or worse “.30L” cans, instead of the perfect 12oz (“.355L”) size.

        There’s also no unit akin to the foot; it’s either way too many centimeters or fractions of a meter.

        The lack of divisors of 10 means that when estimating, you only get (even) fifths and halfs instead of thirds, fourths, and halfs. Fifths are much harder to visualize than thirds or fourths.

        Also, you’re not fooling anyone with your “25mm” drive sockets.

        Even when we’re talking building materials i’m still far more comfortable talking about ‘mils’ than 4×2 or whatever you guys use.

        Well, dimensional lumber is a whole different can of worms. A modern 2×4 S4S is 1.5″ by 3.5″. European dimensional lumber comes in all sorts of sizes; there’s no super-common set of sizes like in the US.

        • Tibor says:

          I disagree with the Celsius being too big. There is little difference between 19 and 19.5 degrees. In fact even the difference between 19 and 20 is minute. 19 and 21 is already noticeable. I can’t think of a non-industrial or non-scientific situation where the difference of 0.5 degrees Celsius is important at all.

        • Artificirius says:

          No, you generally get 355 mL cans. Nothing under a full litre generally gets fractional L measurements. Not that it matters for day to day life. A can of pop is a can of pop. Whether it is labeled is mL or fl oz is irrelevant.

          In day to day life, all measurements are useless, since the vast majority of people won’t actually use them for anything. For people whose life or work requires them to use actual measurements of things, then fractions are an utterly unavoidable fact for either system, and the various sixteenth and thirty-second fractions of an inch for fine measure are not, accuracy wise, different than millimetre.

          Moving a decimal point around is easy.

          • The Nybbler says:

            I’ve gotten those short cans of coke in Europe. backs me up with a 330ml offering (so way too many ml rather than fractional liters in this case). I’ve definitely gotten the way-too-small 300ml cans too.

          • Artificirius says:

            Sorry, I wasn’t meaning that there are only 355mL cans. Just that sub litre liquid measure aren’t done with fractions of L measure, as a convention. Whether it’s ‘too many’ mL seems to be a subjective thing.

      • Tibor says:

        I grew up with metric.

      • Nornagest says:

        Yes and no. Base-12 units have more factors; they’re easy to divide into thirds or quarters, which is a big advantage in a lot of day-to-day scenarios. So I’d say inches and feet have a leg up on metric there. Miles are an obstacle, but I rarely find myself needing to convert between miles and feet.

        Unfortunately, the customary units for weight and volume don’t even have a consistent base (volume is quasi-binary, but then there’s teaspoons and tablespoons; weight makes no fucking sense), so cooking, where this sort of conversion would be most useful, is actually harder in them.

        • CatCube says:

          A tablespoon is 1/2 fl oz, so it continues the binary.

          The teaspoon being 1/3 of a tablespoon breaks it, though.

          • Nornagest says:

            Sort of, but it skips a level. There are common measures for gallons down to 1/4 cups and some half steps, but there isn’t one for 1/8 of a cup — that’s defined as 1 fl oz, but I’ve never seen a measuring cup for it.

            The tablespoon fits better than the teaspoon, though, I agree.

          • The Nybbler says:

            I have liquid measuring cups with ounce markings (no number, just an unmarked gradation below 2). Measuring spoons for dry ingredients of 1 fluid oz are not common but available, they’re usually marked as 2 tbsp.

            (in a fit of rationality, somewhere it was decided that the larger US dry measures — based on a gallon of roughly 269 in^3 — would not be used for measuring dry ingredients, and instead the fluid measures based on a 231 in^3 gallon would be used for everything. Yay!)

            I believe the teaspoon was originally 1/8 ounce (same as the dram) but got bloated to its current size.

    • Nornagest says:

      Metric has the advantage of making it easy to go between mass and volume if you’re talking about things about as dense as water, like people, dirt, most liquids in common use, and hardwoods — a kilo is the mass of a liter of water, which is pretty easy to visualize. Contrast customary, where a gallon is eight-some pounds. Yeah, “a pint’s a pound, the world round”, but I can only visualize pints in terms of milk cartons (almost all other liquid-producing industries use fluid ounces or have gone metric).

  5. Tibor says:

    What feeling do you get from Azbuka (the Russian alphabet)? I’ve been able to sort of read it with some effort ever since someone told me to read it like Greek with sharp edges, but it somehow looks aggressive and not very pleasant to the eye to me. Latin alphabet is much prettier, in my opinion. I don’t know if it is a feature of the alphabet itself or some associations I might have with it.

    But for example this picture shows Russian in azbuka and then in roman letters and the second just seems much nicer to me. Similarly, I think the Japanese should just fix a particularly well done Romaji. That would make their language considerably easier to learn. Since it is not even tonal, there is no reason not to use the Roman alphabet. I can of course still understand that they would be reluctant to do so, just as I would not want to switch to writing in Azbuka and the English won’t switch to driving on the right (or the other way around).

    • Creutzer says:

      “azbuka” means “the alphabet” in the sense of “the canonical sequence of letters”, not the script. You’re talking about the cyrillic script (“kirillica” in Russian).

      For the purpose of representing Russian in particular, it’s slightly superior to romanisations, which have to rely on either diacritics or digraphs or both.

      My experience is that, having learnt cyrillic as an adult, I read it rather slowly. But this is not a big deal, and Russians in turn learn the latin alphabet quite early. So there would be little benefit to switching. It’s not at all comparable to the preposterous Japanese situation.

      • Tibor says:

        I see, I thought the word azbuka referred to the script itself. In Czech the word is used specifically to describe the Russian alphabet, so I thought that was what the Russians were calling it, not that is simply means “alphabet”.

        I don’t know, most other Slavic languages use Roman letters and work just fine. True, Czech does not have the soft and hard versions of letters, for example (I believe that is what is indicated by the small “b” next to Cyrillic letters), but Polish or Slovak do (although perhaps only with the letter “L”, I’m not sure) and neither uses Cyrillic.

        • Creutzer says:

          Yes, but Polish uses lots of digraphs and some diacritics, and Slovak and Czech use lots of diacritics. That’s my point. From a purely aesthetic point of view, Russian written in cyrillic, which has few digraphs and no diacritics, is clearly superior because it looks much calmer and neater.

          • Tibor says:

            Creutzer: Czech uses diactritics a lot but only really has two accents, the line like in “á”, which just means the vowel is supposed to be pronounced longer (something other languages often only denote implicitly based on letters around the vowel) and the hook like in “č”. Portuguese has five for example. I would change the hook for ~ like in Spanish. The Czech “ň” and the Spanish “ñ” are pronounced exactly the same but “ñ” is prettier.

            There is also “ů”, which should really be removed from the language because it is pronounced exactly like “ú”. But in the middle ages “ů” would actually be written like “uo” and pronounced like “uo”, then they switched to “ů” but kept the pronunciation and eventually the distinction disappeared entirely (still a couple of hundred years ago).

          • J.S. Bangs says:

            If Russian has no diacritics, then what do you call that thing hanging out above the letter й ?

          • Creutzer says:

            Czech and Polish orthography are both very regular, and from that perspective unobjectionable. But as I see it, there are two problems: first, you need a special keyboard and thereby lose one benefit of using Latin. Second, they interrupt the flow of handwriting. That’s why I count them as a slight disadvantage.

            Also, c, z and s with diacritic ~ look weird, and using the háček on n as well is pleasingly consistent.

            @J.S. Bangs: I forgot about that. I apparently perceive this thing as one letter, not as letter plus diacritic, but you’re completely right. Still, it’s less pervasive than in Polish and Czech. And the letter is quite rare and in native words occurs only at the end of words, which means it leads to less of the aforementioned interruption in the flow of handwriting.

          • Tibor says:

            @Creutzer: I don’t know, I still like the tilde more than the hook above all letters. It is not so sharp looking and makes the font look rounder, which I find prettier (maybe that’s why I don’t like the Cyrillic which is full of sharp edges). Although I think that the Greek alphabet might be too round, so that it is hard to read in print.

            I guess the incidence of accents is still higher in Czech than in Portuguese (which seems to have quite a lot of them), but most European languages have some accent which interrupt the writing. In German, the umlaut consists of two dots (although I’ve seen people write it as a solid line in handwriting, which also looks prettier, I’d say), which means you have to rise the pen twice. But actually, unless you write in cursive, which most people don’t today, you still have to raise the pen for every letter. I do recall that it was annoying to write the accents when we learned to write at school – we learned cursive, now they’ve switcher to some hybrid font, essentially something close to how people normally write (although I do know a few people who write in cursive and even one person who writes everything CAPITAL LETTERS, which is extremely slow, at least when I do it).

            On the keyboard, it either does indeed make writing a bit slower or necessitates a different keyboard. The Czech keyboard uses the numbers above letters for accented letters (almost all of them, you still have to write ň by first pressing the accent and then the letter, but it is a very rare letter), so it ends up being faster than writing Portuguese on a Portuguese keyboard, where you only have ç as a key and you have to write áàâõ using a two key combination. But the price you pay for that is that the symbols @#$%^&* are in less handy places (SHIFT+a key close to the enter key, SHIFT+keys above letters actually does numbers).

            I think the Czech (and German) keyboard should also switch from QWERTZ to QWERTY. I use four keyboards on my computer, half of them use QWERTZ and half QWERTY and it is sometimes quite annozing. Well, at least it is still better than the French AZERTY.

    • Montfort says:

      Personally I like Cyrillic, as it gives Russian and other languages that use it a more exotic feel while not being nearly as intimidating as, say, learning Chinese writing.

      As far as Japanese goes, are you referring just to kana, or to all of their character sets? Kanji does actually afford them some advantages (besides attractive appearance, backwards compatibility, tradition, etc) – it allows them to distinguish between homophones. Granted, it doesn’t do so perfectly – to the best of my understanding, many homophones are very close in meaning and native speakers may sometimes disagree about which to use in a given situation. But to get this effect in the roman alphabet, they’d have to invent non-standard spellings like English does (e.g. carat/caret/carrot).

      • Creutzer says:

        But to get this effect in the roman alphabet, they’d have to invent non-standard spellings like English does (e.g. carat/caret/carrot).

        Except they wouldn’t have to do that at all because there is no real need to distinguish between homophones in writing. Have you ever been in a situation where spoken language was being used and you had to ask for clarification which of the homophones was intended? No? Well, I have. Once in my entire life. And it was about the homophony between intensional and intentional, both of which had a technical meaning in the context in question. In short: Indistinguishability of homophones is not an actual problem.

        There is no excuse for a writing system that people become proficient in at, what, age 16? (I’m probably exaggerating, didn’t check the number.) Not that English writing is particularly defensible, either…

        • Montfort says:

          there is no real need to distinguish between homophones in writing. Have you ever been in a situation where spoken language was being used and you had to ask for clarification which of the homophones was intended?

          Even if I were prepared to accept this as true for English, I wouldn’t think it would necessarily apply to Japanese.

          • Tibor says:

            Why not? Also, homophones exist in other languages as well. In Czech the word “kohoutek” means either literally a small cock (as in male version of a hen) or the water tap (I think it is because the old-fashioned water taps look a little bit like the cock’s crown thingy on top of his head). It is very unlikely that you would mix those two (also, they’re declined differently, since one is living and the other is not, but even in nominative there is virtually zero chance you’d confuse the two meanings).

            In German you also use Wasserhahn (“watercock”) to mean the water tap but I am not sure if you can omit the Wasser. If you do, it is again the same thing.

            And most homophones are like this, their meanings are very different and are hard to confuse in context.

          • Creutzer says:

            Even if I were prepared to accept this as true for English, I wouldn’t think it would necessarily apply to Japanese.

            If the Japanese had to constantly go out of their way to ask which of several homophones their interlocutor just uttered, I guarantee you one of those homophones would die out very quickly.

          • Montfort says:

            Because different languages are different? Japanese is not particularly closely related to English, German, Czech, or even Chinese (it is possibly related to Korean, but that is apparently debated).

            My intuition would be that most languages have some number of homophones, but that doesn’t imply that the number or inherent ambiguity is consistent across languages. Just as some languages have larger vocabularies or use more phonemes than others, so too might they have different homophone distributions.

            But besides the inapplicability of evidence from other languages, I’d also note that Japan allegedly has an uncommonly high number of homophones, and those homophones can have many meanings (~20 in some cases), which can be very different or almost identical. To give an example (pathological) Japanese sentence in the roman alphabet:

            “Kisha no kisha ga kisha de kisha shita.” allegedly means “a reporter in your company came back to the company by train”

            (I keep saying “allegedly” because I am no great scholar of japanese. Given his name, perhaps Onyomi could chime in?)

            I don’t mean to imply that it would be impossible to transition to the roman alphabet – for years basically all Japanese typing had to be done with kana, and they muddled through somehow. But the kanji do serve a purpose, even if you don’t see it as sufficient justification for their existence.


            If the Japanese had to constantly go out of their way to ask which of several homophones their interlocutor just uttered, I guarantee you one of those homophones would die out very quickly.

            And if their spoken and written vocabularies had the same distribution of words used, I’d say you have a point. But they don’t. I mean, they could just give up using the more confusing homophones, sure, but that’s something you’re asking them to give up. It’s a reason to keep kanji, whether or not you think it’s sufficient reason to keep kanji.

          • The original Mr. X says:

            @ Montfort:

            If it were really impossible to tell what somebody was saying based on the sound of the words alone, it would be impossible to communicate verbally. Since people obviously can communicate verbally in Japanese, it follows that the sound of Japanese words is in fact sufficient to convey meaning, and hence that you don’t need separate symbols for each homophone to communicate.

            “Kisha no kisha ga kisha de kisha shita.” allegedly means “a reporter in your company came back to the company by train”

            And “Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo” is apparently a grammatical sentence in English. Coming up with trick sentences proves approximately nothing about how easy or hard it actually is to communicate in day-to-day conversations.

          • Montfort says:

            I believe I understand your argument, and it seems broadly the same as Creutzer’s. Perhaps I can explain myself more clearly. If the role of written language were solely to record spoken language and written language could perfectly encode spoken language then there would be no point in distinguishing homophones.

            However, that’s a big if, and the premises are inaccurate:
            1. The role of written language is not solely to accurately record spoken language. Most of us do not write the same way we talk. Languages often have vocabulary, grammatical rules or cases, and constructions used almost exclusively in written language.
            2. Not all information passed through spoken language is recorded in written language. For instance, inflection creates a great deal of meaning in spoken Japanese (and sometimes English), but is not encoded in writing.

            Specifically, we can see these two objections in Japanese: When writing, one can use homophones which would normally be indistinguishable in speech (for instance, some homophones mean largely the same thing but different “spellings” carry different connotations). If forced to write all words phonetically, this would no longer be possible. (Similarly, there are a large number of jokes that rely on this homophony to work).
            Additionally, spoken Japanese can encode some of the different homophones and other context in intonation/inflection (which is not written and varies by region), so some distinctions present in spoken Japanese would require different “spellings” for homophones.

            Once again, I’m not saying it would make it impossible to communicate, but you lose something besides the nice-looking characters. There’s no “real need” to distinguish homophones in the sense of straight-up losing the ability to talk about weather, for instance, but there is a “real need” in the sense that the feature is currently being used to create meaning in a way not possible without it.

          • onyomi says:

            Yes, Montfort is right that the written style of Japanese is different and, without using kanji, more ambiguous than the spoken language, which makes less frequent use of Sino-Japanese compounds (which are more ambiguous in Japanese than Chinese because Japanese did not, like Chinese started to 2,000ish years ago, develop tones in compensation for the loss of phonemic complexity of Old Chinese). (The example sentence depends somewhat on substituting obscure words for common words, and would more likely be something like “kisha no kisha ga densha de kisha shimashita,” the meaning of which most Japanese could probably guess based on context, though they’d likely not phrase it that way in speech.”)

            Generally speaking, this is not a problem unique to languages using logograms: rather, all languages have the problem of to what extent they want their written language to closely reflect the spoken reality of some particular vernacular, and to what extent they want the written language to conserve etymological information and historical distinctions which the descendants of the spoken language it was based on no longer differentiate.

            For example, English spelling seems very illogical compared to e. g. Spanish. But the spelling of say, Irish, and, I have heard, Thai, is even worse. Yet if you are a linguist you love these weird old spellings because they actually tell you something about how the word was pronounced a thousand or more years ago, and, in many cases, something about the etymology. It would seem logical, for example, to just reduce the three words, “rain, reign, and rein” to just one spelling (IPA, ideally), since they are all pronounced the same. But to do so increases ambiguity of the written language, which in all languages offers fewer context cues and tends to be somewhat more terse, and also hides the differing etymologies.

            So there is a sense in which a conservative approach to writing is a nuisance for beginning learners but an aid to scholars and what we might call “advanced users.” It’s not just about the convenience of a few scholars: I find Japanese with no kanji to be much slower and more annoying to read than Japanese with kanji, for example. Books are printed in Japanese without kanji–they are for children and have to use a limited vocabulary.

            Korean also provides an interesting example of adopting the alternate strategy: basically dropping the kanji. My Korean is not nearly as good as good as my Mandarin or Japanese, so I can’t comment exactly on the experience, but I find reading Korean to feel more like staring at an undifferentiated mass, rather than having the very iconic kanji jump out at me like flagposts. As someone already able to read Chinese, it also makes things harder, rather than easier, because most of the words of Chinese origin are easier to learn when I know their hanja etymologies.

            In other words, choosing between reflecting a particular spoken language and representing a more cosmopolitan, historical written language, Korean chose to focus more on the former, while Japanese preserved more of the latter. I tend to prefer the latter approach for a variety of my own reasons, but both have pros and cons.

            Written languages are always a step or more behind spoken reality; the question is how tightly you want to force it to follow: if the distance becomes too great, learning to read and write correctly becomes unnecessarily difficult, but excessive narrowing of the distance cuts the spoken language off from older written sources and other languages which share a common history (example: written Cantonese and Mandarin are mutually intelligible; spoken, they are not; if you replaced the Chinese characters used to represent Cantonese with a phonetic system, it would become easier for Cantonese speakers to learn to write, but at the expense of making their language harder to learn or read for speakers of any other Chinese language).

          • onyomi says:

            One other point: switching e. g. Cantonese or Taiwanese or Shanghai dialect or whatever to a phonetic writing system would not only make it harder for non-natives of those areas to learn and read those languages, it would also make it harder for native speakers of those languages to read any Chinese documents or writing printed before that time or in other regions.

            Also, though I’d reiterate the important difference between style and function of written and spoken languages, I do agree with Creutzer’s point that, especially in the spoken language, people wouldn’t keep using homophones that consistently resulted in ambiguous sentences read aloud. To take the example sentence, even in writing, much less in speech, a speaker/writer of modern Japanese would probably use a more liberal mix of words of Chinese, Japanese, and Western origin, like “kisha no ripootaa (reporter) ga densha de kisha e modotta (native Japanese verb),” eliminating the ambiguity of using almost exclusively formal Sino-Japanese words (brevity of literary Chinese+sound changes of modern Mandarin can become so ridiculous as to allow you to write a whole paragraph using only one syllable (with different tones), which doesn’t imply that a more normal paragraph would be unintelligible if read aloud, even using classical Chinese and reading in modern Mandarin).

            In related news, Japanese Prime Minister becomes object of ridicule for incorrectly reading kanji out loud. “Ah hah!” one might say, “this proves kanji are too difficult even for educated native speakers!” But this is actually a relatively unusual case of someone reading out loud a very formal style of written language rarely used for anything intended to be read out loud (and certainly no one says “unnun” in daily conversation). Sort of like running into some word in English you not infrequently see written but almost never hear spoken out loud, such that you’re not totally confident when the rare occasion arises for actually saying it. He still should know this word, but there’s also the fact that if you record anyone long enough you will capture them uttering a serious brain fart at some point.

  6. Deiseach says:

    Stories I never thought I’d read:

    (a) “Wexford hurling star Lee Chin” – granted, I have no cause to be so surprised after Seán Óg Ó hAilpín

    (b) An Irish lad being “shocked” by “drinking culture” in sports, though granted again, for an amateur organisation and sports played by (still pretty much) amateurs, the GAA has become very professional in its attitude to training etc.

    Wexford hurling star Lee Chin has admitted that he was floored by the apparent ‘drinking culture’ that exists in professional ice-hockey, writes Paul Keane.

    The star of Davy Fitzgerald’s resurgent Model County setup spent a week with pro side the Vancouver Canucks recently as part of the AIB commissioned ‘The Toughest Trade’ documentary series.

  7. Anonymous Bosch says:

    Trump will nominate Scott Gottlieb as FDA commissioner instead of Jim O’Neill or Balaji Srinivasan, the two candidates our host was excited about. Lower your estimates of Thiel’s influence and raise your estimates of cable news. (Gottlieb is a frequent FOX guest.)

    His selection would be a victory for the mainstream pharmaceutical industry, which has little appetite for upending the FDA approval process. Even Democratic critics of the industry, and of Gottlieb himself, privately indicated they preferred him to another frequently mentioned contender, Jim O’Neill, an associate of Silicon Valley billionaire and Trump adviser Peter Thiel. In a 2014 speech, O’Neill called for allowing drugs to be marketed after they had been proven safe, but before they had been shown to be effective.

    I hope everyone learned a valuable lesson about privileging the hypothesis.

    • Deiseach says:

      No, this actually sounds quite sensible and moderate – instead of getting everyone’s backs up (by forcing a candidate that the pharma companies don’t like and will go out of their way to oppose and trip up), they appoint a compromise candidate who takes away the excuse from the pharma companies of “we don’t like this guy because his position is too extreme”, so it’s more likely they’ll be willing or amenable to changes he may make.

      As well, it gives an interesting look into the question of “Does the FDA hold up new drugs too long?” You could say “Yes, because the big pharma companies don’t want the situation to change and don’t like the idea of quick approvals, as seen in their opposition to O’Neill”, but it also shows that the fault isn’t completely that of the FDA. If life-saving drugs really were being held up by unnecessary red tape, the big pharma companies could lobby for changes; that they’re not doing so – well, that depends on how cynical you want to be about it.

      They don’t care if people die while waiting for treatment, so long as they are the ones who get their drugs passed or the less (but still cynical) the “life-saving” drugs aren’t that life-saving or life-changing, so a year or two delay won’t make that big a difference really to mortality rates.

      • Anonymous Bosch says:

        There’s zero indication that Gottlieb plans to institute any of the reforms discussed by Scott. And it’s not like he’s some rando with no record to examine, he was deputy commissioner during the Bush admin. By far the most likely scenario is he continues the status quo.

        You’re engaging in exactly the same hypothetical privileging Scott did except instead of a hypothetical where Trump appoints whoever Peter Thiel wants it’s a hypothetical where an established Republican bureaucrat (with an established gravy train route between him and big pharma) suddenly develops an interest in reform.

        • Deiseach says:

          My point is that he is an established Republican bureaucrat, so he has a better chance of getting some reform (and no, I’m not expecting him to announce drastic overhauls within the first ten minutes of getting the job) past the vested interests.

          If the very mention of O’Neill or Thiel had the pharma companies getting into siege mode and “Whatever he says, we say ‘no!'”, the idea that “hey, this is a guy we know, we know what to expect from him, he’s not likely to do any crazy stuff” is going to reassure them somewhat and make them more amenable to discussion and change.

          Worst thing that happens? He continues the status quo, as you said. Not ideal, but surely better than a reformer who crashes and burns because from the get-go everyone is fighting them tooth and nail? And then in future whenever anyone raises the prospect of reform, everyone shakes their head and points to “He tried it, and look what happened”?

          • rlms says:

            But the point is that he has no reason to try to get reform. Are you saying that Trump actively wants reform, and that Gottlieb is a canny choice to push it through? I think the excitement about O’Neill and Srinivasan was from the idea that they’d push for reform, and succeed due to Trump’s indifference.

          • cassander says:


            I’d bet more today that trump actually wants reform than I would have a few months ago. The administration has moved pretty aggressively on de-regulation. Nothing has come of it yet, of course, and nothing might yet, but he’s made far more noise in that direction already than I expected in january.

          • Deiseach says:

            I’m saying that if O’Neill, Srinivasan, Thiel, etc. are still involved somehow or somewhere with the Trump administration, they can still propose their reform-minded ideas and have them considered, and that getting these ideas passed on to the “pharma companies like this guy” person in charge will have a better chance than outright proposals by “he’s an outsider and we’re gonna say ‘no’ to everything” selection.

            The whole “a spoonful of sugar helps the medicine go down” gambit, you know? I’m not suggesting anything one way or the other because I genuinely don’t know the mindset of Trump re: any of this and who is in/out of favour or has decided to drop away. All I am saying is that generally a compromise candidate has a better chance of getting things done because the other side feels like it has won something or is having its points taken seriously, so they are more inclined to go in with “negotiating” than “opposed no matter what” mindset.

            And that if Trump (or whoever is guiding policy) selected a compromise candidate, then that was sensible and reasonable and unlike the image being peddled of “forcing his picks on everyone no matter how bad they are”.

          • Nyx says:

            I think this is a tad optimistic, yes, maybe DJT is playing 11d chess and his selection of a boring mainstream bureaucrat that gives every appearance of not wanting to reform the FDA is actually a calculated move that will end in total reform and nootropics being added to the water supply of every major city or whatever it is that people want, but that logic could apply for any awful pick.

  8. Levantine says:

    Roger Schank: Enough with the AI is going to take over the world stories. ….. Enough with AI people who barely know the first thing about AI. …. I can fight back. Not with words, which I know don’t really convince anyone of anything, but with a new AI tool, one that uses what I know about AI, in other words, what most people who worked in AI in the 60’s, 70’s, and 80’s likely know about AI ………

  9. BBA says:

    Does anyone have any experience with sleep studies, CPAP, etc.? I’m looking into getting tested for sleep apnea – no matter how much sleep I get, I still have trouble getting up and feel tired during the day.

    • skef says:

      No, but: an easy thing to try on your own first that might work.

      How is your heartburn/acid reflux/etc.? Mine was so-so, and I thought tied mostly to eating and not a big deal. Then I tried omeprazole for a week and started getting better sleep than I had in many years. (I didn’t even put it together how bad my sleep had gotten until it got better.)

      So anyway, something you might rule out before a real sleep study.

    • Corey says:

      Yep! I did this about 10 years ago or so, and it made a big difference. If you drive, it’s important – falling asleep at the wheel (which I used to struggle to not do) can ruin your whole day. My primary care physician suggested this after finding me napping while waiting in the exam room, two appointments in a row.

      Like any healthcare, never ever go out-of-network, for the sleep study or (if indicated) buying the CPAP. (e.g. Apria will bill $900 for a mask that runs $140 in-network).

      Insurance will have you rent the machine for a while (6-12 months) and check (via SD card and the company who sells it to you) that you’re using it and it’s working well, before approving its purchase.

      For the study, they’ll wire you up with EEG, EKG, leg electrodes (to see about restless legs), O2 saturation monitor and such, and watch you sleep. If they see apnea they’ll put a CPAP on you and have you sleep some more, futzing with the pressure until the apnea goes away. Usually this can be done in 1 night (after which you will NOT be well rested) but sometimes takes 2. Also you’ll need to shower to get EEG-electrode goop out of your hair (they might have a shower there for the purpose). The electrodes can be detached if you need to go potty, there’s audio monitoring so you just say so.

      Many masks just cover the nose but I need one that covers nose & mouth because I cannot keep my mouth shut while sleeping (childhood allergies mandated mouth breathing).

    • Well... says:

      My wife got diagnosed with sleep apnea and started using a CPAP while we were dating. What do you want to know?

      • BBA says:

        Basically, how well does the machine work? Is the improved sleep worth the discomfort and inconvenience of having to wear this thing? Also, I’m curious about the study itself, and whether it’s really necessary to spend a night in a hospital bed covered in electrodes, etc.

        And (for anyone in the NYC area) if there’s a particular specialist to seek out or avoid.

        • Well... says:

          My wife would say yes, it’s totally worth it. She said it’s like she never even realized what a good night’s sleep was until she started wearing a mask, which didn’t take very long at all to get used to. (Though occasionally she does grumble about her life sentence of having to wear a big thing on her face every night.)

          If I remember right (it was back when we were dating, maybe 10 years ago!) she did have to spend the night in a hospital or clinic covered in electrodes, or being monitored in some other way. This wasn’t a problem for her because she’s very comfortable with those kinds of medical environments. For me, with my mild phobia of hospital equipment, it would have been a nightmare.

          Recently her original CPAP died and she had to get a new one. That involved having to go in and see a specialist again so that everything would be properly calibrated, but it did not require an additional sleep study. Though as I understand it, it could have under other circumstances.

          I’m not in the NYC area, got no help for you there.

    • dndnrsn says:

      I was diagnosed with sleep apnea back when I was considerably fatter. I could never adjust to the machine – it wasn’t the discomfort of the mask, which was fine, but rather the feeling that the air was “stale.” Eventually I lost a bunch of weight and daytime tiredness without an obvious cause (patterns in caffeine consumption, having slept terribly due to some other factor) went away.

  10. Matt M says:

    Re: the camgirl survey in the link topic

    As someone who spends way too much time/money on camgirls, I find it to be a really interesting market. Is anyone here familiar with the story of Olivia4Naked? I can get into it if people are interested, but basically this girl became one of the Internet’s biggest camgirl success stories, sometimes raking in thousands of dollars per hour, with *very* little nudity, nothing sexually explicit, and most nights not even uttering a single spoken word. It was modern day performance art. Super fascinating. To me at least!

    • Anon. says:

      I googled her, and the first link I clicked on included this:

      There was that time she dressed up as Bin Laden – the beard on her chin had a matching one at her pussy. There was that time she did a striptease dressed as a “sexy” Hitler, complete with a nazi flag, podium, and memorized a famous speech of his and recited it…in German.

      You should definitely get into it.

      • Matt M says:

        This isn’t nearly as interesting as it sounds. Dressing up was kind of her gimmick for awhile. Pretty much every ethnic stereotype, profession, whatever, was represented at some point. For the most part it was just as if she went to Party City before halloween and said “I’ll take one of everything.”

        Although her “hobo” costume was used pretty frequently and was a really weird mix of arousing and offensive.

        The Nazi thing got some mainstream press and condemnation. The site actually shut her down during the middle of it, even though she was raking in several thousand dollars during it (and word is, the site gets about half). She “retired” a few weeks later.

      • Matt M says:

        Here’s a fan tribute video that tries to capture some of the magic. If you don’t want to spend ten minutes, recommend starting at like 4:20 and lasting as long as you can.

        The lack of content/information/web presence out there for her adds to the mystique imo. She always played very loose and inconsistent with facts about herself. There are a ton of legends. For awhile it was rumored she never spoke because she was deaf. Then, for fun, she hosted a talk radio segment for a few weeks. But it’s utterly amazing to me that for someone who was probably making six figures worth of voluntary donations, there aren’t any real fansites or anything to enshrine her legacy or what have you. Sometimes I almost wonder if I hallucinated the whole thing…

    • Dr Dealgood says:

      Do you mind if I ask some questions? I’ve never met someone who admitted to watching camgirls so I’m kind of curious.

      What do you see as the benefits of camgirls over free non-interactive porn? For example, paying to watch a camgirl live versus seeing the same video up on a streaming site a week later.

      Is there a difference in your mind between paying a camgirl or a stripper IRL? How about a camgirl and a regular pornstar?

      • Matt M says:

        Personally, I use camgirls and strippers for social interaction more than arousal. It’s a substitute for dating, not a substitute for porn. The girl I spend most of my time/money on now is topless only and I never really…. erm…. satisfy myself to her (and the bouncers don’t look favorably upon that sort of thing in strip clubs). The benefits are that you build up a personal relationship. Yeah, it requires that you pay them a bit, but the arrangement is sort of nebulous. I throw whatever amount of money I feel like at them whenever I feel like it, and I basically get a cute girl as a friend I can chat with online and sometimes see her boobs. The non-specific attribute of the arrangement makes it feel less creepy and weird. Also there are a ton of camgirls, it’s ultra competitive. Meaning you can find exactly what you want, down to a compatible personality.

        The line between camgirls and pornstars is getting increasingly blurry, and there’s a lot of moving in both directions. Famous pornstars can make a lot of money camming with virtually zero effort just on their name recognition. Meanwhile, building up a cam following is a great way to break into porn as you basically bring a pre-installed userbase with you. If you want to do professional porn at all, that is. A ton of camgirls produce and sell their own amateur stuff, sometimes solo, sometimes with other girls, sometimes with guys.

    • Randy M says:

      It was modern day performance art.

      Isn’t performance art a fairly recent phenomenon? Or is modern day already post-(post-?)modern?

  11. IrishDude says:

    Federal Court Rules Snuggie is a Blanket

    Fascinating article about a subject I knew little about: tariff engineering. Companies dedicate resources to trying to get their products into different categories that are subject to lower tariffs. They use legal means to change classification or actually alter their product. Some examples:

    *The tariff rate on imported blankets is 8.5 percent while for imported “pullover apparel” it’s much higher at 14.9 percent. Therefore, Snuggie spent money on lawyers to sue the U.S. government to get their classification (successfully) changed from clothing to a blanket.

    *”America’s high tariffs on sugar have encouraged companies to shift to importing cake mixes and other sugar-rich products instead”

    *”when the U.S. imposed a tariff on motorcycles with 700-cc engines and larger in the early 1980s in a bid to protect motorcycle company Harley Davidson, Japanese competitors simply started making a 699-cc version instead”

    *”Halloween costumes walk a similarly thin line, as NPR’s Planet Money has reported. By using a Velcro closure instead of a zipper or button, some products are more likely to be counted as “festive articles” than clothing, which is subject to much higher tariffs.”

    *”Friedman says he has tracked the practice of tariff engineering as far back as 1882, when the Supreme Court ruled on a case involving a sugar importer. Duties on sugar were based on the product’s color, so an enterprising company colored its sugar with molasses to get around the tax.”

    *”For years, Ford has imported its Transit Connect van as a passenger van — imports which the U.S. taxes at a rate of only 2.5 percent. But once the vehicles enter the United States, Ford sends some of the vehicles to a nearby facility, where workers rip out the rear bench seats and replace the rear windows with solid panels to make the vehicle into a cargo van. In the process, Ford circumvents a hefty tariff of 25 percent on imported cargo vans.”

    *Example of unsuccessful tariff engineering: “In one 1991 case, a court ruled against a company that had circumvented high duties on feathers by importing them as feather dusters instead. After the dusters arrived in the country, the company took them apart to make them into boas or put them on hats. Because the articles weren’t sold in the condition they were imported, the court decided this was “artifice.””

    Tariff engineering is such a waste of resources. If we can’t achieve my preference of no tariffs, it seems like charging the same tariff rate across the board would be the next best policy. It would allow lawyers, engineers, and lobbyists to engage in more productive activities.

    • dodrian says:

      In the UK there’s the Jaffa Cake Controversy. Why there are the different tax rates in the first place I don’t know.

      • IrishDude says:

        My hypothesis is the tariff rates are different primarily because companies/industries want special favors, and politicians find it valuable to provide those special favors. It would be interesting to compare lobbying dollars (or political contributions) for industries to tariff rates for that industry.

        EDIT: On the Jaffa Cake topic, my hypothesis predicts that the domestic cake makers would have weaker lobbying than the biscuit makers. Not sure how easily that can be checked. What complicates things is that there could be historical lobbying differences that get baked into the tariff rates, and inertia keeps those rates going long after the lobbying stops.

        • BBA says:

          The Jaffa Cake case is about a VAT, not a tariff. It applies to domestic and foreign goods equally.

          • IrishDude says:

            Got ya. Either way, I’d suspect special interest considerations play a role in differential VAT treatments between cakes and biscuits. This is one of those topics, like a soda tax, that could also be playing to the public being less averse to taxes of ‘sinful’ products like sugary cakes.

          • rlms says:

            Actually, biscuits and cakes (like most food) are normally untaxed. However, chocolate-covered biscuits are considered luxurious, and subject to the normal 20% tax rate. Chocolate cakes aren’t, hence the attempt to classify Jaffa Cakes as such (I think the argument was that cakes, including Jaffa, go hard over time whereas biscuits soften). Since the taxation of chocolate biscuits is an exception to the 0% rate on most food, it seems unlikely that the difference would come from lobbying. VAT replaced the Purchase Tax that was instituted in WWII to encourage people not to buy “luxury” products (as resources were needed for the war). I can’t find the different categories used there, so I don’t know if the chocolate difference originated from that, but it seems plausible (the 0% rate on books comes from then).

          • IrishDude says:

            Interesting about the WW2 origins for the tax on luxuries. There’s an inertia to taxes/regulations/bureaucracies such that even when the underlying reason for them goes away they seem to hang around forever.

          • The original Mr. X says:

            Interesting about the WW2 origins for the tax on luxuries. There’s an inertia to taxes/regulations/bureaucracies such that even when the underlying reason for them goes away they seem to hang around forever.

            Cf. income tax, first introduced in the UK to help fund the Napoleonic Wars.

          • LHN says:

            And Europe remains Napoleon-free to this day![1] Does anyone really want to take the chance of that changing by messing with the tax system?

            [1] Observation void 1852-1870.

          • The original Mr. X says:

            And Europe remains Napoleon-free to this day![1] Does anyone really want to take the chance of that changing by messing with the tax system?

            Hmm, good point. Perhaps we’d better re-introduce the wig tax, just to be on the safe side.

          • John Schilling says:

            And Europe remains Napoleon-free to this day!

            If we bring back Napoleon and run him against Macron, Fillon, and Le Pen, how does he do in the polls?

          • bean says:

            Hasn’t repealing Income Tax due to the fact that Napoleon isn’t a threat been on the Raving Loony platform forever?

          • Deiseach says:

            If we bring back Napoleon and run him against Macron, Fillon, and Le Pen, how does he do in the polls?

            Given that several (well, three) locations are tussling over the bones of Marengo, Napoleon’s horse, perhaps surprisingly well?

            Fianna Fáil councillor Bernard Moynihan believes returning the remains of the famed horse to the town of Buttevant would be an excellent way of boosting tourism in the area, which is the site of the annual Cahirmee Horse Fair.

            However, standing in the way of the bones of Marengo returning to Cork are its current owners National Army Museum in Chelsea in London, and a rival claim from the Bartlemy horse fair, around 40km away.

            …“Well you know, I’ve had no contact with Bartlemy on this issue,” Moynihan said. “I’m working on the absolute, verifiable facts I have from local historians that Marengo was sold at Cahirmee Fair.”

            Marengo is said to have carried Napoleon into famous battles such as the Battle of Waterloo.

            The horse survived the colds of Moscow, and survived the French retreat from Russia in 1812.

            The horse passed into British hands after the Battle of Waterloo, and lived until the year 1831 when his remains were preserved and later passed to the Royal United Services Institute.

            Marengo is believed to be the horse captured in paintings such as Napoleon Crossing the Alps by Jacques-Louis David.

    • Corey says:

      In nerd-news circles, there was a nice example of this a few years back: X-Men ruled human by a court, to settle just such a tariff question (different rates apply to human and non-human dolls; to which group do X-Men action figures belong?)

    • suntzuanime says:

      Here is an article that provides a more in-depth background on tariff engineering. (I suspect the Tribune article drew on this article for the background it provided.)

    • BBA says:

      For a non-tariff-related example: In Europe “alcopop” drinks like Smirnoff Ice, Mike’s Hard Lemonade, etc. are made by mixing vodka, soda water, and flavorings. In the US they’re made by fermenting malt into something meeting the legal definition of “beer”, then filtering out all the malt flavor and adding the intended flavorings for the product. This is for both tax and regulatory reasons: the beer excise is less than the hard liquor excise, and in many states beer can be sold in supermarkets while liquor has to be sold in dedicated liquor stores.

      Note that the “added flavors” for malt beverages can themselves contain alcohol, and in the early 2000s many “malt beverages” derived up to 99% of their alcohol content from those “flavorings,” thus making them almost totally non-beer while still meeting the legal definition of beer. A 2005 regulation closed that loophole by requiring a majority of the alcohol content in a malt beverage to come from the malt.

      This topic also brings to mind the case of White City Shopping Center v. PR Restaurants, in which a judge officially ruled that a burrito was not a sandwich. And that was purely a contractual dispute, so even in your ancap utopia we won’t be rid of this bullshit.

      • IrishDude says:

        This topic also brings to mind the case of White City Shopping Center v. PR Restaurants, in which a judge officially ruled that a burrito was not a sandwich. And that was purely a contractual dispute, so even in your ancap utopia we won’t be rid of this bullshit.

        Hey, if businesses want to voluntarily bind themselves I’ve got no problem with that. This dispute seems like a good learning lesson going forward for Panera to better define ‘sandwiches’ or specify burritos if they make non-compete contracts. On this subject, I wonder whether a court would rule a hotdog is a sandwich.

        My vision of ancap isn’t that it would be utopia, by the way. Just that it would be a net improvement morally and economically to a system where people submit to political rulers.

  12. rlms says:

    Is eating human brains morally wrong (or dangerous, if it’s cooked)? Also, does anyone know if that episode was about Hinduism in general, about the Asghori, or about a wide variety of religious sects? It seems objectionable in the first case, but not the others.

    • Eltargrim says:

      (or dangerous, if it’s cooked)

      Dangers associated with human brain consumption are at best only partially mitigated by cooking.

      Various prion diseases, including CJD, vCJD, and Kuru, tend to have prion concentrations in the brain. These prions spread the disease upon consumption, and to my knowledge are not degraded by any cooking process.

      I’m not going to say eating human brain is inherently dangerous, as prion diseases are generally quite rare, but there is nothing that can be done to mitigate any risk from prion disease aside from abstaining from consumption.

      • rahien.din says:

        These prions spread the disease upon consumption, and to my knowledge are not degraded by any cooking process.

        Prions so potent and so durable that surgical tools that touch CJD-infected CNS tissues are simply thrown away. Adequate sterilization methods degrade surgical instruments just as badly as the prions!

        Good luck with that frying pan…

      • keranih says:

        Quibble: CJD is not a TSE (transmissible spongiform encephalopathy) like kuru, scrapie, nvCJD, BSE, the mink & feline variants, etc. CJD is an inherited tendency to produce/collect the misfolded proteins, rather than a contagious one.

        And the degree to which they can be shared from organism to organism is not yet well defined – it’s very difficult to give scrapie to goats, while genetically susceptible sheep can pick it up from the environment that scrapie-laden sheep were on years previous. (We’ve never seen people with a TSE from sheep either, despite having known about scrapie for centuries.)

        So, yes, nvCJD/BSE should be treated like the incredibly hot risk that it is…(*) but that doesn’t mean all TSEs are like that. The trade off risk from consuming brains is really pretty low, and if that ceremony is recycling vital protein and binding your community together, I can see people thinking it is worth the risk.(**)

        (*) I mean, “incredibly hot for something that killed somewhere around 300 people across the globe over two decades.” Not “malaria hot” or “rabies hot” or “dengue hot.”

        (**) This whole week is just a lesson in “there is no accounting for taste” isn’t it?

      • Le Maistre Chat says:

        Various prion diseases, including CJD, vCJD, and Kuru, tend to have prion concentrations in the brain.

        “Concentrations” implies that they’re present in other types of tissue. So I guess rationalists can’t eat human flesh at all unless it’s a matter of survival?

        • The Nybbler says:

          “Concentrations” implies that they’re present in other types of tissue. So I guess rationalists can’t eat human flesh at all unless it’s a matter of survival?

          I’m personally OK with that.

      • Nancy Lebovitz says:

        Any thoughts about why prion diseases are concentrated in the brain rather than equally present in the rest of the body?

    • Le Maistre Chat says:

      I was a practicing Hindu for a few years and I never heard of the Asghori.
      Like modern (esp. American) Christianity, there is no disincentive for heresy in Hinduism, which leads to a bewildering array of beliefs and practices of widely varying popularity. The Asghori might be an ancient tantric sect, or they could be a recent invention being sensationalized by the media as a weakman, ala Westboro Baptist Church. Even with personal experience it’s hard to say.
      Since Hinduism is actually pretty obscure to Anglophone culture, I can see USINPAC’s concern with misrepresentation. It’s the same as any minority’s concern with media representation, except here it seems to be ONLY Hindus complaining, rather than minority+affluent white allies, which is what you’d see if Islam was being depicted negatively.

      • rlms says:

        This article suggests they are very obscure (about 100 members). I don’t know when they originated.

        I think mainstream Hindus have a very good reason to complain about misrepresentation if the report on the Asghori presented them as “one of the many types of Hindus”, e.g. by filming them as part of an episode on Hinduism in general. But I don’t think there is a good case if there was a whole episode on them, or they were presented as “one of many fringe groups of various religions that have wacky beliefs”. Just producing a documentary about the WBC or ISIS isn’t misrepresentative.

    • Deiseach says:

      Definitely not about Hinduism in general; a quick and ignorant off the top of my head* is that the Aghori are one of the sub-sects of Shiva worshippers who take the ascetic, renunciate, destruction and overturning of social structures elements and the breaking through to enlightenment by deliberately breaking taboos so as to liberate the mind/soul from bonds of earthly attachment (this is the philosophy of avadhuta). They therefore are ascetics and hermits, do severe penances and practices, go to cremation grounds, etc. They smear themselves with the ash from cremation grounds and are alleged to use human bones (e.g. using skulls to make cups) but whether they practice actual cannibalism is another question; the guru of the Aghori is Dattatreya and this incident may explain the origin of the idea:

      Lord Dattatreya, an antinomian form of Shiva closely associated with the cremation ground, who appeared to Baba Keenaram atop Girnar Mountain in Gujarat. Considered to be the adi guru (ancient spiritual teacher) and founding deity of Aghor, Lord Dattatreya offered his own flesh to the young ascetic as prasād (a kind of blessing), conferring upon him the power of clairvoyance and establishing a guru-disciple relationship between them.

      Something comparable in Western thought would be Dionysius and the Bacchantes for the same enlightenment-via-transgression, or the Stylite saints and “Holy Fools” in Christianity (e.g. why Rasputin could be taken as a holy man).

      *Look, the kind of stuff you find yourself looking up on Wikipedia to make sense of who the various sages, gods and events are in Indian mythological TV series on Youtube covers a wide if shallow range.

  13. Dr Dealgood says:

    So, has anyone else been reading Ozy’s Trans Turing Test results?

    I’m honestly a bit disappointed, because so far the three submissions I’ve seen have all opened up with some variation on “gender identity is the same as biological sex the vast majority of the time but sometimes it isn’t.” Rhetorically it’s a very weak opening for a gender identitarian, because it opens the door for legitimate questions about why subjective self-perception should matter more than objective biological fact. Generally if you encounter those folks in the wild the first thing they do is start trying to muddy the scientific waters if not reject ‘biotruth’ outright.

    Beyond that, only one of them bothered to engage the question about the so-called ‘cotton ceiling.’ This is IMO the most interesting point of contention: it pits the two very popular ideas of disparate impact and sexual autonomy against one another. If refusing to have sex with transsexuals is bigotry, because every woman (or ‘woman’) deserves equal consideration as a sex partner, then a lot of other even less palatable conclusions follow. But by the same token, allowing lesbians the freedom to distinguish between women and transsexuals in the bedroom opens the door for straight men to do the same.

    Plus the comments are boring.

    • rlms says:

      I think the reason for the lack of engagement with the “cotton ceiling” is that the authors were not gender identitarians, and either couldn’t or wouldn’t argue convincingly in favour of it.

      I wouldn’t imagine that many of Ozy’s readers (or even radical feminists in general) think that lesbians are allowed to have preferences against trans women, but straight men aren’t.

      • dndnrsn says:

        I think part of the confusion is that the people most likely to say “yeah this is a good thing” with regard to that are neither Blanchard-Bailey folks nor gender identitarians, but TERFs, who don’t really fall into either group, do they?

        • rlms says:

          I thought there was a big overlap between BBs and TERFs (at least between BBs, and TERFs who read Ozy’s blog). I’m not sure though.

          • dndnrsn says:

            Aren’t most TERFs fairly socially constructionist, moreso than the B-B supporters?

    • Le Maistre Chat says:

      Generally if you encounter those folks in the wild the first thing they do is start trying to muddy the scientific waters if not reject ‘biotruth’ outright.

      The “lol biotruth!” progressives are baffling. Back when I was 14 and rejected fundamentalist Protestantism, the biggest reason was that I liked science and was fed up with the community constantly undermining biology with creationism. Back in those days, it seemed that all secularists were Darwinists. Then in the late Noughties I started seeing the “biotruth” sneer from progressives, as though Darwinian biology was a rival worldview.
      I DO realize that the charitable interpretation is that they’re only sneering at EvPsych, but if we evolved from apes, why wouldn’t that be a useful predictor of behavior?

      • Dr Dealgood says:

        One thing which has helped me immensely is separating out people who “believe in” science and those who understand it.

        Believers can turn on a dime without realizing it. Fred Clark’s history of how abortion went from being seen as a Catholic preoccupation to a centerpiece of Evangelical thought within a generation is a good example of that.

        If anything, understanding science will make you sound less sincere in your defense of it. People who hear the certainty and conviction of a pseudo-scientist are convinced, whereas real scientists reflexively hedge. So it’s not surprising to me that sociologists and other pretenders to science were able to turn the I Fucking Love Science crowd against the hard sciences.

        • Incurian says:

          FWIW, this comment made me feel all warm and fuzzy inside. This is the culture war I’d rather be fighting.

        • Nornagest says:

          I get the sense that science as in I Fucking Love isn’t even a belief, more of an aesthetic. One can fucking love Hubble photos or Discovery Channel documentaries (at least, before it turned into 24/7 sharks and home improvement, although I suppose someone must still love that), but it’s pretty hard to fucking love natural selection, let alone a fast Fourier transform.

      • Iain says:

        To steelman “lol biotruths”: evo-psych done well is valuable. Evo-psych done poorly is an attempt to promote the societal norms of the day into eternal truths based on a superficially plausible just-so story, and deserves to be mocked. Consider: “women like pink because their brains evolved to pick berries”, which is somewhat complicated by the fact that “pink for girls, blue for boys” is less than a hundred years old.

        It’s not a rejection of science — just a rejection of science done badly.

        • cassander says:

          “pink for girls, blue for boys” is less than a hundred years old.

          As I often say, most of what we think of as storied and ancient traditions were inventions of the mid to late 19th century. this goes double for any “national” traditions.

        • lvlln says:

          I think this shows an overactive immune response, which is a pattern I think I notice among the “lol biotruth” crowd. Because shoddy evo-psych has been used in the past to promote societal norms of the day into eternal truths, often with oppressive implications, they respond not by being super-cautious to evo-psych and being on the lookout for shoddy research, but rather by just rejecting the whole field altogether and automatically equating any sort of evo-psych with the worst forms of abuse perpetuated by shoddy versions thereof.

          To speculate further, I think this also reflects a lack of understanding of science. By conflating science with beliefs that are scientifically accurate, they don’t recognize or acknowledge that it’s possible to correct previous shoddy research in any field – that this is basically what science is. Instead, once they see that it’s been shoddy and produced silly conclusions, they consider it discredited.

          That’s all just me trying my best to make sense of it based merely on my observations, though. I would be surprised if I weren’t missing some huge chunk of the story.

          • Douglas Knight says:

            Scott’s commentary is not perfectly clear, but his source agrees with Iain. It traces how the comment Iain makes got distorted into the claim that the colors reversed.

          • LHN says:

            The correspondence certainly isn’t less than a hundred years old. A quick Google Books search show it treated as a commonplace in 1856. (“A layette (that is, baby-clothes) must be prepared. Shall it be a layette of pink, for a girl, or a layette of blue, for a boy?”)

            Mid-nineteenth century as cassander suggests is still possible, though the way it’s presented suggests that it was established at least somewhat earlier.

        • Brad says:

          evo-psych done well is valuable.

          Would you please point out a relatively self contained paper that you consider an example of evolutionary psychology done well?

    • The Element of Surprise says:

      But by the same token, allowing lesbians the freedom to distinguish between women and transsexuals in the bedroom opens the door for straight men to do the same.

      So the underlying assumption here is that men doing this is somehow worse than women doing it?

      • Dr Dealgood says:

        Well I mean that seems to the goal that all of this is driving at.

        • Deiseach says:

          I think it’s because if you insist on “the genitals aren’t important, it’s the personality” (I’ve seen rather too much “my bigoted grandmother said why did I have to pick a girlfriend instead of a boyfriend, and I told her that I didn’t fall in love with a set of genitals but with a person – ha ha, I sure schooled her!” which doesn’t seem to recognise that Bigoted Grandma can then retort “So find a guy with a personality you like, if it’s not the genitals you care about when having sex!”) about trans women (or men) then the corollary of that for lesbians/gay men is “so you can hack yourself to find the genitalia associated with the opposite gender sexually attractive, it’s not innate and unalterable born this way, so why can’t you be straight?”

          If you’re telling a woman who isn’t oriented to having sex with people with dicks that she should be equally open to be having sex with a woman with a dick as a woman without one, then you’re not far from telling that woman that she should be open to having sex with a guy with a dick, or at least your argument that “but she’s a lesbian and lesbians don’t like sex with people with dicks, that’s what being a lesbian means” is weakened.

    • ADifferentAnonymous says:

      I think what you’re seeing is Ozy’s blog and/or the ITT format selecting for moderates. This was also an issue in the previous ITT there, SJ vs. anti-SJ.

      Not sure what you mean by most posts “not engaging” the cotton ceiling question; they all answer it, and they mostly acknowledge the same dilemma you point out. If by “engage” you mean “definitively choose one side”, then they don’t do that because it’s dumb?

      • Dr Dealgood says:

        Not sure what you mean by most posts “not engaging” the cotton ceiling question; they all answer it, and they mostly acknowledge the same dilemma you point out. If by “engage” you mean “definitively choose one side”, then they don’t do that because it’s dumb?

        You can’t answer a question by just repeating it. The whole point of the question was to see which side of the dilemma the author fell on: do you bite the gender identity bullet and tell lesbians that transsexuals are women who they have no non-bigoted reason not be attracted to, or do you acknowledge that having a cock and balls is something of a natural barrier to lesbianism.

        The latest one, Gender Identity #6, literally refused to answer the question. #4 had an interesting strategy of initially playing dumb about what the cotton ceiling was (reads blogs but can’t use Google?) and following it with a one-paragraph non-answer. Only #3 and #2 gave actual opinions and hesitantly at that.

        It’s a shame because it’s the only question where the rubber meets the road, so to speak. If you come down on the side of breaking the cotton ceiling, you’re advocating for lesbians to be pressured into straight sex. If you come down on letting the issue alone, you’re implicitly recognizing that ‘trans lesbian’ is a nonsense concept. Not engaging with it is the coward’s way out.

  14. rlms says:

    Question for anyone who’s done some academic philosophy:
    Are there any philosophical questions that seem very difficult or controversial to non-philosophers, but are considered settled amongst people who actually study them?

    • Protagoras says:

      Hmmm. Philosophers prefer to focus on unsettled issues, making this unusually complicated. A majority of philosophers seem to think atheism is too obvious to be worth much further discussion, but the subfield of philosophy of religion is particularly packed with the minority who disagree, so whether that comes anywhere close to “settled” depends on how you define “people who actually study it.” I would say that it is the near unanimous opinion among philosophers that subjectivism/relativism, as understood by non-philosophers, is an incoherent view, and similarly that the common non-philosopher’s view of free will makes no sense, but in neither area is there a consensus position about the correct view, and in both cases there are people who will defend what are intended to be more sophisticated derivations of the fatally flawed common sense positions.

      I guess there are debates that people don’t know much about, but often have strong intuitions once they encounter them, where there is a general philosophical consensus that those strong intuitions are misguided; e.g. logical issues like material conditionals. Having difficulty coming up with more plausible examples. Maybe that the meanings of words and sentences is determined by the intentions of the people uttering them? Whatever other debates there are about meaning, there’s a pretty good consensus that that view doesn’t work.

    • Philosophisticat says:

      I agree with what Protagoras says. I’ll add that there’s a lot more consensus on whether certain arguments for views are good than about whether the views themselves are correct. Even for these there’s always someone out there who defends the views, but this is true in the sciences too.

      Also, it tends to more often go the other way – plenty of things that seem obvious to people but philosophers think are very difficult.

      To add to Protagoras’ list, off the top of my head:

      -certain moral issues – the wrongness of homosexuality, for example.
      -certain moral views – Divine Command Theory, for example.
      -psychological egoism
      -stuff about souls?
      -there’s a kind of naive falsificationism common among laypeople and scientists alike that philosophers of science think is untenable.

      Things that are not consensus but where I expect there are notable differences in how philosophers and nonphilosophers pattern:

      1) The wrongness of eating factory farmed meat
      2) The wrongness of incest

      • Purple says:

        How do you think philosophers pattern about factory farming and incest?

      • Philosophisticat says:

        I think philosophers, particularly ethicists, think eating meat is wrong at a disproportionately high rate relative to nonphilosophers, and think incest (between consenting adults) is permissible at a disproportionately high rate.

        • Purple says:

          Ah, sorry, I thought you meant they make arguments different from those of the general public, e.g. incest is wrong but for a reason other than genetic defects/it’s nasty.

          Sounds like I’m half a philosopher though.

    • skef says:

      The first to come to mind is “Doesn’t that depend on how you define the words?” No, that’s not how language works.

      I think it’s at least unclear to many people whether something has to be true for you to know it, but that’s the consensus view in philosophy and it doesn’t usually take too long to convince someone, even though there is some linguistic evidence weighing against.

  15. nimim.k.m. says:

    Given that we have had constant stream of statistics related questions (along the line “I don’t get this or that, please help”) lately (and will probably as long as Scott is going to keep on blogging about medical research and related topics), should we get some kind of a more or less organized statistics self-study club going on?

    edit.Or something less formal, like a collaborative list of book and internet material recommendations?

    • Dr Dealgood says:


      I’d for one like to know of a good book on statistics. Statistical education is weird and spotty, so something that starts at the beginning and walks the reader through how this stuff works would be very helpful. Something like “Mathematics: It’s Content, Methods and Meaning” but for stats (and maybe with fewer intrusive aside about communism).

      • Schibes says:

        I’d for one like to know of a good book on statistics.

        Well, Nate Silver seems popular enough on this blog, what better place to start than with his book?

        Its top reviewer on Amazon raves:
        “This is the best general-readership book on applied statistics that I’ve read. Short review: if you’re interested in science, economics, or prediction: read it. It’s full of interesting cases, builds intuition, and is a readable example of Bayesian thinking.”

        The only possible strike I can think of against this book is that it might actually be a notch below the statistical expertise level of many SSC readers, but as such it could also be a valuable catching-up tool for others here.

      • cactus head says:

        For starting at the beginning: there’s All Of Statistics by Larry Wasserman. Fairly expensive though, unless you can find a pirated PDF. Amazon

        For machine learning/big data stuff: Elements of Statistical Learning by Trevor Hastie, Robert Tibshirani, and Jerome Friedman. This one’s not great as an introduction but it is available for free online.

        For a really practical-minded book on linear regression models (the kind that you learn in undergraduate math courses) leading to mixed-effects models, generalised linear models, and generalised linear mixed models, I suggest Data Analysis Using Regression and Multilevel/Hierarchical Models by Andrew Gelman and Jennifer Hill. It’s not up for free like ESL is, but it has a homepage with data and code.

  16. Do you think it is important for a moral framework, like utilitarianism, to have a sound answer to the question: “Why should a selfish, rational agent choose to behave morally?” Or do you think it is a Wrong Question? Or does not deserve answer for some other reason?

    I think this issue is relatively rarely considered (compared to arguing about what is and what is not moral) but is fundamental to the entire enterprise of ethics.

    I haven’t seen Scott or Effective Altruists addressing it. Thoughts or links?

    As for my opinion, I’m currently halfway through writing a morality sequence that considers origins of morality and its relation to rationality. On one hand, I feel the moral framework has to justify itself and give a strong reason. On the other, the problem has a vibe: “If you need to ask, you won’t get it anyways.”

    • Salem says:

      A moral system needs a justification, but it doesn’t have to be internal to the system.

    • skef says:

      Do you see a difference between your question and “Will a well-informed selfish, rational agent choose to behave morally? Explain.” And for that matter “Will a well-informed rational agent choose to behave morally? Explain.” (So: what role is “selfish” playing in the earlier questions? Shouldn’t the compatibility of selfishness and rationality be a question on this table?)

      Starting with “should” confuses the question because it’s ambiguous between ethical requirements (which I take it would be intrinsically begging your question) and function (which seems redundant with specifying “selfish” and “rational”, those presumably being the functions you’re interested in).

      • > Do you see a difference between your question and “Will a well-informed selfish, rational agent choose to behave morally?

        No, I don’t see any difference. I use “should” as in “Kasparov should have moved the pawn to win”. I agree I have put a lot of redundancy in my question. To me selfish is contained inside rational. You have deconstructed this question perfectly.

        Let’s take the distilled version then: “Will a well-informed rational agent choose to behave morally?

        How do you think this question ties to utilitarianism (edit: or your preferred framework)? Do you think the answer is “yes”?

        • skef says:

          My own suspicion is that value is external to rationality, and therefore it doesn’t contain selfishness. Trace back through reasoning and you’ll always find prior interests that aren’t entailed by it.

          Is it rational to destroy yourself? Sure, if that’s what you want.

          So my answer would be “No, but neither will a rational agent choose to behave selfishly. Rationality is a tool for, among other things, sorting through and deciding between competing interests.”

          • I think we understand selfishness slightly differently.

            I view selfishness as an aspect of decision-making algorithm. Contrast the two algos:

            An agent has to decide which action a to take.

            selfish algo:
            1: for each action a from A.
            2: predict state S(a) after taking action a.
            3: compute utility U(S(a)) according to agent’s fixed preferences.
            4: end loop
            5: select action that lead to highest U

            non-selfish algo:
            all the same except line:
            3: compute average utility U(S(a)) across all agents or across society or some other value implied by a moral system.

            Selfish agent’s preferences can include caring for other agents. They may include a wish of self-destruction.

            I might be using the word in a confusing way…

            Thanks for you response.

            Edit: What I’m trying to say is that I see selfishness as an aspect of decision-making algorithm rather aspect of agent’s values.

          • rlms says:

            What is the point of defining a separate class of non-selfish agents, when you could just consider them to be selfish agents with a utility function that takes others into account?

          • skef says:


            Suppose I said “I think we understand horses slightly differently”, and went on to describe a cow. In the philosophy business that’s called Humpty-Dumptying. You can stipulate non-standard uses of words, but it’s best to do so rarely, and up-front.

            By your definition someone who gives away all their belongings, and then dies to give away their organs, is being selfish if (as would presumably be the case) those acts align with their preferences. This is a species of argument that tries to make selfishness necessary, usually in preparation for smuggling in some assumption based on the more traditional understanding.

            The problem with that approach is that it doesn’t really make any distinction if the preferences are arbitrary. If someone prefers to follow a moral framework, they implement the second algorithm by way of the first. All you’ve really said is that to perform an act (or an act under a description) someone will have some pro-attitude towards that act that outweighs other attitudes. Well, yes.

            In ordinary use the concept selfish is less related to preferences than it is to interests. Both are a bit nebulous, but that’s life.

          • skef says:

            And, incidentally, by leaving preferences arbitrary rather than entailed by rationality, you’re making the point that there is no direction of value in rationality in a different way. Therefore the answer to any question of the form “Will a well-informed rational agent choose to ___” will be “not necessarily”, unless ___ is “act according to their preferences”, which is close to vacuous.

          • @rlms

            I think the following scenario captures a “selfish” agent that cares about some other agents.

            Imagine you are a parent of a child. Omega kidnaps you and one hundred other parents along with their children.

            Then Omega asks you: “I will kill almost all of the children. You have two options:
            1) your child survives with 100% certainty
            2) five randomly selected children survive”

            Which do you pick?

            I expect in such a situation most human parents would act “selfishly” saving only their own child. Because they would consider primarily their own preferences, rather than preferences of all the parents (as per algorithm).

          • @skef

            I’ll be back later, gotta go for now. Thanks!

          • @skef

            I plead guilty on usage of “selfishness”. This was not intentional. I attempt to clear this up and will try to find a better way of expressing my concept. Anyhow, I am not planning on doing the smuggling you’re talking about.

            I do think preferences and rationality are almost completely independent (I mean in general, not in humans).

            What sort of values do you think are entailed by rationality?

            To simplify the question further, I think there is value in asking “Why should a paperclip maximiser choose to act morally?”

          • skef says:

            Assume it’s acceptable to deduce what values a being has from how it or parts of it function (so sort of from it’s behavior, but with insight into its potential as opposed to actual behavior). So the being may be unaware of its values but still have them in this sense.

            On that assumption, a rational being will have some epistemic values. It seems hard to get rationality going without some tendency to form accurate representations, and revise them in (easily detectable) cases where they conflict.

            On the practical side, a rational being will have some of what I would call “technical” values, which is a super-set of instrumental rationality. If what you take as your means don’t make your ends more likely, you’re doing something not just wrong but specifically irrational — you’re not being sensitive to certain reasons.

            A lot of these values will just amount to one or another form of consistency.

            Beyond that, I doubt that other values are entailed by rationality.

          • What is the point of defining a separate class of non-selfish agents, when you could just consider them to be selfish agents with a utility function that takes others into account?

            Part of the answer is recognizing that utility has a different meaning in economics than in philosophy. In economics it describes people’s preferences, which determine their choices. In philosophy it means surplus of pleasure over pain, happiness, some not very precise concept along those lines.

            The two are related but need not always be the same; someone might choose to do something because he thought it was right even though he expected it to make him less happy.

          • rlms says:

            I interpreted LoopyBeliever as using utility in the economic sense, since using it just as a synonym for happiness doesn’t really make sense if you’re talking about agents and world states in an abstract way.

    • beleester says:

      I feel like, for most humans, this is the wrong question. We (usually) want to behave morally. We like getting the warm fuzzies for doing what’s right. Ideas of fairness and altruism seem to be hardcoded in children from a very young age.

      Indeed, I’d note that a lot of moral arguments take the form of “If you followed system of morality X, you’d get repulsive conclusion Y.” In other words, they’re attempts to codify our moral intuitions, or test how our moral intuitions handle unusual circumstances. It’s rare for someone to first invent a system that they call “moral” and only afterwards explain why they like it.

      Asking why we want morality at all is a useful question, but I don’t think it’s necessary to know why we want to act moral to study how we can act moral. Sorta like how economists can study the supply and demand curves for widgets without needing to know why people need to buy widgets in the first place.

    • Philosophisticat says:

      A few thoughts:

      -You’re smuggling in a view about rationality by throwing “selfish” in there (after all, perhaps rational agents aren’t selfish). I think few philosophers would be particularly worried by the thought that there wasn’t self-interested reason to be moral, as long as there were other reasons to be moral.

      -I think different versions of the worry have different force. For example, I think there’s more reason to want an explanation for why it would be rationally permissible to act morally, than an explanation for why it would be rationally required to act morally.

      -The debate over this tends to end up revolving around different views of reasons and rationality. Some people think that all reasons are based in some way on the agent’s existing preferences (or similar contingent psychological features), and some don’t. It’s a difficult debate, and I don’t know of a conclusive case either way. It sounds like you may simply be assuming the former view. A great many philosophers do not.

      -Perhaps you’re thinking only of the rationalist community, but the relationship between morality and rationality is not rarely considered among philosophers. It is high on the list of most-discussed-ever topics in metaethics. It’s a pretty tough topic for nonphilosophers to have worthwhile thoughts about, so it’s not too surprising that rationalists in general wouldn’t have much to say about it. (Though to be fair, most rationalists are not in a position to make competent judgments about comprehensive moral theory with remotely any confidence either, and that doesn’t necessarily stop them).

      -I gather a significant number of rationalists are noncognitivists, or in any case have some more nebulously non-objectivist metaethical stance. This kind of worry looks a lot different when you don’t think there are objective moral facts at all.

      -People who are worried about this in the way you seem to be tend to be attracted to contractarian accounts of morality.

      • Thank you! Great deconstruction.

        1. I seem to define “selfishness” in a weird way.

        3. I struggle to think what else could they be based on. Could you give an example?

        4. To be fair I do have an aversion to philosophy that is not strongly rooted is some science (physics, evolution, game theory, ai). That is a definite skew towards rationalists.

        5. and 6. I am a noncongnitivist and contractarianism approximates my moral position best.

        Pretty decent inferences on your part.

        • Philosophisticat says:


          There are lots of different views about what we have reason to do that aren’t limited to our present desires. Some people think we have reason to do things that are pleasurable, and to avoid doing things that are painful, regardless of what we want. Some people think we have reasons to take into account the things we will desire in the future, regardless of whether we desire them now. Some people take a cue from epistemology, where we typically think we have reason to believe what the evidence supports, regardless of whether that satisfies any of our desires. And some people think we have reason to do what is morally right, regardless of whether that satisfies any of our desires. And on the other end, many people think there are some desires, or patterns of desires, that are irrational – for example, being entirely indifferent to pains or pleasures that occur on future Tuesdays. There are many different grounds for these views, but best fit with our ordinary judgments about peoples’ reasons is one of them. I think it’s actually quite hard to give a good argument for why our reasons must be based on existing desires. I used to share your “of course, how could it be otherwise?” thought on this but I’ve come to find the support for the view pretty thin.

          For the most sophisticated modern defense of the view that all our reasons are grounded in desires (the Humean view, it is often called), I recommend Mark Schroeder’s work, particularly his book Slaves of the Passions.

          • I see. Well, I agree there many forces pulling us in different directions in decision-making. I tend to call the net force “preference” and sort of sweep the complexity under the carpet. In AI you typically have one simple utility function that guides the actions. I fully appreciate that whatever mechanism evolution gave us is not so neat and well-structured.

            Again thanks. Will have a look.

          • Philosophisticat says:

            The views I mentioned aren’t descriptive views about what in fact moves people – if you have a broad view of preferences, then sure, people’s actual behavior can be explained by preferences – they’re normative views about what kind of behavior is rational.

    • rlms says:

      The question of whether morality is meaningful (whether moral claims are true, or even can be true etc.) is pretty much the entire field of meta-ethics. It is on a different level to Effective Altruism, which presupposes that moral claims matter, and generally presupposes that utilitarianism is the correct moral framework.

      Personally, I think the point of moral frameworks is to let people argue about moral issues where they have clashing intuitions, by choosing frameworks that agree with their shared intuitions and seeing what they say. I don’t think moral claims can be objectively, eternally true. Perhaps you could create a moral framework by arguing game-theoretically from first principles, but I doubt it would end up having much to do with the “interesting” parts of human morality, which seem to be arbitrarily caused by biology.

      • I agree with your assessment of EA.

        Arguing from game theoretic principles and considering evolutionary utility of morality is the way I approach reasoning about morality. Not too sure what you mean by “interesting parts of human morality”. Do you mean altruism? Or helping strangers who never will be able to repay you? I don’t think they are arbitrary at all.

        I think this line of reasoning is taking me somewhere promising. But it’s easy to delude oneself in philosophy.

        • rlms says:

          By “interesting parts” I mean things like abortion, animal rights, the death penalty etc. I can’t see how you could get from “consider an agent with utility function U in a world consisting of similar agents” to “abortion is only OK in the first trimester” or similar. I agree that considering the evolution of moral intuitions could be useful.

          • Well, obviously you’re not going to get to something as specific as details of abortion.

            But I feel I have an explanation that takes me to why would agents evolve to care about other weak defenceless agents like animals.

            The explanation involves evolution of such altruistic preferences as a method for pre-commitment in games. E.g. I may be more willing to think you won’t backstab me if I can see you genuinely care about weak. Thus, cooperation even on one shot prisoner’s dilemma is possible. It has been argued that displays of emotions have evolved as a mean for honest signalling and pre-commitment.

            The reasoning chain is rather long but I have it roughly complete. I’m about halfway through writing it down at 8k words. If this sounds interesting to you, come by and drop a comment wherever you think I’m missing something. I can see already see from our chat that I should rephrase what I mean by “selfishness”. Thanks!

        • skef says:

          Arguing for the evolutionary utility of morality is only effective if evolutionary utility has some form of value, otherwise you’ve just dissolved one phenomenon into another. If the goal is just getting rid of “oughts” in favor of “ises”, you can just go the evolutionary psychology route. These rearrangements don’t typically result in reasons for acting. “I need to look for a good job as part of my project to be more evolutionarily successful.” “Hmm … OK.”

          Why do you think evolutionary utility matters? If we all got clobbered by an asteroid next week, it would be a bummer for us if we found out, but that’s just the being-a-product-of-evolution talking. Of course we’d think that.

          From my admittedly limited impression from your posts today, you seem to want to find a place where you can get game-theoretic principles and evolutionary utility going as oughts, rather than just ises. But you also seem to want that place to be where almost every other preference is at least arbitrary. Can you explain your reasoning about that place? Why should I care about or act in accordance with evolutionary success, even given that someone might explain my actions in terms of evolutionary success from a third-person standpoint?

          • I only consider evolutionary utility of morality because I want to understand the intuitions that implement morality. I don’t expect to find “oughts” but I expect to find why people care so much about arguing “oughts”.

            I don’t care about increasing my “evolutionary success”.

            I think contractarianism captures my position best. I think it would be nice if everyone agreed to follow utilitarianism, but I am slightly too cynical to believe no one will defect. Why we see “moral progress”? What are its limitations? Whether people usually genuinely care about others or do they self-deceive most of the time?

            When do people adhere to contracts and when do they fall apart?

            What structure is required in the future so that we have a stable society even when majority of power is in hands of a few?

            More mundane stuff: How to be more trustworthy so I can improve my relationships? Should I trust people more than I do now?

            (e.g. people are often hesitant to ally with me during board games because they perceive me a treacherous Machiavellan — I sort of like this image of myself but truly I’m a very cooperative non-manipulative honest person (said a every scheming Machiavellian))

            Side note: Maybe from the amount you have seen it was ok to assume what you assumed, but I don’t think I have ever read anyone who would actually argue for maximising his evolutionary success.

          • skef says:


            I think I’m trying to probe a different question. Why do you have a position on oughts at all? Once you arrive at the conclusion that what you want/think is important/has value is just the product of an evolutionary process with no further meaning, isn’t the rational conclusion that it’s all a bunch of BS? Is there any argument against, say, wireheading possible from that standpoint? That’ll fix up your preferences right good.

            I can see a role for contractarianism along these lines: “Look, evolution has stuck us with these tendencies, and there’s enough commonality in the distribution of these tendencies to support some stable equilibria, so it makes sense to go with some of those.” But a) how compatible with this attitude is writing a sequence about it? And b) isn’t trying to figure out what tendencies you might tinker with after all at least as rational a response, and maybe a more rational response?

          • Evopsych tells you why humsns have ethics and what ethics is for. Contractarianism tells you, up to a point, why you should follow the ethics of your society. Neither tell you what ultimate value is, and thats a feature not a bug, since it allows ethical progress and innovation to be accounted for.

    • Dr Dealgood says:

      If you take a step back, a lot of non-Abrahamic moral systems explicitly do this.

      (Warning: lay interpretation of Nietzsche follows.)

      The distinction between master and slave morality is, in large part, about what the proper function of morality is. To the Greeks and Romans, virtue was that which increases one’s own excellence and power* and the reward for a moral life was happiness**. To the Jews Christians and Muslims, righteousness is submission to divine will and seeking personal gain is inherently unrighteous.

      The Death of God hasn’t yet hit home enough to shake the majority out of that mindset of submission and self-sacrifice, but it has forced them to scramble for another target of their devotion. This search will prove fruitless as the role of God cannot be so easily replaced. The foundation of slave morality is gone and we must choose between nihilistic hedonism and self-overcoming***.

      I’d strongly suggest that you at least read ‘On the Genealogy of Morals‘ before writing your sequence. It is very relevant to the questions you’re examining.

      *The Greek and Roman words for virtue, arete and virtus respectively.
      **Eudaimonia, also sometimes translated as welfare or flourishing.
      ***The Last Man and the Superman, in Nietzsche’s terms.

      • Thanks! Ordered the book.

      • Protagoras says:

        I consider myself something of a Nietzsche scholar, and this post at least does not seem to horribly misrepresent him in any important way. Ignores lots of related issues that are relevant, but considerations of space are a legit excuse. Not to encourage you to engage in more lay interpretation of Nietzsche (you’re bound to go astray eventually in that minefield), but at least this time you seem to be OK.

  17. onyomi says:

    Democrats continue to rediscover federalism now that they’re not in power (of course they would not have been suggesting that red states should be free to go their own way had they just recently won the presidency and Senate). Increased federalism? Abolishing the income tax?! In the New Republic!! Please don’t throw us in that briar patch!

    • HeelBearCub says:

      I mean, he’s just wrong.

      Really he is just indulging in this fantasy where he gets to see red states harm themselves and somehow that has no bad consequences for blue states.

      I mean, China’s pollution is affecting L.A.’s air quality. Global sea level rise will affect blue states the most. It doesn’t work the way he fantasizes it.

      • caethan says:

        The financial nonsense annoys me too. For example, take New Mexico from the chart he cites in the Economist. Says that New Mexico got $317b from the feds from 1990-2009 and sent $116b in taxes over the same time period. Clearly New Mexico is a goddamn moocher, right?

        Take a look here: 1997 breakdown of federal spending by state, from the census bureau.

        New Mexico has a hugely disproportionate amount of federal monies going into direct salaries and procurement instead of individual payments (i.e. welfare) and state/local grants. Why? Los Alamos is in New Mexico, all the procurement money is going there to support it. Sure the locals get some benefit from Los Alamos, but most of the money is direct costs to running a big nuclear research center, not handouts to New Mexicans. 32% of federal spending in New Mexico (in 1997) is DoD or DoE salaries or procurement!

        You could do a much better job of this kind of thing – I’d start with per-capita spending on direct payments (Medicare, SS, SNAP, etc.) and compare with per-capita taxes. (All there in the census/IRS results.) But the naive “lol, red-states get more money from the feds than they give back” is just stupid.

        Which is not even to get into the point that maybe we should be helping our poor citizens wherever they live, even if they (or their neighbors!) disagree with our preferred politics. E.g., Mississippi, which always gets trotted out in these discussions, has high federal intake in large part because of a big population of poor rural black folks. But no, let’s cut off welfare payments to poor black Mississippians because their white neighbors reliably vote Republican.

      • Sivaas says:

        That should resolve itself pretty quickly though, no? Part of the Democratic platform is open borders, so plenty of poor Mississippians who think they’ll get a better deal from the Blue States (and would have voted that way anyway) will be headed there in pretty short order, and I’m sure the Red Staters won’t mind seeing them go.

        • shakeddown says:

          Part of the Democratic platform is open borders

          That’s definitely outside the mainstream. Unless you mean between states, when I’m pretty sure is also in the republican platform.

          • Sivaas says:

            I think “open borders” was the term I needed to be more specific with, I think that was hyperbolic when “extremely open to refugees/immigration” would have sufficed. Maybe even that’s too much, but wasn’t that one of the major conflict points of the election?

            I recall “we should take in anyone who wants to come here” being a fairly common line, often quoting the poem from the Statue of Liberty.

            EDIT: I also want to point out that I mistakenly thought we were at maximum nesting depth, my original comment is a reply to caethan rather than HBC.

          • shakeddown says:

            But I still don’t understand your objection. Caethan’s example was black people in Mississippi, and you’re saying that the democratic platform is okay with people moving between states. Black people in Mississippi are generally American citizens, so where does immigration come into any of this?

          • JayT says:

            I think he’s saying that the Democrats like to point to states like Mississippi to show that Republican states take more federal funding than they give, but if you were to actually try and split the blue and red states up into two different countries, a lot of the people that cause states like Mississippi to require a lot of federal funds would want to live in the “blue country”. So if the blue country were to actually lean towards open boarders, they would end up with a lot of the expensive people from the red country.

          • Sivaas says:

            Yes, JayT explained my point more accurately than I could.

            In the hypothetical where this Blue State- Red State split happened and it led to bad results for poor people in Red States, those people are going to look at the Blue States, which have better support for the poor and political leanings that the poor people aren’t particularly opposed to, and give some serious thought to moving there.

            It’s possible that the Blue Staters reject Red State outcasts, but I think there’s going to be a strong movement to accept them, because “accept refugees” is a pretty core part of Democratic policy (at least right now, in contrast to Trump), and many of these people will be minorities that the Democratic Party is very interested in supporting.

            This will weaken the “blue state utopia, red state slum” idea that the OP link is expecting, as a large portion of those costs will simply migrate to the Blue States. It also means the Blue States won’t end up abandoning Red State minority poor, because they won’t be Red State anymore.

            Maybe the transition costs will be too high for this to be universal, but the Blue States are going to have public support for getting the refugees in (at least initially when their political situation is the same as it is now) and may try to help with that. The Red States don’t have any incentive to keep them Red either.

          • Deiseach says:

            What interests me greatly as a question to be solved, pace that Bluexit thinkpiece, is the question of migrant labour and agriculture in the new West Coast Blue Enclave. California has a huge agricultural sector (though it only accounts for 2% or so of the state economy) and it is admitted that this is dependent on immigrant labour, which is illegal/undocumented immigrants, because that is cheap labour.

            Now, the Blue Nation can probably absorb a lot of that produce to feed itself even if it doesn’t export it (I imagine the writer would like to gloat, had he thought about it, that Red State USA depends heavily on California for year-round availability of what would otherwise be seasonal produce), but what I want to know is this:

            The writer was gloating about the Blue State Utopia where everyone would have UBI and it would be a paradise of innovation, smart rich people, $15 an hour minimum wage at least, free college, free health care, wealth creation, equality and diversity with the more immigrants, the merrier. So if they’re going to make the agricultural sector pay the going rate plus conditions to their workers, are they willing to absorb the increase in food prices that will entail? Or are they going to maintain the winking at the welcoming of immigrant labour because it’s cheap labour, and as long as we’re not talking about Indian PhDs coming over to work in Silicon Valley but Carlos and Juan harvesting crops and working on building sites, well, we’ll turn a blind eye to it – will they encourage seasonal labour to return to their countries of origin when work is scarce, or will they make them legal and let them live year-round? And if they’re legal, then this opens up other opportunities than manual labour to them, so there’s the chance they’ll go for better-paying jobs, which again means that the requirement for low-paid seasonal labour depends on migrants who have no better opportunities:

            On the demand side California farms rely on access to a ready supply of labor both for full time and seasonal employment. California is a large producer of labor-intensive crops such as fruits and vegetables. The comparatively low wages of farm work does not draw sufficient labor from the industrial and service sectors of California’s urban centers. The state’s agricultural sector therefore depends on low wage immigrant labor.

            Under the federal Fair Labor Standards Act (FLSA) are exempt from overtime pay provisions, and young workers with parental consent are also exempt from the child labor provisions (USDOL 2008). The Migrant and Seasonal Agricultural Worker Protection Act provides the right to timely remuneration with earnings statements and safe
            transportation during work hours, but does not provide guarantees available to other workers under federal law. However, the California Agricultural Labor Relations Act (CALRA) established the right for collective bargaining for farm workers in California. CALRA is administered by the Agricultural Labor Relations Board, which investigates claims of unfair labor practices. In addition, California farm workers must be paid the California minimum wage, significantly higher than the Federal minimum wage rate.

            Never mind the legal niceties of “California farm workers must be paid the California minimum wage”; farms (which are large agri-business concern) don’t directly hire the workers but rely on sub-contractors who provide work crews, and if you’re an undocumented migrant who has signed up with one of these companies, how much can you insist that they are obliged to pay you the legal minimum wage in the first place, even if they don’t decide to charge the farm the legal wage (so the paperwork holds up) then keep a slice of that and pass the remainder on to you?

            It’s all very well for the author to imagine Blue Utopia of all the colours of the rainbow, but he hasn’t given any thought to “so how are you going to pay for all this?” (apart from “our rich guys will invest in Wall Street”) and “are you tackling the existing problems of not treating the migrant labour the same as native labour?”

            I laughed out loud at his “high speed trains linking islands of Blue stranded in Red territory to the east and west coast Blue States” because hey, those trains have to pass over Red territory, and if the Red states decide to refuse you planning permission to build those rail tracks, or they decide that yes, you can build and run your railways but you have to pay a toll to each state you pass through – and that’ll be 1000% the price of each train ticket – what can they do about it? Not to mention who is going to maintain the train tracks etc. and protect them from sabotage – there’s a lot of way disgruntled Reds can mess up the perfect Blue systems and if Red state governments are so inclined, there’s “oh we’d love to help but dash it, the CCTV systems were all down for some reason” or “we can’t identify them because the cunning villains all wore black balaclavas” or the likes.

          • Nancy Lebovitz says:

            Silicon Valley figures out how to automate farm labor.

          • The Nybbler says:

            Not all technology comes out of Silicon Valley. For example For another example, the GPS-receiver company Garmin. I don’t know that Silicon Valley has any particular advantage in automating the parts of farm labor which haven’t been automated; one common criticism of SV tech is it’s all about making the next app to get the advertising money, not doing anything practical outside the digital world. There’s a lot of truth to this, I think, and the big exceptions tend to be medical rather than agricultural (because SV execs want to live forever)

    • Deiseach says:

      Red states are nearly twice as dependent on the federal government as blue states.

      Government is California’s largest industry, like most states, with about 2.5 million employees. The second largest industry, according to the Census, is Healthcare and Social Assistance.

      Go ahead, end your federal Amtrak subsidies. In their place, we will build fantastic, new high-speed rail systems of our own. They’ll run past our state-of-the-art wind farms, fiber-optic networks, and highways that recharge our self-driving cars as we travel.

      See kids, this is what happens when you legalise marijuana. Just Say No! Will they build monorails? Oh please, please tell me they’re going to build monorails! 😀

  18. Scott Alexander says:

    In the PCT thread, Null Hypothesis describes a theory of obesity.

    If I understand it right, it’s that in the short-term, everything is calories-in/calories-out. In the long-term, exposure to too many insulin spikes builds insulin resistance and metabolic syndrome, which causes fat cells to to be less willing to release fat in response to starvation, which causes people to be hungrier after shorter periods without food, which causes them to eat more and get fatter. Eventually a new equilibrium is reached where maybe someone has twice as many fat cells, each of which is half as willing to release fat, and so their body has enough energy.

    This seems really good to me. It combines the most important insight of the Gary Taubes crowd (primitive populations are much healthier than moderns and it seems like it has to involve sugar and insulin resistance in some way) with the most important counterargument of the mainstream (low-carb diet doesn’t really lose weight better than any other diet, and there’s no advantage to eating less sugar in the measurable short-term). The missing synthesis is that sugar/insulin only matters on the scale of years and so wouldn’t show up in a lot of the sort of experiments that have been done.

    Now that I hear it said this way, it seems really obvious and I’m wondering if everyone else knows this and I was just somehow completely missing it before.

    • Jaskologist says:

      Ace of Spades recently had a post about that, so I guess the idea is out there.

    • onyomi says:

      If this is right, why do Japanese people, who eat white rice two or three times a day, not gradually get really fat as they get old? Does white rice not spike insulin?

      • dndnrsn says:

        The explanation I’ve seen is that they’re eating some protein and fat with it, which blunts the insulin release, and they’re not eating huge portions.

        • onyomi says:

          Have you seen the portions at a Japanese ramen store?

          I could believe, however, that the fact that they usually eat it with some other food probably helps by slowing down digestion.

          Overall, however, in order for the “insulin resistance as primary cause of obesity” thesis to be true, I think it would have to be the case (and is the case, I have read somewhere, though not sure the source is reliable, as I am unsure of all sources on nutrition) that meat and fat increase insulin overall more than we think, and carbs maybe not as much as we think.

          My overwhelming sense of human dietary history is roughly:

          dawn of time-10,000 years ago: everyone hunts and gather, emphasis on the gather; eat a fair amount of meat, but also a lot of calories from e. g. fruit and nuts; fat and protein consumption make up a relatively high proportion of calories, but everyone gets a lot of exercise and food is scarce, so no one is fat, except maybe the tribal chief.

          10,000 years ago-1900: almost everyone except the few remaining pockets of hunter-gatherers farms, gets vast majority of calories from grain and, to a lesser extent, legumes, fruits, and vegetables. Animal husbandry for food is very expensive and king dominates hunting land; only king gets to eat large amounts of meat frequently; only king is fat and has diabetes. Overall, these people are smaller, weaker, and lighter than the hunter-gatherers, though still probably surprisingly strong for their size due to all that farm labor (but even the less wealthy city dwellers get most of their calories from grain and are very small and thin by today’s standards and even by hunter-gatherer standards).

          Now: everyone eats like a medieval king; everyone is fat.

          Is this picture at all wrong? If not, then it implies that the only way people get fat is by eating a lot of meat and being sedentary, not by eating a lot of carbs. Moreover, the farmers are, if anything, smaller than the hunter-gatherers; the hunter-gatherers are prevented from getting fat on their diet of bear fat and pemmican by the fact of having so little food, overall, and being extremely active.

          • Douglas Knight says:

            The king was fat, but did he have diabetes? History should answer this, but I don’t know the answer.*

            After the war American meat consumption went steadily up until 1980. In 1980, everyone freaked out about meat and switched from beef to chicken, without changing the total. Since that day meat consumption has been steady, but carbs and total calories have kept going up. Obesity seems to have have started going up around 1970, close to but not quite the same as 1980.

            * Henry VIII had diabetes and probably had a low-carb diet. I suspect retired football players would be a good comparison. Especially since 1980 (steroids?), they gain tremendous weight, on a diet that I’d guess is much lower in carbs than the usual American diet. After they retire, it all turns to fat.

            Here is someone claiming that it was an aristocratic disease.

          • Nornagest says:

            dawn of time-10,000 years ago: everyone hunts and gather, emphasis on the gather; eat a fair amount of meat, but also a lot of calories from e. g. fruit and nuts; fat and protein consumption make up a relatively high proportion of calories, but everyone gets a lot of exercise and food is scarce, so no one is fat, except maybe the tribal chief.

            If recent hunter-gatherers are anything to go by (they aren’t representative, but they’re probably illustrative of the range of strategies), it is dangerous to make this kind of generalization. There is a ton of variation in where they get their calories and how they get them — some tribes get up to about two-thirds of their calories from meat (Inuit almost 100%, but they’re an outlier), some one-third or less, though I’ve never heard of a hunter-gatherer culture that was vegetarian or nearly so. Some eat lots of fish, some none. Some eat insects, some don’t. And so forth. Things on the vegetable side are similarly varied.

            If there’s anything consistent, it’s that hunter-gatherer diets are more diverse than agriculturalist — though even this is only true in the long run. You’ll get individual bands e.g. eating a particular species of berry almost exclusively when it’s ripe, then moving on to something else when that gets exhausted.

          • onyomi says:


            Yes, I think there was, and is greater variability in the hunter-gatherer diet than the agriculturist diet, both among and within groups. And, overall, the hunter-gatherer lifestyle was probably healthier and produced bigger, stronger people than the poor agriculturist diet/lifestyle (assuming you don’t get gored by a wild boar). But it’s also unable to support a large, non-nomadic population, and might be more likely than rice and beans to induce fatness and related illness if you take away the “tons of exercise and regular periods of scarcity” parts.

          • dndnrsn says:


            Have you seen the portions at a Japanese ramen store?

            A Japanese ramen place in Japan, or a North American “Japanese” (where I am, most Japanese restaurants outside of the high-end are operated by Koreans)?

            As for your thesis in general:
            This chart only goes back to 1960, but it looks like the dramatic increase in US obesity among adults and overweightness and obesity among kids dates to 1976-1980 forward. I doubt there was a similar spike earlier in the century – I don’t see anyone talking about it, at least.

            My explanation would be that there was a change in food culture in the 1970s. More snack foods became available (Hostess chips used to have 3 flavours, for example, and that changed in the 1970s), portion sizes got bigger (if you went to a McDonald’s back in the day, you got one choice of fries size, smaller than a “small” today) it became more acceptable to eat whenever (I’m told once it was considered rude to eat on public transportation), etc. It is much, much easier to overeat consistently and put on lots of excess fat with omnipresent processed snack foods than with having extra servings of mashed potatoes too often.

            @Douglas Knight:

            Source on Henry VIII? Googling suggests he ate a diet that was heavy in carbs including refined sugars. Your comparison to a retired football player is a good one – by some accounts, he only got super fat after he was injured in a jousting accident and couldn’t exercise any more.

          • The original Mr. X says:

            the hunter-gatherers are prevented from getting fat on their diet of bear fat and pemmican by the fact of having so little food, overall, and being extremely active.

            Aren’t hunter-gatherers generally less active than farmers? I recall reading an account by somebody who visited hunter-gatherers in southern Africa, and reported that they all considered farmers to be a bunch of dopes for following a lifestyle which required them to do more work to feed themselves than the hunter-gatherers put in.

          • onyomi says:

            “Aren’t hunter-gatherers generally less active than farmers?”

            That is an interesting point I hadn’t considered. I assumed hunting and gathering consumed more calories than farming, in part because you don’t get to store and preserve stuff as much (farmers work really hard during planting and harvest but also have a lot of time off, during which time they live off stored-up grain, pickles, etc.).

            As in other things, the hunter-gatherers are probably more variable. If you live near the equator in an area where the gathering is fairly plentiful, maybe you have to do less work, overall. If you have to follow buffalo and whales around their natural range, maybe less so.

          • onyomi says:


            Especially at a Japanese ramen store in Japan.

            And if you order the “big size” (oomori), the bigness usually comes largely in the form of giving you twice as many noodles, not twice as much pork, toppings, etc.

            Regarding changes in habits: I do think most of the rise in obesity comes from taking in more calories overall, not to changes in macronutrient ratios (though arguably different macronutritent ratios can affect satiety differently and lead to taking in more or less calories).

            Most notably for me: I asked my parents, who grew up in the 50s, how eating had changed since they were children: they said food when they were kids was more expensive, less varied, less convenient, and more likely to be home-cooked (grandma’s apple pie, even if it contained even more fat than McDonald’s apple pie, was not available 24/7, for 75 cents at the drive-thru). These factors alone seem enough to make a big difference without appeal to changing macronutrient ratios and/or prevailing dietary wisdom.

          • dndnrsn says:


            Ramen: how often do people eat that much?

            Regarding the rest of your post, that sounds about right. I also assign some credence to the idea that modern processed foods are hyperpalatable, or a superstimulus, or something in that vein. There’s something about modern processed food that makes it eminently binge-able. Even, weirdly, if it doesn’t taste as good. My mother makes cookies that are better-tasting (more interesting) than the stuff you buy in a box at the supermarket – but something (texture?) means I can resist hoovering down a couple dozen of them in a sitting. The same is true of baked goods bought from a quality bakery. There’s just something about the mass produced stuff that cripples my self-control if I have any of it – so I just avoid it.

          • onyomi says:

            Re. Ramen

            Pretty often. Large quantities of rice: pretty much every day.

            Re. hyperpalatability:

            Yes, I think today’s foods are designed to be addictive by hitting all the salt-fat-sugar pleasure buttons without necessarily being very interesting, which would be expensive. I, personally, don’t find it any easier to exercise restraint with homemade apple pie than store bought (rather the opposite); the key difference for me is that the former isn’t available all the time unless I spend all my time baking.

            The history of cooking has been finding new and interesting ways first to preserve food and then to concentrate and intensify its pleasure-inducing flavors, basically by making it easier to digest large quantities of fat, protein, and carb. The history of the past 60 years or so of food has been figuring out how to get some version of that calorically-dense, hyperpalatable food available really cheap and fast… which is sort of the death blow.

            People can deal with unconcentrated cocaine (coca leaf); they can probably deal with concentrated cocaine if it’s only rarely available; most can’t deal with (use in moderation) plentiful, cheap concentrated cocaine (except by abstaining entirely).

          • dndnrsn says:

            You’re right that home-cooked stuff is usually less available. My mother’s cookies are dependent on my mother making cookies; supermarket cookies are always there for a few bucks a box. I think there’s also a contextual difference. The one time of the year she makes a really huge amount – around Christmas/New Year’s – they’re present in a social environment. There’s more of a brake on consumption when it’s a platter of cookies after dinner with guests around.

          • Douglas Knight says:

            Googling suggests he ate a diet that was heavy in carbs including refined sugars.

            Googling says that his courses were all named either meat or dessert. But bread was a side dish and it is hard to determine how much he ate.

            What is your standard for “heavy”? That he ate dessert at all? I gave my standard when I talked about the football players: I’m pretty sure that his macronutrient ratios were tilted more towards protein than Americans today, or even Americans in 1980. In absolute terms, he probably ate more carbs than Americans in 1980.

          • dndnrsn says:

            I’d define “heavy” by absolute amount. Ratios get screwy – someone who eats a diet that’s 20% carbs isn’t eating a “high-carb diet” but if they’re eating 6k calories a day, that’s 300g carbs, and I’d say that’s pretty heavy carb consumption.

        • Mr Mind says:

          Which is really nonsense, since proteins spike insulin more than carbs.

          • dndnrsn says:

            More per gram, but digested more slowly, seems to be what I’m seeing, and people tend to eat more carbs at a sitting than protein. A moderate (4oz) portion of lean meat is usually ~20g protein plus a little fat, while a moderate (1 cup) serving of cooked white rice is ~45g carbs and ~4g protein.

      • Deiseach says:

        Singapore is having problems with obesity and diabetes, is this because they have a more Westernised diet or because they are consuming more of the modern sugary/refined carbs food and drinks?

        • Tibor says:

          Judging solely on the hawker stands and what they sell (there are a couple of places around Singapore with several dozens of hawker stands offering a variety of foods, each stand usually only offers one or two things but usually also different things than other stands), their diet is not very western. In fact, the one “western food” hawker stand they had there made me laugh. They had something like a schnitzel but cut to pieces so you can eat it with chopsticks and served together with eggs I think . I bet the Chinese laugh at “chinese” bistros in Europe as well though 🙂 I haven’t tried the “western food”, so I can’t tell what it tastes like. I am going to Singapore again this summer, so I might give it a try.

          However, one can get strange “western food” in Europe as well, apparently. In one Italian restaurant in France, they served me a schnitzel with spaghetti, which was really strange…unfrotunately, they didn’t speak English and I don’t speak French, so I just picked almost randomly, using my very imperfect Spanish to try to guess the meaning of the French words on the menu. Later, I learned that this is a legitimate type of Schnitzel – Milanesa is what they called it so I suppose that this is how they eat it in Milan? But I am still not convinced about the spaghetti. The waitress actually asked me what I wanted as a side dish (that much I understood, she also spoke some poor Spanish like me) and since I didn’t know what Milanesa was, I decided to have it with half of spaghetti and half of something that ended up being more sensible (I think it was potatoes).

          • dndnrsn says:

            Sounds like what you had would be called veal parmigiana.

            The dish consists of a sliced filling pan fried in oil, layered with tomato sauce and cheese, and baked in an oven. In some kinds, the sliced filling is first dipped in beaten eggs and covered in flour or breadcrumbs before frying. Some recipes use hard grated cheeses such as Parmesan or Pecorino Romano, while others use softer melting cheeses like Mozzarella or Caciocavallo, or a combination of these.

            In the United States and Canada, veal parmigiana or chicken parmigiana is commonly served as a grinder or a submarine sandwich. It is also popular with a side of or on top of pasta. Diced onions or green bell peppers, sautéed or raw, are sometimes added.

            Usually menus call it “parmesan”, I suppose because parmagiana is harder to say – the cheese is almost always mozzarella.

          • Tibor says:

            @dndnrsn: Nope, it was a Milanesa. Basically a romance language name for a schnitzel, but I didn’t know that. And I still doubt they actually serve it with spaghetti in Milan 🙂 It just does not work together.

    • Douglas Knight says:

      This seems to predict that low-carb diets are easy and effective. But that just isn’t true.

      Of course diabetes correlates with obesity and that’s why we have this concept of “metabolic disorder.” But the correlation is very weak.

      Is the theory
      insulin spikes → subclinical insulin resistance → specific claims about fat cells → hunger → weight gain?

      First of all, lab animals are getting fat and they aren’t subject to insulin spikes. Pet cats are getting fat and they don’t like sugar.

      Subclinical insulin resistance is widely measured. If it was a good predictor of obesity, someone would have noticed. I guess this theory is that insulin resistance predicts the derivative of weight, not current weight, but I still think someone would have noticed. It is true that there are several tests of insulin resistance and the common GTT is considered inferior to the glucose clamp. I don’t know why it is considered inferior, but maybe it fails to distinguish exactly the subclinical differences we are talking about.

      Another way that insulin resistance could lead to hunger without having producing an obvious correlation with obesity is that another control system, call it “willpower,” intervenes. People get the same level of hunger signals, but different people respond by eating or accepting hunger, resulting in differential weight gain.

      Anyhow, you should break down long chains of causes into individual causes and study them. The specific claims about fat cells sound so specific that they should be easy to study. This sounds like the kind of thing that bro science has specific opinions about that medical science doesn’t. But it is exactly the portion of bro science that leads to claims about low carb diets that are false.

      Added: Jaskologist’s link to Ace of Spades has an interesting graph of insulin as a function of time. It claims that in obese people overnight fasting insulin gets back to zero, but not between meals. This is a good step of breaking down the complicated node and introducing new metrics.

    • dndnrsn says:

      The rise in obesity and accompanying problems is fairly recent – if you go back and look at high school and university graduation photos from just a few decades ago, it is shocking how thin everyone is. Even the one or two fat people you see are not especially fat by the standards of today. So I don’t think it makes much sense to say that “moderns” are less healthy in terms of stuff like heart disease, diabetes, than hunter-gatherers. Someone from 1917 is a “modern” same as someone today.

      Clearly something or some combination of things started changing in the middle of the 20th century.

      Personally I have had the greatest sustained fat loss from bro science: low carb, close to zero starches, high protein, high omega-3s, saturated fat not something to worry about. My (bro scientific) theory, based entirely on personal experience, is that successful dieting is 90% psychology.

      • Douglas Knight says:

        Your last two sentences don’t seem very compatible. Do you or anyone else advocate omega-3s for fat loss?

        • dndnrsn says:

          There’s significant overlap between “recommends high protein and low carb” and “recommends high omega-3s”. Sometimes there’s some “mumble mumble inflammation mumble mumble” relationship between omega-3s and fat loss going on. Sometimes the omega-3s are to help your body deal with the weightlifting you’re doing to hold on to the muscle while you lose fat. Like I said, bro science.

          To explain further what I meant: the best diet is one you can stick to. I know a guy who can eat whatever and fit it into calorie counting. I can’t do that: give me 100cal of Doritos and I will want 1000cal of Doritos – I’m not good at moderation; I can do 0 beers or 12 beers but not 3. There are people who succeed with low-fat, high-carb diets and jogging but I can’t stand that.

          • Douglas Knight says:

            Oh, OK, I thought “90% psychology” meant placebo effect, or something, that people had varying psychologies and thus needed varying diets. But it sounds like you’re saying that low carb is (pretty much) universally easier, but some people have more of the relevant willpower, or haven’t developed bad habits around carbs, so they don’t need to bother with it.

          • cassander says:

            I wouldn’t overstate understate the benefits of sheer calorie restriction. My dad has spent the last couple decades slowly building up weight over a few years, getting serious about atkins for about 6 months, losing the weight, then repeating. Whatever you think the benefits of Ketosis are, it’s undeniable that he eats fewer calories when he’s on it. If he orders a burger, he gets it without a bun. A steak, no potatoes. And sure, some of those calories get substituted, but you invariably get into places where you don’t or can’t substitute, and so consume less. It’s a weak form of portion control and I don’t doubt that its responsible for a lot of his weight loss.

          • dndnrsn says:

            I don’t know if it’s universally easier, but I find that it certainly is for me. Perhaps there are some people who eat a slice of cheese and go wild, or a handful of almonds. But the things that wreck my diet are usually slamming an entire box of cookies into my face, or something like that. For me effective dieting is largely about structuring things so that doesn’t happen, or if it does happen, is clearly delineated from “ordinary eating”.

            Something that suggests that maybe not eating carbs are hard for people in general is that Nutrition Action (which is the newsletter of the Center for Science in the Public Interest) will periodically do one of their “saturated-fats-are-bad-you-guys” articles and say “to reduce fat, instead of this [skin-on chicken breast with a baked potato] eat this [skin-off chicken breast with a salad]” and while they acknowledge the different macros they don’t really acknowledge that the switch is not just about reducing the fat. This suggests to me that not only is overeating carbs a common problem, it’s enough of a problem that it can’t even be said openly. (Alternatively, the CSPI are just really really on the saturated-fat-is-bad train and don’t want to acknowledge another narrative).

          • Douglas Knight says:

            dndnrsn, don’t be ashamed to make universal statements. If that’s what you mean, say it. I think people make too many disclaimers.

            cassander, you mean understate, right? Of course weight loss is about calorie reduction, but your theory of low-carb seems to be that it is an arbitrary restriction. Some people (Scott Adams?) put forward such theories. But most people claim that low-carb diets make them less hungry. It’s not just about knowing when to stop at mealtime, but not suffering in between meals.

            I think most experiments show that low-carb diets have an advantage for the first six months, which matches your father’s experience. If it is really true that it has and advantage in the short term but not the long term, that’s pretty weird and needs to be explained.

          • dndnrsn says:


            Yeah, that’s the trick of low-carb diets. People usually eat a few hundred calories less a day without really thinking about it due to stuff like skipping the bun on a burger. It’s less that there’s some magical thing that goes on when you don’t eat carbs, and more that if you can get someone to eat, say, 250 fewer calories a day, they’ll lose ~25 lbs a year. Plus, low carb means there’s a big water weight loss early on, which helps with morale.

            If I were to promote a diet that was about avoiding food dyes, and it caused people to lose weight, and my theory was that food dyes cause weight gain – my theory would wrong, but the results would still be there. What would be happening is that people are cutting out highly-processed calorie-dense foods – what else is there food dye in?

          • dndnrsn says:

            @Douglas Knight:

            If there’s a universal claim I can make based on my own experience, it’s that fat loss is about strategy rather than the actual diet in question, so the current research attempts are misguided. I’ve gained and lost fat over the past 4 or so years, more lost than gained (the graph would look like a downwards zigzag) and it’s basically been about figuring what diet I can stand day-to-day, and how I keep from breaking it.

            For example, I need an external enforcement mechanism. This isn’t the sort of thing that an RCT in a hospital ward would figure out.

            EDIT: And so, the attempt to find a “best diet” is worse than misguided, because it leads to people being handed a set of “do this” instructions, instead of being taught the tools to figure out their situation. Then, when they fail, they conclude it’s hopeless.

          • onyomi says:

            Regarding the idea that you can either eat a lot of carbs and not a lot of fat or else eat a lot of fat and not a lot of carbs, many seem to assume it’s something about the combination: maybe carbs stimulate the insulin which stores the calories in the fat, etc.

            But what if it’s really just that fat and carbs produce different kinds of satiety and most people just don’t have the inclination to eat enough of just one of them to consume too many calories?

            Example: let’s say you need to eat <2000 calories a day to lose weight. What if, eating all fat and protein, you just feel you can't eat another egg or piece of camembert somewhere around 1,800 calories. Or eating all potatoes and rice you also can't eat another bite without feeling sick somewhere around 1,800 calories. But if you have as much fat and carbs as you might desire you can easily eat 3,000 calories.

          • dndnrsn says:

            I could eat another egg or piece of meat or bit of cheese. The point at which I think “I cannot eat more cheese” comes waaaaaay beyond 1800cal. It’s just that, for whatever reason, I can control myself better with some foods than with others.

  19. houseboatonstyxb says:

    Google is saying this site has been hacked. My technical support people say it’s not actually anything bad, but this still might be a good reminder to make sure your password here isn’t the same as your password for anything more important.

    So, how does one change the password?

  20. WashedOut says:

    I have an upcoming fitness exam in which I need to be able to run 2.5 km (1.6 miles). I am not aware of any time limits, I just need to not die during or immediately after. I consider myself quite fit in most domains except running, which I never do (cardio kills gains). I don’t even jog.

    I know this might be the ultimate case of barking up the wrong tree but are there any runners on SSC who could suggest ways I can train to achieve this? I have about 2 months.

    Should I start distance jogging, or start with short interval training and increase the distance gradually?

    • oldman says:

      How are they defining running? Can you just walk it? Can you do something that is technically jogging, whilst only being about as strenuous as a walk?

      Also where are you at now? How sure are you that you couldn’t just do this tomorrow?

      • Chalid says:

        This. If you already are basically fit, a 2.5 km jog isn’t much. You should just try it and see if you can do it.

    • Aapje says:

      The danger with high intensity workouts is that if you use bad technique, you can hurt yourself relatively easily.

      Given the lack of a time limit, I would just start jogging. I would suggest building up the distance. Start with jogging 1 km (500 meters away from your home and back again). Try to go slower than you think you can (it’s easy to overload muscles that you don’t use normally, especially if you are fit in general). Then recover for some days at least, where you evaluate the damage.

      If you have a lot of muscle pain, do the 1 km again for your next run. If you have no bad effects, run 2 km. Then go to 2.5 km. Once you can do that comfortably at a casual pace, try to go a little faster, but not too much. Once you are happy with the pace, you can go into maintenance mode where you do that run once a week to stay in running/jogging shape.

    • HeelBearCub says:

      The easiest way to start training running is to do time based intervals. You can easily walk this distance at a brisk pace now (I assume). Start with 1 minute of jogging followed by 3 minutes of walking. Increase the length of jogging as comfortable until you are at 4 and 4. Then start decreasing the walking until you are jogging the roughly 16 minutes it will take you to do 1.6 miles at a fairly leisurely pace.

      Pay attention to whether you have specific new flexibility needs. Calves, quads, hamstrings and hips are (I believe) likely culprits in injuries. A competent physical therapist can help you if you find that you are having pain that is more than just muscle soreness.

    • Rosemary7391 says:

      I find this quite helpful to give me structure:

      The entire plan is supposed to take 9 weeks, with a run 3 times a week, and in theory gets you to running twice as far as you need to. You could possibly skip forward a bit seeing as you’re fairly fit, but you shouldn’t need to do that to reach your goal.

      These pages have more advice on actually doing it and things that can go wrong:

    • powerfuller says:

      I remember that 4 Hour Body guy suggested speed walking as a good way to start increasing running endurance (like walk 15 minutes, then the next day do 15 minutes but walk a longer distance, repeat for a few weeks until you can’t possibly increase the distance without running), and a few popular sports articles seem to agree; I’ll try looking around for some more reputable sources.

    • Anon. says:

      Couch to 5k is good. Get a good pair of shoes for running, they make a huge difference.

    • Incurian says:

      Should I start distance jogging, or start with short interval training and increase the distance gradually?

      Yes. Also, don’t be discouraged after your first and second run when it feels like you’re about to die, it gets much easier. And don’t over-train, you should know the difference between bad pain and sore pain.

  21. Le Maistre Chat says:

    Long, long ago, there was a civilization so rich that they used diamond-tipped tools to cut other tools from sapphire. Its capital was a walled city twice the size of Uruk, surrounded by more than twice that in suburbs. They had advanced dams, and while we find evidence of flooding, they continued to prosper for a thousand years. Then one day an impactor from space destroyed it. DNA from the ancient graves shows that their race was utterly wiped out in its homeland, though survivors fled south to found new cultures…

    Set up for a fantasy novel? New Age woo? No, it’s the Liangzhu culture of prehistoric China.

    • Scott Alexander says:


      “A 2007 analysis of the DNA recovered from human remains shows high frequencies of Haplogroup O1 in Liangzhu culture linking this culture to modern Austronesian and Tai-Kadai populations. It is believed that the Liangzhu culture or other associated subtraditions are the ancestral homeland of Austronesian speakers.”

      That went in a different direction than I expected!

  22. Mark V Anderson says:

    Past threads on SSC have discussed the comparative governance of small countries vs large ones. Ony clearly expresses a preference for smaller countries. And I think Tibor has said the same. I don’t think anyone has said that big countries are usually governed better. I would like to see some discussion on whether or not Ony and Tibor are correct.

    Intuitively I tend to agree that smaller countries are better. In my book “Simplify Government,” I argue that more complex government leads to less efficient and effective outcomes. And I think that smaller countries are almost certain to have simpler governments than larger ones, so in that respect I think smaller countries would likely govern more effectively and be more accountable to its citizenry. The counter-point to this are the many complicated and dysfunctional state and local governments that exist now in the US. Would these governments run better if the Federal government wasn’t a distraction to citizens?

    And I see strong reasons for favoring large countries. The EU began as a way to avoid future European wars by uniting previous warring countries into one confederation (with some large country like features). And it has been extra-ordinarily successful at that – I think it is unimaginable today that any of the EU countries today would declare war on another one, any more than Kentucky would declare war on Ohio. And I think part of the rationale of the EU was also to try to duplicate the economic advantages of the single market that exists in the US. And I think the EU has been successful there too, with borders becoming much less significant than they were. And despite all the Euro bashing that has occurred over the last ten years or so, it seems to me that a single currency has also been very effective in simplifying lives of people traveling between countries or doing business with each other. Admittedly I have little direct knowledge of the impact, having never even been to Euro-land. But from my distant perspective it appears to be a great thing to have a single currency and many unifying laws across the EU.

    And I do find it difficult to imagine life in the US if it was broken up into many smaller countries. I can now drive for thousands of miles across the country without encountering a single government official questioning why I am traveling or where I am going. I can move to anywhere else in the country and get a job and live there as easily as my own location. Of course I would have to follow their local laws and customs, which might be different from my current location, but still under the US constitution and laws. There is a lot of freedom in having a large country available for travel and living and commerce without substantial borders between different locations.

    Does anyone know of empirical research on the advantages or disadvantages of large countries vs small ones? Like perhaps how do large vs small correlate with the rankings of countries in Freedom House or other types of rankings?

    • Le Maistre Chat says:

      I don’t think anyone has said that big countries are usually governed better.

      There’s just one quality of life advantage I know of. In a big country, dreams stay with you, like a lover’s voice fires a mountainside.

    • Dr Dealgood says:

      The EU began as a way to avoid future European wars by uniting previous warring countries into one confederation (with some large country like features). And it has been extra-ordinarily successful at that

      Has it?

      The modern EU has only been around for a very short time, less than a quarter century. And its predecessor, the ECC, was for the first 34 years of it’s existence faced with the existential threat of the Eastern Bloc.

      In terms of peace in Europe, EU has a better track record than the League of Nations and the Concert of Europe. But it seems early to call it extraordinarily successful.

      • JayT says:

        Also, do we really think that the members of the EU are any less likely to attack one another today than they were in 2001? I’m not a Europe expert by any stretch, but I don’t remember there being any real chance of war between the member nations around that time.

      • cassander says:

        I’d say that NATO deserves more credit for peace in europe than the EU.

    • Aapje says:

      @Mark V Anderson

      I think it is unimaginable today that any of the EU countries today would declare war on another one, any more than Kentucky would declare war on Ohio.

      It also seems quite unlikely that any of the EU countries would declare war on Turkey, Marocco, the US or even Russia. In general, the European countries seem to have mostly become very wary or war, to such an extent then when they do want violence to be used, they look at the US to do it.

      How much (if any) of that is due to the common market? Aren’t you just telling a just-so story/EU propaganda, where the actual reason is much more complex and involves the repercussions of two nasty wars being fought on European soil.

      IMHO, I see a lot of ressentiment developing because people are forced into ‘solidarity’ that they see as unjust.

      • Mark V Anderson says:

        @Aapje. I don’t have evidence that the EU has achieved this, but I don’t know that we can ever be sure given all the confounding variables. But it is true that avoidance of war was the most important reason for the original common market of 6 countries, and there have been no wars between members since it was started, after two devastating wars in Europe in the first half of the century. Right there is pretty good evidence. And I think it makes very good sense that having a kind of supra-national shell over several countries and making travel and commerce very easy between them greatly quells the urge to war on fellow members. Of course there have been many reasons for the “long peace” since WWII, as Steven Pinker calls it, but I think the EU played a big part.

        Yes, EU countries have fought few wars with other neighbors either, but I think that is as much a consequence of the peace within the EU as anything else. People in the EU find it hard to imagine warring on fellow members, and then as they get used to more peace they find it harder to imagine war on other neighbors too. As I said, the EU is only part of the general drift towards more peace in the world since its experience with the terrible world wars, but the trend has been stronger in Europe than elsewhere, partly because of the EU.

        • Aapje says:

          What Tekhno said. I expect that future historians will blame the failure of the EU on overreach.

          The strategy for any problem in the EU is to double down, which is a very risky strategy that cannot but fail eventually… and spectacularly.

      • Tekhno says:

        The EU proper didn’t exist for the vast majority of the “long peace”. It only slowly came into existence as a real political union. The Schengen Agreement wasn’t signed until 1985. The Maastricht Treaty wasn’t signed until 1993, and monetary union didn’t come until 1999 (Eurozone). It may have evolved from the limited common market that already existed, but most of the hard political, national, and monetary elements that make the EU the EU didn’t appear until very late, and after accruing monetary union, the EU started to get into big trouble, in my opinion, due to mismatches in appropriate monetary policies for the North vs the South.

        • Mark V Anderson says:

          The EU proper didn’t exist for the vast majority of the “long peace”.

          The Common Market began with the Treaty of Paris in 1951. This treaty included France and West Germany, two main antagonists in the previous wars. IMO, this was very important to the beginning of the long peace. The addition of other countries has only added to the peace.

          Perhaps Aapje is correct that the EU has extended their scope beyond what is sustainable. I don’t know the details well enough to judge, and Aapje is in the middle of things. But the EU has been greatly successful in their results thus far. The possibility that Greece might have been able to exit their recession earlier if they could have de-valued the drachma pales in significance compared to the decades of peace in part brought about by the EU / Common Market.

          • The original Mr. X says:

            One thing I’ve noticed, which tends to complicate discussions, is that Americans often assume that the EU is basically a free trade zone like NAFTA. It’s not. The Common Market, back in 1951, was basically a free trade zone like NAFTA; now, though, the EU has its own court system, its own legislative body, its own executive (all of which have the power to overrule national governments), its own laws (which take precedence over national laws), its own foreign minister (albeit nobody really listens to him or her), and there’s currently talk of giving it its own army. The EU really is a totally different beast to the old Common Market, and I don’t think you can say “Yeah, the Common Market helped keep the peace, the EU is basically the Common Market, therefore the EU’s helping to keep the peace.”

            (I hope it doesn’t come across as if I’m picking on Mr. Anderson here. It’s just that I’ve noticed this sort of thing before, especially in the aftermath of Brexit — “Oh yeah, the British voted to leave because they all hate free trade or something” — and his comment just happened to remind me of it.)

          • Aapje says:

            its own laws (which take precedence over national laws)

            More or less. The member states have a variety of ways to ignore EU regulations. For example, the EU apparatus merely makes rules, but has almost no executive ability (this is why the EU has fewer employees than a decent EU city). So the implementation of laws is left to the member states, where it is assumed that they will implement the law. The member states can choose to fail at their task, like happened in Germany with the diesel scandal, where the government chose to ignore the 2014 warning letter by the EU about the issue and chose not to fine VW. The only reason why this issue was made public is because VW did the same in the US and the US government refused to keep quiet.

            The more I learned about how the EU actually operates, rather than the theory, the more I began to dislike it.

          • rlms says:

            “The more I learned about how the EU actually operates, rather than the theory, the more I began to dislike it.”
            Would you prefer it if it was a domineering, bloated organisation that ruthlessly enforced its laws the same way in each country, regardless of how appropriate that was?

          • Aapje says:


            To me, one of the advantages of good government/bureaucracy is a relative lack of capriciousness. In my view, if a government does make a law, they better mean it.

            Selective enforcement creates an environment where the rules are not actually the rules. This undermines the entire democratic process, because the mechanism that we use to allow people to influence the law in a supposedly fair way is then undermined by granting a lot of power in a far less democratic and transparent way. Nobody ever voted for selective enforcement of emissions laws or for immunity for VW. And if you are not called VW and actually are strictly held to the law, it’s very hard to object, because the ‘law is the law.’ So this gives a lot of power to the selective enforcers to help one group/business over the other, which is unchecked power that enables a lot of corruption.

            To answer your question directly: I think that the EU does too much and has a very, very poor democratic system that results in minimal checks and balances; preventing citizens from correcting it. Most of this is by design, as the EU mantra has been that a small elite knows better than the conservative populace and has to do things that the populace would not support, but that this is OK, because the elite is obviously correct and the people will see that they are correct later on.

            Basically, the same rationalizations usually employed by dictators. The divide and conquer that the EU uses to blunt opposition is also typically used by dictators.

          • Aapje says:

            To give an example of the absurdity of EU politics: I cannot actually vote for EU parties. EU citizens vote for national parties.

            So surely this then results in a mess in EU parliament? No, because those national parties then join up in just a few blocks who usually vote en masse. So you get the absurd situation that we have Dutch political parties that campaign against each other, but they actually end up in the same block and vote together. So the elections are often just politicians lying to the citizens to get cushy seats for their political party. In the EU, they will vote for things that they disagree with on the national level, due to this ’rounding off’ into blocks. However, the politicians that campaign with agenda A during national elections can’t reasonably be expected to suddenly defend agenda B (the agenda of the block in EU parliament that they are part of) during EU elections. So the system makes them lie.

            A lot of countries/federations have two chambers, where one has members who are chosen regionally and the other has members chosen by all citizens. The EU has a single chamber which is neither/both and thus is a malfunctioning mess, as it has no clear purpose or mandate.

            PS. A lot of people seem to see the EU as a semi-equivalent to the US, but when it comes to the political system, it is actually quite different, and not for the better. And I already really, really dislike the US system, so you can imagine how much I hate the EU political system.

      • Tibor says:

        In general, the European countries seem to have mostly become very wary or war, to such an extent then when they do want violence to be used, they look at the US to do it.

        Except for when Sarkozy decided that France will play world politics and messed up Libya…then everyone pretends that it was the US. To be fair, European militaries are in such a bad shape that they probably would not have been able to pull of the attack without US support. But it was still mostly a French operation.

        Otherwise I agree. It holds for European countries in general and for Germany in particular.

    • Incurian says:

      I can now drive for thousands of miles across the country without encountering a single government official questioning why I am traveling or where I am going.

      Not in the southwest.

      • Mark V Anderson says:

        Not in the southwest

        Are you referring to those awful Customs checks close to the Mexican border? I got caught in one of those in Texas some years ago. Isn’t that only within 50 or 100 miles of Mexico? And I do hope those will eventually go away, as the populace gets further way from its 9/11 hysteria. Of course the Trump phenomena seems to argue against the reduction of hysteria, but I am thinking that is just a temporary blip. Wishful thinking perhaps.

        • random832 says:

          There’s a map of a “constitution-free zone” floating around accompanied by the assertion that it’s legal (regardless of if it’s common) within 100 miles of any border (including the coasts and the great lakes). And another one with circles drawn 50 miles around every international airports, with no information about what this means. I’ve also seen one that superimposed it with a map of federally owned land for maximum hysteria.

        • Incurian says:

          Are you referring to those awful Customs checks close to the Mexican border?

          Yes, and they are awful. I had one on my commute in Arizona where I never had a problem during hundreds of encounters, but in New Mexico and Texas where I’ve had perhaps a couple dozen encounters the douchebag rate is like 60%.

      • Tibor says:

        Also, in Europe, if you drive for a few thousand miles, you’re either in the sea or not in Europe anymore 🙂 Other than that, the EU has the Schengen zone, you can drive through it without being checked.

        Also the idea of driving so far is a bit strange for Europeans. I think most Europeans consider 200 km (about 130 miles) quite far away. It is probably because 200 km away is quite possibly a different country (and 500 km away almost definitely is) and that means a different language and to some extent a different culture. North America is culturally extremely homogeneous.

        • The original Mr. X says:

          Also the idea of driving so far is a bit strange for Europeans. I think most Europeans consider 200 km (about 130 miles) quite far away. It is probably because 200 km away is quite possibly a different country (and 500 km away almost definitely is) and that means a different language and to some extent a different culture. North America is culturally extremely homogeneous.

          As someone or other once said, in America, one hundred years is a long time; in Europe, one hundred miles is a long distance.

    • Kevin C. says:

      “I can now drive for thousands of miles across the country without encountering a single government official questioning why I am traveling or where I am going.”

      Just a quick reminder that this isn’t so much the case for those of us in the non-contiguous states.

      • Mark V Anderson says:

        Okay, but I was on a cruise ship out of Seattle a couple of years ago, and I don’t think I saw any government officials on our stops in Alaska.

      • Paul Brinkley says:

        I think government officials concerned about people driving between Hawaii and California mostly just let the problem solve itself.

    • Tibor says:

      The EU likes to boast about how it secured peace, even got a Nobel peace prize (as did Obama for becoming president, or Yasser Arafat for ceasing to be a terrorist…). But I find that claim rather dubious. First of all, the EU started in 1994, since then, there have been two wars in Europe (Yugoslavia and Ukraine), albeit not in the EU (well, Croatia is now also an EU member, but anyway). Before 1989, Europe was divided into two solid blocks. In the eastern block it was completely impossible for any Russian vassal to do anything than the Russians did not want, as they repeatedly demonstrated. In the west, the threat of the Russians invading was more than enough to keep peace. If anything kept peace (save for a couple of rebellions in the East which were quickly suppressed) in Europe as a whole, then it was the NATO and nuclear weapons. What keeps it peaceful now, I believe is once again the Russian threat as well as the memory of two World (but mostly European) wars in the last century. The EU makes internal European trade easier and that is a great thing. But that is not enough to keep peace – prior to WW1 Europe was also economically interconnected and people argued that starting a war in Europe would be completely insane, since everyone would lose (and they were right of course).

      Apart from this though, the increasing (albeit now it’s kind of seems to have ground to a halt, hopefully it will not resume again) centralization of the EU is rather a source of conflict between European countries, as does democracy in general. Instead of everyone having their own rules, now they have to agree and since some things are done by majority voting, sometimes a country ends up with rules it did not want. I would very much like to see the EU roll back to its roots – a free trade (that includes movement of labour) zone, no more and no less. Had France left instead of the UK, it would theoretically be possible to at least get rid of the idiotic agricultural quotas (there are maximum quotas for how much can be produced in each country…to keep the prices up and competition down and it is one of the things the French governments care about the most in the EU) and subsidies. This way, it is less likely.

      Maybe you can’t have both worlds, maybe the nature of politics is such, that you always either drift towards a political union with all its drawbacks, or closed down protectionist states. But I doubt it. The Eurofederalists paint a false dichotomy – either a federation ruled from Brussels or Marine Le Pen-like government in every European country. I think you can have a group of cooperating independent countries instead. Switzerland manages to do that just fine, so do Norway and Iceland.

      • Aapje says:


        I’d like to add that the EU construct is a major source of irritation. There is a lot of ‘Germany/Merkel are Nazi’s’ and general irritation. This despite the stability of EU project of the last 15 years mostly being bought with German money. I don’t see how future generations of Germans are willing to sacrifice that much and as we all know, people tend to seriously get angry if you take away their income, even if it was a temporary gift. So the ressentiment that is already quite strong in countries like Poland and Hungary is really going to ramp up if the Germans reduce the transfer payments.

        Of course, Eastern Europe is quite afraid of Russia, so that will make them reticent to break off. But I see some of them blocking a huge amount of legislation, effectively stopping the EU in its tracks.

        You already see this with the refugee crisis where the majority promised to distribute some refugees over all countries and these countries simply refuse. They are not even refusing as EU countries typically do, in secret or by creative interpretation of the laws, but open defiance. This clash of cultural values cannot but sink an ‘ever closer union.’

        Britain tried to get rid of the ‘ever closer union’ statement and we all know how that worked out. Ultimately, the EU apparatus is cursed with the issue that true believers go to Brussels/Strasbourg and thus there is an insular culture among the EU elite that sees only one solution to every problem: more integration.

        However, these people don’t understand human nature, just like the communists didn’t. They have the mantra that extreme economic integration needs political integration, yet they fail to understand that political integration needs cultural integration; which is simply not going to happen. The Greeks are the Greeks, the German are the Germans and the Polish are the Polish. There are simply no mechanisms in place to eliminate cultural diversity between these countries.

        • Mark V Anderson says:

          @Aapje. It sounds like you think that the EU has become pretty much un-manageable with so many countries. It does make sense that doubling the number of countries without changing how voting and such is handled doesn’t make any sense. And I think that is what has happened in the last 10 years or so? Perhaps the EU needs to dial back its scope of business since it will never get a consensus on many key issues with such a large group.

          • Tibor says:

            That’s basically what Aapje and I were saying. Dial it back and have the decisions be done on unanimous basis, as they used to be. That way, you will have a slower but more stable integration and it will stop at a more natural place. A lot of EU integration is actually done either without letting the citizens have. I doubt the Euro would have been passed if it had been a subject to referendums in each country. In retrospect (although it was clear to many people back then as well), this would have saved the EU a lot of trouble.

            That also shows a nice example of why separate can be better. The Slovaks were (maybe not as much today, but they’re kind of stuck with it) very supportive of introducing the euro. The Czechs, on the other hand, were always very skeptical about it. In the end, both countries get what most of their citizens wanted. Had Czechoslovakia stayed a single country, this would not have been possible. And the Czechs would be very angry at the Slovaks today if we had introduced the euro because of them. And since both countries are in the Schengen zone*, there is nothing preventing me from going to Slovakia without getting checked at the border, or more imprtantly moving or working there. I have to change the money, but that’s not a big deal.

            *as are all EU countries, apart from Romania and Bulgaria, whom others did not want there yet, and Britain, which is soon not even going to be in the EU anyway

          • Aapje says:


            The big problem is that the EU leadership wanted to achieve competing goals. They wanted to expand to make more countries strongly bound by trade, with the goal to keep the peace and make countries democratic.

            But they also wanted an ever closer union, with extreme integration, to achieve a humanitarian walhalla, a shared foreign policy and other lofty goals.

            However, these goals inherently conflict, because if you bring in semi-medieval countries like Romania and/or young democracies, then it is absurd to think that you can keep making new advanced regulations that depend on a well-functioning state with little corruption. These countries don’t have the maturity to handle the power they are given and the large number of them makes it very hard for the EU leadership to stay on top of things. Especially as the ill-functioning countries start to use their power to prevent investigations and punishment for other ill-functioning countries, as they know that if they allow this against the neighbor, it will probably end up used against themselves later on.

            Greece is a good example of what happened when a country with a poor legislative culture gets the power to print money: they print money and spend and spend and spend & lie and lie and lie; until the lie can no longer be maintained and the chickens come home to roost.

        • Tibor says:

          I don’t know about Poland or Hungary, but I think that the Czechs don’t care about the subsidies all that much, despite being a net receiver of EU money. The government (the current prime minister is a social democrat, so that also goes together with wanting to have more money to redistribute) cares about it more than the citizens, I think.

          That might be for two reasons. The Czech republic is a net receiver of EU money but it is still richer than Poland or Hungary, which means that the difference between what we pay and what “we” get is smaller. Also, the popular (and I believe correct) opinion is that the subsidies are often used for senseless projects, sometimes outright dubious things (like a private golf course or a hotel), which should not be subsidized from public money (IMO if you have to have these subsidies they should all be used to improve the infrastructure which roughly benefits everyone), so then if you account for that we pay more than we actually get. Plus you have a bunch of businesses which specialize in rent seeking – helping people write good proposals for the EU subsidies. This is another cost.

          It might also be my own bubble a bit, I have a disproportionate number of libertarian-leaning people around me. But disproportionate number of a small minority is still a minority and I get the feeling that most people generally see the EU subsidies as something corrupt. I don’t know any people in agriculture or anyone who actually gets these subsidies, they probably have different views. But if there were a popular vote for or against subsidies, I think the Czechs might vote for abolishing them.

          The Czech government also did not go “full Orbán” on the refugee thing. I think it ended up most Brussels-supportive of the Visegrád group countries – i.e. Hungary, Slovakia, Poland and the Czech republic, although that still meant about as much support as came from Denmark, which is not much. What annoyed me about this thing in particular though was that back in 2015, Merkel and the German media were all about how backwards all these “eastern Europeans” are (and they wrote the same way about Eastern Germans, who seem to have similar views of this). I read a couple of commentaries which tried to psychoanalyze those countries, saying that it is is somehow the legacy of communism that affects the people’s thinking and that the Eastern Germans are still influenced by Nazism, because the Deutsche Einheitspartei (the DDR communists), while officially anti-fascist, did not do much to uproot Nazi thinking. But all these commentaries completely ignored that Britain, Denmark and France were doing pretty much the same, if perhaps a bit less conspicuously (and it is easier for Denmark than for Hungary due to geography) and Spain actually built a wall in its African exclave Ceuta. In the end, it was Austria which went against Germany and organized that fence building with some non-EU Balkan countries but there were no articles about Austrians being Nazis. Tthis was a bit later and the sentiment in the German media changed a bit since late 2015, so that might be a reason. I think that now everyone is kind of glad that Austria did that, because Merkel’s deal with Erdogan was a catastrophe and unfortunately, no actually good solutions seem to be politically feasible, because the right just wants to build fences and the left wants to give refugee status to everyone who comes from Africa or the middle east.

          • Aapje says:

            I admit that I painted with too broad a brush, there are obvious differences between the Eastern European countries and I rate the Czech people/culture quite highly.

            The subsidy system is quite distorting in that it’s ‘free money.’ If you don’t use the subsidies, the money just goes to other countries, so there is a big incentive to do things to get the subsidies, rather than do what is needed. Personally, I feel that there are far too many subsidies and that those which are useful should mostly be granted at the national, rather than EU level.

          • Tibor says:

            @Aapje: I did not take any offence, just wanted to specify things. I think that there are still a lot of Czechs who would want the government to be like Orbán. And for example Zeman (the largely ceremonial president, who gets more media attention than he should) is coveting those people.

            Then there are few weirdos like Tomio Okamura, who’s an anti-immigration half-Japanese (well, apparently, his father is half Japanese and half Korean and his mother is Moravian) immigrant, but though he had some success at first (about 7% in the last parliamentary elections), his party basically broke down and its support plummeted. Now he spends times talking at neo-nazi “worker’s party” rallies, which is so absurd it is hilarious (well, to be fair, he did that at least once). I think he started a new party recently after the previous one melted away.

            Btw, Okamura also had, possibly still has, this weird business where his (mostly Japanese) customers send him plushies and they then take them somewhere in the world and take a plushie-selfie, then send both the picture and the plushie back to the customer. Although quick googling seems to show that the business was not terribly successful, so it probably does not exist any more. He did have some actual success with other businesses though.

          • Aapje says:

            few weirdos like Tomio Okamura, who’s an anti-immigration half-Japanese[…] Now he spends times talking at neo-nazi “worker’s party” rallies, which is so absurd it is hilarious

            Some people are just born into the wrong body. Having a very xenophobic brain and being a tiny minority leaves him unable to realistically rally around his own ethnicity, yet the those feelings can’t be suppressed so they come out in weird ways.

          • Tibor says:

            @Aapje: An acquaintance of mine actually went to see that rally, it was I think a year and a half ago or something like that. In fact, there were 3 demonstrations taking places on the Wenceslaus square in Prague at the same time and he went to all of them to observe – one was by neonazis, one was by sort of people who might vote for Wilders or the AfD but are not really nazis and one was by hippies who had transparents like “let’s take 1 million refugees!” or “we’ve never been a nation!”. Granted that the square is fairly large (or rather long), but still, the city official who allowed all three at the same time and place has to have a very peculiar sense of humour. In any case, the police managed to keep the neonazis from attacking the hippies, but not for the lack of effort on the nazi side. A funny thing was that in response, the hippies just literally sat there, “nonviolently protesting”, but very much depending on the police muscle to keep them safe. Well, at least they’re kind of cute, not like the aggressive leftists like Antifa.

            Then at the nazi demonstration (my acquaintance has long dreadlocks which is not all so safe when surrounded by skinheads in military boots), Okamura had his speech. After a couple of beers, one of the demonstrators apparently shouted “Why do we listen to him?! He’s Japanese!” Some others calmed him down: “Man, he’s a good Japanese!”. I have to give it to the first guy that, unlike the others, he was consistent in his world view.

            The third demonstration was less interesting and also badly prepared, somehow they forgot to bring loudspeakers, or the loudspeakers did not work and a lot fewer people came than they expected. They then spend most of their time complaining about how badly organized they were and accusing each other, although mostly people who weren’t there. I cannot help myself not to feel that that tells something deep about the sort of people who would go to this demonstration.

  23. thoramboinensis says:

    ISIS has been using drones to drop improvised explosives on Iraqi troops.

    Police departments around the world have been training eagles to take out illegal drones.

    How long will it be before the US military starts deploying tactical eagles against ISIS? It’s hard for me to think of an image more patriotic than a Bald Eagle taking out an ISIS drone.

    • the anonymouse says:

      It’s hard for me to think of an image more patriotic than a Bald Eagle taking out an ISIS drone.

      Yup. Pretty sure I just got an America boner.

    • Le Maistre Chat says:

      The fact that they haven’t started means the answer is “too long.”

    • John Schilling says:

      The imagery is great, right up to the point where ISIS learns to put tilt switches on those explosive charges and film the resulting carnage for their propaganda. Bonus points for having ISIS troops chow down on Eagle McNuggets right afterwards.

      The Eagles may be a useful stopgap against merely criminal use of drones; they aren’t going to cut it against even field-expedient military drones.

    • Civilis says:

      Police usually have limited R&D budgets and PR issues with using guns in peaceful urban areas. The US military has a huge R&D budget and a lot of guns and doesn’t have a problem using them in areas where ISIS might be found. Also, given the number of Americans in the US military from rural backgrounds, the correct response to an ISIS drone might well be “pull!”.

      On the other hand, the US Military has been training dolphins for use against undersea mines, which is conceptually very similar, so it’s not that ridiculous of an idea.

  24. onyomi says:

    Very un-pc, but with “International Women’s Day,” female Atlas Shrugging, and that interesting video where Clinton and Trump swapped genders (though that was in the “no culture war” thread I believe, so maybe we can discuss that more here too), I thought this would be worth discussing (but don’t worry, I’m going go say unflattering things about white men, as well as feminists!):

    We’ve talked a lot about automation taking away jobs, a little less about foreigners taking away jobs, but almost never about what seems to be the elephant in the room: women joining the workforce. In other words, to play fast and loose with the numbers, the workforce doubled while the population stayed roughly the same (or grew at a static or declining rate).

    If I imagine my field of academia, for example, with the same number of jobs needing doing, but only men are allowed to fill them… well suddenly it’s a hell of a lot easier for me to get a lucrative, desirable position (though the field is also greatly impoverished by the loss of all the female researchers, who, in the liberal arts, are, if anything, bigger contributors than men nowadays). Take away women and foreigners and suddenly a field where everyone agrees the job prospects have declined dramatically (though there are disagreements about why: one which rings truer to me is not so much “fewer jobs,” but “fewer good jobs,” as in, jobs which used to be done by someone with a PhD and a “bread winning” job are now done by adjuncts with MAs with pay commensurate for a part-time gig; this may reflect some bias about jobs done by women: suddenly non-stereotypical breadwinners flood the field=>suddenly the field is full of non-breadwinning jobs).

    But academia is not ideal because, of course, if you took away the female professor jobs you’d probably simultaneously be taking away a lot of the female students, so demand for teaching would go down too. Imagine other fields where the demand would remain relatively static whether or not women were working. The difference there would be even more stark. My thought about the hypothetical “day without women at work” is not that men would look around and say “gee, we couldn’t get anything done around here without women” (which may be true), but rather look around and say “hey, looks like it’s time for a promotion for me! Also, now I get to treat this space like a locker room/boy’s club.”

    This also sounds very un-pc, but it’s also unflattering to men, of course: basically that, when forced to compete with (and socially deal with the presence of) competent women in the workforce, men, who have always gotten to dominate by accident of birth don’t always do so well. But as mentioned in the other thread, the result of feminism is largely more acceptance of women taking on masculine roles, not more acceptance of men taking on feminine roles (though some of that has happened too); that is, women are now allowed to be women or men, but men are still expected to be men: no wonder men are having a hard time: the competition for being a man just doubled in intensity!

    None of this is to suggest that women shouldn’t be in the workforce, but rather that it just seems like something we have to think about seriously before wringing our hands and saying “oh well, it looks like there are no more jobs for blue collar men in the globalist, robot-driven future.” If anything, we should be pretty amazed that the economy has managed to absorb so many new workers in such a short time without even greater unemployment and displacement.

    So all these men are displaced by women and foreigners in such a short time and we are surprised that they had the temerity to vote for a masculine business guy who promises to stymie the foreigners and who has seemingly less than perfectly progressive attitudes about women (though I think that was exaggerated; clearly Trump has no problem hiring women in important positions). I think if Blue Tribe especially wants to win back the votes of white men they will have to stop just calling them entitled whiners (even if it’s true that historically, white men had lots of advantages over women and minorities, we must also always keep in mind that those white men are not today’s young white men; they are dead, retired, and/or the really old guy running the company, not the young white guy competing for the entry-level job with the women and minorities) and actually think about what to do, be it, be more accepting of men staying home and raising kids, or what have you.

    • Dr Dealgood says:

      (It’s still me, just need to change my gravatar back later.)

      I thought it was pretty uncontroversial that women entering the labor force depressed wages. You can’t effectively double the supply of labor without reducing prices.

      That said, this alone shouldn’t have had a huge effect on the old-school male dominated resource extraction and manufacturing jobs. Women are largely working in the service sector, by their preference: how many women choose to become oil riggers or auto workers? Even with pressure from men competed out of the service sector, wages in the other two sectors of the economy shouldn’t have been crushed as thoroughly as they have been.

      I might be wrong on the economics here, but while your analysis has some merit for men doing white collar work the (former) blue collar majority should have been able to weather it. If anything, it seems more like men are being forced to compete for feminine* jobs than the reverse.

      *I don’t really have any other good way to describe what I’m talking about. Maybe ‘soft handed’ would work better? Academia is absolutely not masculine, and industry isn’t much more so at the white collar level. Put my hands and my father’s side-by-side, you can easily see which of us is competing with women in our respective fields.

      • Mark V Anderson says:

        I thought it was pretty uncontroversial that women entering the labor force depressed wages.

        It seems pretty obvious, but I have very rarely seen this spelled out. One sees article after article about how median wages have increased little over the last 30 years (exaggerated in my opinion), but how many folks bring up the greatly increased presence of women in the workforce as a major cause? I think this has probably decreased wages more than any other factor over this time period, but instead one hears about pro-business laws and greedy execs from the left, and lazy workers, imports, and immigrants from the right, with all of them pretty much ignoring the elephant in the living room. Of course discussing the main cause of their distress does nothing to advance the agendas of either group, so they both prefer to ignore it. But I do think because it is rarely discussed, it rarely occurs to the average lay person as a major cause of depressed wages.

      • onyomi says:

        That said, this alone shouldn’t have had a huge effect on the old-school male dominated resource extraction and manufacturing jobs. Women are largely working in the service sector, by their preference: how many women choose to become oil riggers or auto workers?

        Yes, but as you say, there is downward wage pressure on the men in traditionally male jobs competed out of the service sector, which is the fastest growing part of the economy.

        That is, at the same time as women entered the workforce en masse, coincidentally (?) the type of work demanded started skewing heavily toward the type of job women are more likely to want to do. Women entering the workforce is still key, then, because if they hadn’t, there would have been more service sector jobs for the out-of-work oil riggers.

        • Dr Dealgood says:

          If you take the decline in the primary and secondary sectors as given, then yes the amount of competition in the service sector is key. But why should that be a given?

          The argument that the number of jobs in manufacturing and raw materials must necessarily decline needs to be supported not just asserted. Disadvantageous trade policies and over-regulation of key industries aren’t laws of nature.

          • onyomi says:

            Disadvantageous trade policies and over-regulation of key industries aren’t laws of nature.

            Oh no, I’m not just assuming that a shift from manufacturing to the service sector was part of the US destiny (though there is probably a general move in that direction which is inevitable due to automation); I do think overregulation, etc. are also to blame. Question is: is it a coincidence we started getting regulation that killed manufacturing right around the same time a much bigger proportion of the workforce became people who tend not to like manufacturing jobs?

      • Corey says:

        More people working *grows* the economy as a whole, e.g. if a man marries his maid, all else being equal, GDP goes down. (Which is not to say it doesn’t depress individual wages).

        Most econ commentary I see that mentions women entering the workforce, instead of talking about wages, comes at it from this aggregate perspective – wondering how to keep the economy growing though we’ve picked that low-hanging fruit.

        • caethan says:

          Wait, what? The only time I’ve ever seen the “when a man marries his maid, GDP goes down” bit before is as a criticism of GDP as a measure of the size of the economy. E.g., here noting that home production is excluded from GDP. I’ve never seen it used like this before, as a criticism of home production!

      • IrishDude says:

        You can’t effectively double the supply of labor without reducing prices.

        Isn’t a countervailing force that women spend the money they earn, thus increasing demand for labor?

        • onyomi says:

          Yeah, but women spent money before they had jobs. Of course, to the extent a two-income household is richer than a one-income household they should get to consume more, but it might be consumption of goods and services which didn’t used to get counted in GDP because housewives did them–which is to say, we’re no richer, overall.

          My sense is that most working couples today make more money than most single breadwinners did back in the day, but that each of those jobs, individually, tends to make less, inflation adjusted.

          Example: used to be man made 80,000/yr. inflation adjusted, woman stays home, takes care of kids and runs the house.

          Now, man and woman both work at jobs which pay 50,000/yr., so they contribute more to GDP, but they also spend 20,000/yr. on daycare, housekeeping, eating out more often, etc. Are they actually any richer? To the extent the woman prefers working and eating out to spending more time with the kids and cooking at home, arguably, yes. But to the extent she views her job as just a means to an end (support her family), no.

          Which is not to say that increased spending on e. g. daycare actually totally negates the increased wealth derived by women working. It probably doesn’t. I don’t think we spend that much more on daycare and cleaning and restaurant meals.

          But I still think it’s likely that the average job pays less, overall, given the much higher supply of labor, than it would otherwise. That is, we can have a world where man jobs pay 80,000/yr and woman runs the house, or a world where non-gender-specific job pays 50,000/yr, everyone works and pays for daycare, etc., but we can’t have a world where non-gender-specific jobs pay 80,000/yr., everyone works, and now every household just gets to enjoy twice as much consumption.

    • dndnrsn says:

      I think you’re misunderstanding the history of academia a little bit. Even when female professors were rare, female students were less rare – at least at the undergrad level. Mrs. degree and all that.

      Further, your thesis runs into a problem – the men who face the least competition from women are the men who are the most likely to vote for Trump. Conversely, the men who face the most competition from women – men in white-collar occupations but specifically the less testosterone-driven white collar occupations (some fields of law reward a far more aggressive personality than others, ditto medicine, business rewards a far more aggressive personality than teaching, etc) – were probably the most likely to vote for Clinton. Likewise, “pink collar” jobs (secretary, typist) were still done by women back in the day, and are either still done by women, or have been replaced by people learning how to type themselves.

    • Scott Alexander says:

      I asked about this before here. The consensus seemed to be that lots of men mysteriously left the workforce around the same time women entered, so no problem. I don’t completely understand this.

      • Jiro says:

        I don’t see there having been a consensus there at all. A lot of things were said, some of which could be checked but which nobody checked.

        • The Nybbler says:

          Research suggests that men avoid college majors and graduate fields that are 24 to 54 percent female

          That’s hilarious, since one of the reasons men in tech are often beaten rhetorically over the head is that our mere presence intimidates women out of the field. Now they’re saying it’s us poor men who are intimidated out of a field once there’s too many women?

          I doubt it’s true, however. I’ve looked at the male-female ratios in many of the STEM fields and there isn’t evidence of a positive feedback effect.

          • rlms says:

            “Now they’re saying it’s us poor men who are intimidated out of a field once there’s too many women?”
            It seems plausible that both could be true, for different fields (I’m dubious about both).

    • 3rd says:

      Is the full performance viewable online? I could only find the two minute clip from the Guardian.

    • Scott Alexander says:

      There isn’t actually a way to integrate neuroscience with psychiatry right now, but everybody agrees that there should be, so every year or so we get another “Integrating Neuroscience With Psychiatry: The Time Is Now” article (or possibly the same one, slightly edited; no one ever reads them all the way through so there’s no way to be sure). We all nod our heads, agree this is the sort of thing that should happen, and move on.

  25. BBA says:

    So, how about that AHCA bill? I think it’s remarkable – it preserves almost everything that the Right didn’t like about the ACA, while undermining the parts that the Left supported. There’s absolutely nothing for anyone to like, which means it’s sure to pass overwhelmingly. Well done, Mr. Speaker!

    • cassander says:

      As someone who had high hopes for Paul Ryan as speaker, I’m shocked by how terrible it is. I didn’t think it would be good, mind you, but I figured it would at least defensible on at least axis or another. I see absolutely nothing to recommend it.

    • Machina ex Deus says:

      “Republicans are set to replace Obamacare’s system of subsidies, where the government gives you money to help pay for health insurance, with a system of refundable tax credits, where instead, the government will give you money to help pay for health insurance.”
      —Jim Geraghty

      • AnonYEmous says:

        it’s pureblown cancer of the highest variety

        just let obamacare fail, to the point where you clearly have to step in, and then do so.

    • shakeddown says:

      I’m unreasonably angry about it having “American” in the title. When roughly half of America is your hated outgroup, you do not get to claim patriotism.

      (edit: I’m generally okay with various people considering me/my vague ingroup-ish group to be their enemies, ingroup/outgroup dynamics are just part of human nature that I have to live with and heck, sometimes they even have legitimate reasons for their grievances. But when they do that and also claim to be representing/advocating for us, it bothers me).

      • HeelBearCub says:

        You are correct. It’s unreasonable to be angry about that.

        I far prefer that liberals cloak themselves in the flag while lighting the torch on the Statue of Liberty, rather than get mad that conservatives love to love America.

        It’s different when they say “we” aren’t American. Get mad about that.

        • keranih says:

          I far prefer that liberals cloak themselves in the flag while lighting the torch on the Statue of Liberty, rather than get mad that conservatives love to love America.

          I am hoping that it is counter-triballing to agree with you strongly here, HBC.

    • Scott Alexander says:

      If I have to say something nice about it – I’m glad it preserves ability of people with pre-existing conditions to get coverage. I thought that was maybe the most important part of Obamacare and it seems like they’re mostly keeping it.

      Otherwise I don’t really understand it and defer to everyone who says it’s awful.

      • suntzuanime says:

        The problem is that the ability of people with pre-existing conditions to get coverage on the same terms only when they discover they need it sort of makes a mockery of the idea of insurance. That was the whole reason behind the hated Individual Mandate/Obamacare Tax, which AFAICT the new plan doesn’t preserve. It’s not clear that the IM/OT was enough to actually make the system work, but it doesn’t seem like the new plan has anything better to offer.

        IMO we should just give the health insurance industry a giant Thank You For Letting Us Fix The System refundable tax credit so they don’t block us from instituting public insurance.

        • HeelBearCub says:

          Not only does the new plan not preserve the tax penalty for non-insurance, what it replaces it with provides the opposite incentive.

          A one time 30% premium penalty applies when buying new insurance if you have not maintained continuous coverage. This means that if you do not have insurance, you are actually discouraged from buying insurance.

          (Haha, too slow Iain! From the shadows I strike …a silly pose.)

        • Iain says:

          The AHCA’s equivalent to the mandate is a 30% rate hike on plans purchased after a gap in coverage. There are lots of reasons that this is worse than the mandate — just for starters, if you are young and healthy but have had a gap in coverage for any reason, this makes it less likely that you will sign up for insurance, which is the exact opposite of what we want — and I am unaware of any good arguments for why it is better than a mandate. (“Because we want to be able to say we got rid of the mandate in our political ads” is not a good reason.)

    • John Schilling says:

      The only winning move in the politics of American health care is not to play. The ACA is at this point doomed to failure – really, it was probably doomed from the outset, just with enough of a delay for Obama and company to be safely clear. But if there was room for some hope in the past, that’s pretty much gone now. And anything that anyone can put forward to replace the ACA, will either be massively unpopular because of the taxes and the fees and the death panels, or it will be doomed to the same mathematically inevitable failure as the ACA.

      And whoever owns that failure, doesn’t get to build America’s next health care system. The Democrats desperately need for the Republicans to “repeal and replace” the ACA with something that has conspicuous Republican branding, before the ACA collapses. The Republicans desperately need for it to be Obamacare that fails, not their own replacement.

      But, having promised “repeal and replace” and now at least weakly controlling all three branches of government, they at least have to pretend they are trying to do so. A plan that sounds like it is saying all the right things but has every mathematically-literate wonk running for the door, a plan that might actually bleed enough GOP senators to not pass, might be exactly what the Republicans need. They can spend a few years glaring at the obstructionist Democrats and traitorous RINOs who won’t let them repeal Obamacare, and then when it is the ACA and not the AHCA that fails, offer up a monumental “we told you so”!

      Just what they need, so long as they don’t actually pass the thing. If they do, it will go down as an own goal of epic proportions.

      • HeelBearCub says:

        You might be right on the politics.

        You are wrong on the facts of whether the ACA is doomed for failure. I mean, it’s certainly possible that Trump and the Republicans will succeed in making it fail. It’s not foolproof legislation that requires no ongoing management. But it isn’t a structure that is inevitably doomed.

        • cassander says:

          It’s not the structure that was inherently doomed, but the deceptive and slapdash way it was put together that doomed it. As originally written, the law was completely unworkable, which is why there have been so many workarounds, not a few that were extra-legal, since.

        • John Schilling says:

          What cassander says. Any legislation that requires as much “ongoing management” as the ACA is doomed in the long run. Particularly if it offloads as much of its work on the private sector as does Obamacare, because once the prognosis is “probably doomed; if everything works out we might muddle through” their incentive goes to maximizing the short-term profit they can extract before that probable doom. Sooner or later, there’s going to be a government that won’t manage the system to the standards of technocratic excellence the ramshackle ACA requires.

          And what do you know, we just elected that government. I am certain that when the ACA goes down, with God only knows how much collateral damage, you will blame this on “Trump and the Republicans making it fail”. But it will be the ACA that fails, and Trump et al don’t have to lift a finger to make that happen. They just have to not be fool enough replace it with their own equally doomed system first.

        • Brad says:

          It was doomed in the sense that it didn’t solve the cost growth problem. It *may* have helped, the analysis is really tricky, but it certainly didn’t solve it. The cost growth problem will sooner rather than later doom every scheme that doesn’t solve it.

          • BBA says:

            And that’s a direct consequence of the dominance of employer-sponsored insurance, which works extremely well for people who have it but screws over everyone who doesn’t. Anything that tries to dislodge ESI is political suicide, anything that doesn’t will ultimately fail. The ACA tried to preserve ESI while the exchanges were built up, then gradually transition away from it through various measures, including if all else failed the Cadillac tax. But most employers were content to keep ESI, so all else failed, and everyone expected the Cadillac tax to be repealed before taking effect, and what do you know, it will be.

            There might have been some way to thread that needle, given a Congress and an administration that weren’t trying to sabotage the ACA (and a can opener). But now the death spiral is inevitable and soon we’ll be a lot worse off than the 2008 status quo. Is the ACA to blame? Mu.

          • Unless either there is a mandate or the insurance companies are free to base prices on actuarial cost, doesn’t adverse selection doom any scheme?

          • Corey says:

            @DavidFriedman: Correct.

            Mark Cuban had an interesting baby-splitting idea: single-payer for catastrophic and/or chronic conditions, private coverage (with less regulation than pre-ACA) for everything else. Maybe you’d get there through Medicare/caid for all with a percentage-of-income deductible, then taking a chainsaw to private insurance regulations. He’s asking on his blog for people to shoot holes in it if anyone’s interested.

          • gbdub says:

            Mark Cuban stole my idea! (But actually it does sound interesting – might have to check out that blog)

            I think the hard part would be deciding where the chronic/catastrophic line would be. There would be constant pressure from interest groups to cover their pet thing (e.g. birth control) that would eat into the advantages over straight single payer.

          • Paul Brinkley says:

            Anything that tries to dislodge ESI is political suicide

            This sounds odd to me. In my conversations about health care management, the one thing that seemed to get agreement from both left and right was to equalize the tax rates on wages and non-wage benefits, so that employers would have no incentive to offer the latter in order to more cheaply attract labor, and then later cause employees to be so anxious about losing their job or even switching jobs (unless they were healthy).

            Is there some bloc of employers warning the government that they’ll outsource everything if tax rates are equalized, even if it’s revenue-neutral, and despite the headaches employers have to deal with when managing non-wage benefits?

          • John Schilling says:

            This sounds odd to me. In my conversations about health care management,

            Yes, policy wonks would like to get rid of ESI. That’s, what, 0.5% of the population?

            Is there some bloc of employers warning the government that they’ll outsource everything if tax rates are equalized,

            Not employers, employees. Repeat after me:

            “If you like your health care plan, you’ll be able to keep your health care plan”.

            For all the talk of how horrible the American health care system was before the ACA, and how Obama made it worse / didn’t make it much better, most Americans have pretty consistently gotten first-rate health care all along without having to worry or even understand how much it cost. And most of those, get their care through ESI. If you try to take that away, they’ll scream bloody murder just like the minority in the private market did when they found out Obama had been fibbing, but there’s I think five times as many of them.

            That’s the political suicide part. And wonkish explanations about how they will eventually be able to buy new health insurance plans that are better than the ones you took away, won’t save your political career.

    • skef says:

      I presume it stands for “Affordable (Heh) Care Act”?

    • Corey says:

      Not too hard to explain.

      Nobody including Congressional Republicans expected Trump to win. So they figured they could just throw one-line ACA repeal bills at President Clinton’s veto pen for a few years to satisfy the base / their own BS about the evils of ACA.

      Then Trump wins, and they have to actually govern.

      Since the problems Republicans actually have with ACA (it redistributes too much, from healthy->sick, rich->poor, and young->old) are opposite to the problems they and their base complains about (it doesn’t redistribute enough to make individual insurance actually usable, on the current level of subsidies), there’s nothing they can do.

      Hesitant as I am to claim 11-dimensional chess, it seems Ryan & McConnell are setting AHCA up to fail, then they can shrug and say “we tried to slay the evil ACA” and move on to tax cuts. (If they actually want it to pass, they’re way more politically inept than prior evidence would suggest).

      • shakeddown says:

        I don’t think they’re deliberately setting themselves up to fail, so much as they already maneuvered themselves into an unwinnable situation, and this is genuinely the best they can do.

        • Jordan D. says:

          I think this is true. And this is why you’re seeing so much conflict right now between the ‘conservative’ wing of the GOP (Rand Paul et al) and the majoritarian wing- Paul Ryan isn’t fool enough to think that their majorities would survive an actual straight repeal, but there’s no actual way to “keep the good parts of Obamacare and get rid of the bad parts” and there never was (there may be ways to substantially improve the law while keeping the popular parts, but that isn’t the GOP’s mandate).

          • Randy M says:

            The people in favor of repealing Obamacare are a mix of those against it for personal (ie, financial) reasons, ideological reasons (against expanding government) and practical reasons (fears of a “death spiral” etc.). The particular parts anyone opposes are going to be quite different depending on why, so getting a replacement that even just the republican base approve of is going to be probably impossible.

          • cassander says:

            I actually think the consequences for repeal would be pretty small. the number of people getting healthcare through the exchanges that weren’t getting insurance before the bill was passed is tiny. the percentage that couldn’t get insurance (as opposed to just weren’t for any of several reasons) is even smaller. If they repealed the exchanges and subsidies but left the medicaid expansion alone (or just block granted medicaid without specifically repealing the expansion) you’d eliminate about half the expense and most of the problems. Granted, that means you’d have to repeal the ban on pre-existing condition bans, which opens the republicans up for attacks, but when the repeal ends up affecting almost no one, that probably doesn’t stick much.

            That said “consequences” and “political consequences” are not the same, and politicians care overwhelmingly about the latter, not the former.

        • Corey says:

          The best they could do would be to just repeal the taxes ACA enacted, touching nothing else.

  26. shakeddown says:

    If I had to choose a high point for the history of the human race thus far, it would be December 24, 1968.
    And then the murders began.

    (Not technically the first line, but works better this way).

    • dndnrsn says:

      You’re talking about the rise in murders in the US, right? If you are, I don’t know if you can peg it to the end of 1968. The UCR has the overall violent crime rate going from 160.9 per 100k (5.1 homicide, 9.6 rape, 60.1 robbery, 86.1 aggravated assault) to 298.4 in 1968 (6.9, 15.9, 131.8, 143.8).

      • shakeddown says:

        Okay, that’s interesting. Can we attribute the steady rise in crime from the sixties to the nineties to the increase in leaded fuel driven miles, in the same way that we attribute deleadification to the subsequent drop?

        (Also just to make sure, I was unsonging this meme).

        • Douglas Knight says:

          Crime didn’t rise steadily for 30 years. Homicides doubled in 10 years, from 1965 to 1975, unbelievably smoothly, but way too fast to be lead. For the 20 years after that there were ups and downs, but pretty much flat. Other countries were similar, sometimes at different times, sometimes a bit slower.

          • dndnrsn says:

            Murder has the advantage of a measurement that it’s hard to miss. Of the four things categorized as “violent crime”, rape is the least likely to get reported, robbery and aggravated assault are in the middle, murder is the most likely. However, the disadvantage of murder as a measurement is that better medical care turns murder into aggravated assault. In the 20th century, at least, advances in emergency medical care tended to follow wars. By 1975, some of the advances in emergency trauma care that came out of the Vietnam war had probably trickled into ERs. Today the murder rate is lower than 1960 – but the aggravated assault rate is a little short of 3x as high.

            By 1965, the overall violent crime rate was 25% higher than it was in 1960. Murders were not up, but rapes, robberies, and aggravated assaults were. A 25% increase is pretty big, so something clearly was already happening between 1960 and 1965.

            Of course, it’s not necessarily that 1960 was a golden age. Statistics before that – probably less reliable – show the murder rate being around 1.5 at the beginning of the 20th century. It increased significantly to a peak in the early 30s about the same as the 1980 peak, then dropped again, rose a bit after WWII, dropped again, then rose again during the 60s and 70s. This is not considering the murder rate-suppressing technological advances – the murder rate would have been even lower in 1900 had they the care of 1920, 1950, 1980, etc.

          • JayT says:

            Like you say, rapes are the least likely violent crime to be reported, though I would guess that the reporting rate has probably gone up throughout the years due to changes in how society looks at things like rape. The same is probably true of assault. People are less likely to handle things on their own than they used to be.
            I wonder how much of the change between 1960 and 1965 could be attributed to the changing views of society, so more crimes were reported, not necessarily committed.

          • shakeddown says:

            On the subject of unreported crime, were there periods (either in the sixties or earlier) where crime (including murder) in black communities would go unreported because of segregation?

          • keranih says:

            periods where murders were unreported due to segregation

            Unlikely. A dead body was still a dead body, and it’s not like these people were being killed by blowfish toxin or massive injections of insulin – they were being shot, stabbed, and beaten to death.

            One might suppose a body could be abandoned in the woods and never found, or not found until too far gone. But it is not reasonable to assume this happened frequently enough to shift the homicide rate.

            What may have happened is that the death of a young colored (to use the language of the time) man – possibly with few ties to the local community – was not investigated by the local (white) police with much vigor. (The reasons for this varied from ‘good riddance’ through apathy though ‘let them police their own, that is more just’ to ‘even if we try, we’ll just be stonewalled’.) But a failure to find the shooter is different than noting the bullet holes.

            Having said that – record-keeping of major events (ie, birth certificates, marriages, etc) in poor communities did lag behind that of wealthier communities. And deliberate police mis-classification of major crimes is hardly a phenomenon confined to a generation past.

    • powerfuller says:

      How about the 2nd century in the Roman Empire? Gibbons was a fan of that one.

      For high points of the human race, I usually go with July 21, 1969. Sure there were murders back on Earth, but the moon was (and still is?) murder-free!

  27. aakumar says:

    I don’t have much of a stats background and given the frequent discussion of medical statistics here, I’m wondering if anyone here happens to know the correct technique or google search term I should use for a problem I’m working on.

    I have a data set with a individuals randomly assigned to a control group and various treatments, plus a ton of other variables at the time of segmentation, as well as a few outcome variables of interest. It’s easy to tell that the treatment has significant effects on the outcome variables in aggregate, which can simply be expressed as a ratio of (Test Outcome)/(Control Outcome).

    However, what I would really like to understand is if and how some of the various variables at time of segmentation slope the T/C response to the treatment. If I were looking at a binary outcome I’d use Somers’ D, but I’m not sure what the best metric or technique to use on the target variable of T/C ratio is (in fact, I’m not really sure what the best google search terms to get started are).

    Any suggestions would be hugely appreciated!

  28. deluks917 says:

    Does slatestarcodex have a discord server? Discord is a really convient alternative to IRC. The current SSC irc is rather inactive. Maybe a discord would have more discussion.

    If someone established wants to set up a discord that would be ideal. However for now I set up this discord. Just click the link to join!

  29. Sniffnoy says:

    Scott: You might want to update your Gene Expression link. The new site is here:

    (Although unfortunately the new site doesn’t have the extensive GNXP archives, you still have go to Unz or one of the older sites for that)

  30. bean says:

    I volunteer as a tour guide at the USS Iowa, and I enjoy explaining battleships so much that I’ve been doing posts to explain them to people here too for the past couple of OTs. This one is on non-US battleships in World War 2. Most recent post (armor) and US battleships in WW2.

    Non-US battleships largely fulfilled the same roles of screening, shore bombardment and surface superiority that the US battleships did, although in very different proportions. Carrier screening is much less prominent in the tales of British and Japanese battleships, and is totally absent from that of the Germans and Italians (due to a lack of German and Italian carriers), while surface superiority dominates. Much of this is due to the nature of the European war, where short ranges meant that land-based air power could project power.

    The Japanese carriers were only ever screened by the Kongo-class battleships, as the other Japanese battleships were too slow for the role, and were being husbanded for the “Decisive Battle”. The British did use their battleships and battlecruisers to protect their carriers, both off of Norway (after the loss of the Glorious) and in the Pacific. Neither nation was able to reach US standards of AA fire, and the ships were included as much to protect against surface attack as against air attack.

    The British used their battleships for shore bombardment in much the same way the US Navy did. The fast battleships participated in the bombardment of the Japanese mainland described last time, while the older ships supported landings from North Africa to Normandy. Ships of both types bombarded various targets in the Mediterranean and off France. The Japanese only carried out a few bombardments with their battleships, most prominently the bombardments of Henderson Field on Guadalcanal, one of which lead to the battle between Kirishima and Washington. The Germans also did a few bombardments, one of which was the only offensive use of the main battery of Tirpitz.

    I’ll handle the surface components of the Atlantic and Pacific wars separately, starting in the Atlantic. The German fleet was not built with a clear strategic purpose, which greatly reduced its utility. The primary role that the surface ships found was commerce warfare, although the strength of the Royal Navy meant that they found limited success. On the other hand, the ‘fleet in being’ that the Germans maintained forced the British to hold a large portion of their fleet in home waters, and limited the aid they could send to Russia via the arctic convoys due to the need for heavy escorts.

    The few occasions when the Germans did send heavy ships to sea had mixed results. Admiral Scheer sunk 113,000 tons of shipping during her 5-month sortie, although that was approximately 10% of shipping losses during that period, and she was the most successful of their raiders. Scharnhorst and Gneisenau also raided the Atlantic, sinking 22 ships. The total would have been much higher if several of the convoys they found had not been escorted by battleships, which they had orders not to engage.

    Bismarck was on her way to raid convoys when she was sunk, an action that bears closer examination. I’m sure everyone still reading knows of the Denmark Strait, where Bismarck sunk Hood after only 10 minutes of action. What is not widely known is that Hood was as well-protected as the old battleships of the Queen Elizabeth and Royal Sovereign classes, and the Admiralty warned their commanders to beware in case they suffered the same fate. The loss of Hood is commonly ascribed to insufficient deck protection, although a recent examination of the wreck suggests that the fatal shell in fact went under the belt and set off the 4” magazine. Hood’s destruction, and the loss of all but 3 of her crew of 1,418, badly shook the British. However, the other battleship in the action, Prince of Wales, although so new that her guns were still having serious teething problems (only firing about 75% of the theoretical rounds possible), managed to hit Bismarck three times, causing flooding and loss of fuel which ultimately helped doom Bismarck.

    A carrier strike from Victorious the next day did more damage, but the result was still not fatal, and the British briefly lost Bismarck. Torpedo bombers from the carrier Ark Royal were sent after her when she was re-acquired, but accidentally attacked a shadowing British cruiser. The magnetic exploders failed, and a second strike with contact fuses was successful, with one torpedo taking out Bismarck’s rudder.

    The British closed in with the battleships King George V and Rodney. During this action, Bismarck did no damage to her opponents, but took at least 300 hits from the British, a mix of main battery and secondary battery shells from the battleships, as well as fire from the accompanying cruisers and destroyers. Within 30 minutes of the action opening, Bismarck was totally disabled, and the British closed to point-blank range to continue pounding her. It took another hour to finally sink her, with continuing debate over exactly what was responsible for her actual sinking. (The answer is that the British sank her, regardless of what actually destroyed the last vestiges of her watertight integrity.)

    Scharnhorst and Gneisenau had returned to France after their last Atlantic cruise, and later made a dash through the English Channel to return to Germany. They avoided taking damage from British attacks during this operation, but were mined in the North Sea before they reached their bases. Gneisenau was destroyed by bombers before she sailed again, and Scharnhorst joined Bismarck’s sister Tirpitz in Norway. Tirpitz spent almost all of her career under British attack, both from bombers and frogmen, before being sunk in 1944. Scharnhorst sortied against a convoy in December of 1943, and was sunk by the British battleship Duke of York.

    In the Mediterranean, the British battleships saw extensive service, as did those of Italy. A few battles are of interest to us.

    The first of these are the attacks on Mers-el-Kebir and Dakar. After the French Armistice, the British demanded that units of the French fleet which were in Africa swear allegiance to the Free French, as they feared that the Germans would get the ships. When the fleets in Algeria and Senegal refused, the British attacked. The action in Algeria went well, sinking one of the French battleships and damaging two of the other three with minimal losses. The attack on Dakar was less successful, with the British battleship Resolution nearly sunk, and the French forces mostly surviving.

    The Italian battleships spent the war (at least until the Italian surrender in 1943) trying to disrupt British convoys in the Mediterranean, usually unsuccessfully. Occasionally they fired at the British from long range, and usually came off badly.

    In November of 1940, the British carrier HMS Illustrious launched 21 torpedo bombers at the main Italian fleet base at Taranto, which held 6 battleships. One was sunk, and two others heavily damaged, greatly aiding British efforts to remain in the Mediterranean. This attack served as an inspiration for the latter attack on Pearl Harbor.

    The only major battle of interest in the Pacific not covered last time was that of the ill-fated Force Z, composed of Repulse and Prince of Wales. Based at Singapore, they were sent north to attack the Japanese landings in Malaya. Insufficient air cover resulted in the loss of both ships to torpedo attack. This was significant because, even after Taranto and Pearl Harbor, the advocates of the battleship were able to point out that no battleship that was underway and alerted had ever been sunk. In many ways, the end of the battleship as the master of the sea is more properly dated to December 10th, 1941 than to December 7th.

    As an aside, after the underwhelming response to Part 2 of Armor, I may wrap the series up here unless I get requests to continue.

    • James Miller says:

      Do you have an opinion on whether aircraft carriers have been made obsolete by hypersonic missiles similar to how battleships were made obsolete by aircraft carriers?

      • bean says:

        Probably not. Hypersonic missiles are expensive, and hard to target. Also, they can be shot down. The aircraft carrier is probably less obsolete today than it was in, say, the early 80s, due to the massive improvements in air defense technology since then. Anti-ship missiles haven’t gotten that much better during that time.

      • cassander says:

        The best answer I have for this question is the Bryan McGrath’s. Anything that makes aircraft carriers obsolete almost certainly makes land based aircraft obsolete as well, at least those with less than continental range. If you can reliably kill an aircraft carrier sitting, e.g. 500 miles away, you can kill a runway that far out as well. More easily in fact, because the runway doesn’t move and is therefore much, much easier to target.

        • Douglas Knight says:

          Sure, but people who say carriers are toast also say that forward bases are toast.

          • cassander says:

            That’s not been my experience. I often hear talk of carrier killer missiles sinking carriers. I rarely hear talk of runway destroying missiles blasting air bases, even though that’s a lot easier, at least outside of the taiwan scenario.

          • Douglas Knight says:

            Well, I know what James reads, and that’s what it says. For example

          • On the other hand, carriers are a lot more expensive than runways, making risking them more of a problem.

          • cassander says:


            An airport isn’t just a runway, it’s runways, taxiways, air traffic control, hangars, repair facilities, terminals, fuel dumps and fueling equipment etc. All that together isn’t that much cheaper than a carrier..

          • Chalid says:

            An airport isn’t just a runway, it’s runways, taxiways, air traffic control, hangars, repair facilities, terminals, fuel dumps and fueling equipment etc. All that together isn’t that much cheaper than a carrier..

            I find this hard to believe – could you quantify it? Google says the latest carrier cost $13 billion. I couldn’t quickly find military airport construction costs, but Denver International was $4.8 billion (in 1995) and it is orders of magnitude larger than a carrier.

          • gbdub says:

            The Nimitz class carriers were about $4.5 billion a piece. The latest one is more expensive as it’s the first of a new class (I think the planned final unit cost will be lower than $13 billion).

          • bean says:

            Ford is way more expensive than she should be, both because a lot of R&D got rolled in and because the program has suffered from poor management and long build times. I expect repeat units will be approximately half the price. It’s more expensive than an air base, but as I’ve pointed out, it can be anywhere in the world in a couple of weeks. Even golf course-less airbases (which are considered hazardous duty by the USAF) take a lot longer for comparable capability.

          • Chalid says:

            OK, but how expensive is an airbase?

          • bean says:

            OK, but how expensive is an airbase?

            That depends very heavily on your cost accounting. I’m aware that’s not helpful, but this is a really complicated question. I have a fair bit of documentation on USAF construction costs (don’t ask) but looking through it, none of it is set up to answer the question “how much does it cost to set up a new airbase?”
            But there’s more than one way to skin a cat, even though I don’t like this one much, either. We can go back to when we were doing base development, and see what it cost then. Unfortunately, that was the 1950s. So I pulled up a copy of Selection and Use of Strategic Air Bases, the report of a study which was set up to figure out the cheapest way to attack the Soviets with B-47s. According to their estimate, an overseas base capable of operating a wing of medium bombers would cost about $60 million.
            Running this against inflation, we get $540 million today. The base in question is probably broadly comparable to a carrier. To try to dial out military inflation, let’s compare it to 1950s carriers, too. The USS Independence was ordered in 1954, the year of the report, and cost about $180 million. But wait, you say, she was a dinosaur-burner, and nuclear carriers are much more expensive. Very well. Enterprise cost $540 million, and was ordered in 1957. However, Constellation (ordered 1956) is listed as costing $264 million. It’s been a while since I read the relevant Friedman, so I can’t say what was going on with carrier costing then or how much nuclear power is driving up cost, but I suspect that if carrier cost and base cost have tracked, we’re looking at a differential of 4-8 in terms of cost/unit of basing.
            Note, however, that the carrier can easily substitute for 4 bases in different parts of the world, and that you need land for the bases in the first place. Which you may or may not have, depending on geography. There’s water near almost everything important.

          • cassander says:


            Gdub is right, the Denver airport cost basically the same as the Ronald Reagan, which was ordered in 1998. Just a raw inflation adjustment will bring that cost up to over 7 billion.


            Half is awfully optimistic. the official budget for CVN-79 is 11.4 billion. IIRC, that’s for a 6 year build. If we went to 5 it would get cheaper, but not to half.

          • bean says:

            @ Cassander
            My bad. I assumed that we’d solved the problems with Ford, and were doing sanity again. You’re right. Kennedy is going to be nearly as expensive. Maybe we need to get whoever was in charge of the Virginia program and put him in charge.

          • cassander says:


            If you want true insanity, Columbia (the first nuclear missile sub for those not familiar) is expected to cost 14.5 billion.

          • bean says:

            I think that probably includes the first-of-class stuff, which always drives up price. Ford costs $12 billion and change with that stuff taken out.
            Also, SSBNs are always crazy expensive. I haven’t looked closely at why, but both the USN and RN lost lots of programs to pay for theirs.

          • cassander says:

            it does, but that’s only 5.7 billion. just construction is almost 9.

        • Protagoras says:

          A runway is much cheaper and easier to build, much cheaper and easier to repair, and is never going to be sunk so repairing it is always an option (or just building another one, since, again, much cheaper and easier to build). So despite their being slightly harder to hit due to moving around, I don’t see how aircraft carriers are less vulnerable in a practical sense.

          • Chalid says:

            Is it also a significant advantage to be able to use land-based aircraft and pilots who don’t need to be trained to land on a carrier? Or is this not a big deal?

          • bean says:

            A carrier has the big advantage of being able to move. Sea surveillance is really hard to do right, even though it’s conceptually pretty simple. (Seapower and Space is the best book I’ve read on this, but it’s hard to find.)

            Land-based air has a lot of advantages over carrier air. The training standard of the pilots is lower (although this can bite you if the pilots aren’t trained for maritime missions, which is a special skillset), and the airplanes are more available. But carrier air is the best way to put a forward base somewhere you don’t have one, and it usually comes with a very good air defense suite. There’s a reason we have both.

          • cassander says:

            An airport isn’t just a runway, it’s runways, taxiways, hangars, repair facilities, terminals, fuel dumps and fueling equipment etc. All that together isn’t all that much cheaper than a carrier. And while it can’t be sunk, it can be blown up and mined, put out of action for months or years.

            And you’re underestimating how much harder it is hitting a moving target. An airport is a known location, you can study it for months then blow it up with GPS or inertia guided weapons that are hard to spoof or jam. To go after the carrier, even if you know roughly where it is, you need a weapon that can see the carrier, distinguish it from its escorts, the sea, and other noise, home in on it, and then blow up all while the carrier’s ewar does its best to trick you. It’s a much, much harder job.

          • Civilis says:

            An airport isn’t just a runway, it’s runways, taxiways, hangars, repair facilities, terminals, fuel dumps and fueling equipment etc. All that together isn’t all that much cheaper than a carrier. And while it can’t be sunk, it can be blown up and mined, put out of action for months or years.

            On the flip side, all those airport components on land can be spread out and individually hardened so one missile (short of a nuke) can’t take out more than one component.

            If you lose a carrier, you automatically lose all the runways, all the fuel and ammo facilities, all the C&C facilities and all the crew and support functions.

          • cassander says:


            One, spreading things out increases the size of the area you need to defend, makes operation of the facility slower and less efficient, and raises costs. That’s not to say that it’s never worth it, but it’s not a free lunch.

            It’s true if you manage to sink it, you lose all those things, but carriers are hard to sink. With one weapon you’re much more likely to mission kill a carrier and force it to retreat than you are to sink it. And that’s little different than knocking out a runway but not blowing up your secondary facilities.

          • bean says:

            On aircraft carriers vs land bases, a couple of points:
            1. Striking at naval targets is not the same as striking at land targets. Normal land-based strike units are seriously handicapped over water.
            (For an example, although this is less likely to happen today, the British used to monitor Italian snooper transmissions. If their reported position was far enough off, policy was to not bother shooting them down. This was common.)
            2. A carrier is mobile. A land base is not. This is important strategically. Base facilities can’t be built quickly (do you have any idea how long it takes to get the grass on a golf course right?), and while carriers are even slower, they can be moved around quickly. And you have everything you need right there.
            3. A carrier comes with a really good AD system, probably better than any on land.
            4. Sea surveillance is hard to do. The Soviets weren’t able to crack it reliably. There have been improvements in technology which might make it easier, but it’s still a really serious hurdle.
            5. Carriers move on a tactical level, too.

          • cassander says:

            (do you have any idea how long it takes to get the grass on a golf course right?)

            You joke, but as a project for work a couple years ago I had to look up the physical addresses of the headquarters of a number of Air Forces around the world. My first resource, of course, was google maps. It’s astonishing how many of them were adjacent to golf courses.

          • keranih says:

            Most of those golf courses were actually on the military installation. The military golf courses were open to military members at a very low use fee, regardless of race or social class, and enlisted men could use the courses as well as the officers.

            (Which is not to say that there were not strict social rules and biases, per my grandfather, but the civilian courses in the South would exclude Negroes and the ones in the North would exclude enlisted men, and out west they’d refuse to waste the water. Or so my grandfather said.)

    • the anonymouse says:

      Not sure if it’s an underwhelming response so much as people like me who are reading, waiting for installments, but don’t have much to contribute to the conversation. Thank you for the ongoing history lesson!

      Actually, if I did have a question, it’s echoed a little above: what’s the probability that aircraft carriers are made obsolete by what I would imagine would be guided-missile submarines? Something that can shoot and hide, packed to the… ahem, gills, with rows and rows of antiship missiles?

      • bean says:

        Not sure if it’s an underwhelming response so much as people like me who are reading, waiting for installments, but don’t have much to contribute to the conversation. Thank you for the ongoing history lesson!

        You’re welcome. It’s mostly my ongoing need for validation…
        (Well, a little bit. I don’t want to waste my time on these.)

        Actually, if I did have a question, it’s echoed a little above: what’s the probability that aircraft carriers are made obsolete by what I would imagine would be guided-missile submarines? Something that can shoot and hide, packed to the… ahem, gills, with rows and rows of antiship missiles?

        I have to echo my earlier response to this. The SSGN is an interesting concept, but after much study the US has given up on sub-launched anti-ship missiles. They’re hard to target, and you want to make sure that the thing you’re shooting at is a carrier and not a neutral merchant ship. And they’re a lot easier to shoot down than torpedoes. The advances of the last 30-35 years have been a net win for the surface ship over all contenders.

        • the anonymouse says:

          Neat! I wouldn’t have thought so, but I have no expertise outside of World of Warships.

          That is, no expertise.

        • James Miller says:

          The advances of the last 30-35 years have been a net win for the surface ship over all contenders.

          I would love for you to write further on this.

          • bean says:

            It’s fairly simple. The naval air defense systems in use in the early 80s were based on rotating radars and required that each missile be guided by a dedicated tracking radar. Due to electronic interference, it wasn’t possible to fit more than two radars to each end of a ship, for a maximum of four. (Missile ships were classified as single-end or double-end. Each end had a twin launcher.) So you had a maximum of four missiles in the air at once. It typically takes two missiles to shoot down a target. So no ship can engage more than two targets at once. Later missiles might let you double up on use of tracking radars, so you can have 4 targets at a time. Now, each launcher fires about once per minute, with two missiles, but you’re usually limited by how many guidance radars you have. Even a long-range missile might only get 10 salvos off per end, for 20 targets killed per ship. More typical systems, more like half that.
            Now, enter AEGIS. Your main radar is phased-array, so it searches practically instantly across the whole area around you. The missile has an autopilot, so it only needs a tracker for the last few seconds of flight. It’s updated by the ship, based on information from the main radar, which is accurate enough to get the missile very close. Also, you now have vertical launch systems, which means that you can fire the missiles basically as fast as you want. Maybe a 1-2 second gap per end. Now you can take out one target per end every 10 seconds (or less), regardless of range. Oh, and the modern missiles are more accurate than the old ones, too.

            I don’t have as good of an quantification on submarines, but modern electronics have (I think) been improving sonar faster than submarine quieting has been getting better. Obviously, hard numbers are classified, and all of my books on this are at least 10 years old. (No, this isn’t by choice. Norman Friedman has been busy with other things.)

      • cassander says:

        I agree with bean on this one. submarines are stealthy and deadly, but they gain those qualities at the expense of long range targeting. SSGNs work best where they can attack targets of known location, so they can pop up (to whatever depth they can fire at) fire off their missiles, then go back down. that’s a lot harder to do against a maneuvering target that you can’t see than a fixed target on land. if I want to go after ships, I’d much rather use a torpedo that can’t be jammed (they’re often wire guided, and the russians use wake homing) and that will strike the target underwater where they do a lot more damage.

        • James Miller says:

          bean and cassander

          For navy vs navy battles why doesn’t this all favor ( submarines + satellites or subs+ drones) vs surface ships where you spot the enemy, have the sub pop up and fire a few missiles that are guided with the help of the satellite or drone? Is it because the missile could be jammed or shot down? My (relatively uneducated) guess would be that it’s harder to shoot down a missile than to have a missile hit a surface ship.

          • bean says:

            That’s really hard to do, actually. The satellite system is non-trivial, and only somewhere between 1 and 3 countries can reliably do it. Even at their height, the Soviets had trouble making it work. Sea surveillance is hard. Then you have to talk to the submarines. Submarines are good at lots of things, but talking is something they are really bad at. And shooting down missiles isn’t that hard. Missiles tend to have predictable courses, and modern SAMs are really good. EW is really complicated, and we can’t say how it would turn out without actually going to war.

            My (relatively uneducated) guess would be that it’s harder to shoot down a missile than to have a missile hit a surface ship.

            Shooting down a missile may be harder than hitting an undefended surface ship, granted. But shooting down a missile isn’t hard enough that it can’t be a good strategy against missiles. We’ve seen recent examples off of Yemen, and the anti-ship missiles lost decisively.

          • gbdub says:

            And Aegis is getting increasingly good at taking out even long range ballistic missiles (I help design the targets)

    • dndnrsn says:

      Any chance of an installment looking at battleships post-WWII? The way they’ve been used in the period where it was totally obvious carriers were the new hotness is interesting but I don’t know much about it. They didn’t get rid of them, so obviously there’s some reason to keep them around.

      • bean says:

        I could definitely do that. Of course, I’d probably spend most of it talking about everyone but the US, as that’s just the story of the Iowas, and it’s not particularly long on the scale I work.
        Actually, I’d spend most of it examining why the US and Britain looked at maintaining their battleship forces after the war. Which would be interesting. Thanks for the idea.

    • wavey davey says:

      I am loving this series! I have some knowledge about Royal Navy capital ships, so I’ll add a few comments here (mainly on battleships, a few carrier references).

      The British Royal Navy had some great battleship designs, notably the Dreadnought which ‘made all previous battleships obsolete’, and the Queen Elizabeth class, which replaced inefficient and labor-intensive coal with oil as a fuel source. With 15in guns and 24kt speed the QE class seems years ahead of its time (being described as the first “fast battleship”). Four of the five ships survived not just WWI but front-line service in WWII, being heavily involved in the toughest years for Britain in that war. Unfortunately, they fought in few true battleship-on-battleship encounters, so it’s a bit hard to determine their true strength in comparison with other designs.

      Other Royal Navy directions were less successful. The battlecruiser, as envisioned by Jackie Fisher didn’t have a happy career, losing ships in devastating explosions at Jutland and against the Bismarck. There are arguments that this wasn’t an issue with the ships, but how they were used – in theory the battlecruiser was not meant to be a line-of-battle ship used against other capital ships, instead against cruisers and smaller warships. In reality, they got used as a kind of scout ‘vanguard’ for the main fleet which put them in range of other battleships and battlecruisers, with unhappy results.

      I have a sense, in both wars, that the RN had one of the largest fleets, even if didn’t always have the best or most modern ships. And that the RN used their ships aggressively and were accepting of risking them being sunk, sometimes in tragic, if not absurd circumstances, where other navies guarded and kept their fleets in port as much as possible. Some of the oddest “oops, we just lost our ship” incidents include:
      – Sending a bunch of (admittedly obsolete) pre-dreadnoughts into mined Turkish waters during the Gallipoli Campaign, and having three of them sunk and three more damaged by mines in a few hours.
      – 25 years, almost to the day, after discovering that sending battlecruisers against capital ships was a bad idea, that leads to ships exploding with tremendous loss of life (Jutland, May 1916), an essentially identical battlecruiser was sent against the most modern ship in the German navy (Hood vs. Bismarck, May 1941). The result: the ship exploded with tremendous loss of life.
      – The aircraft carrier HMS Glorious, doing an emergency evacuation in hostile waters from the greatest military threat to Britain in modern history, decides to separate from other ships so it can perform a court-martial because the Captain has had an argument with one of his subordinates. The ship has no combat-air-patrol, no proper look-outs and is running on only half its boilers. The resulting interception is one of the rare occurrences when an aircraft carrier got sunk by a battleship.

      Okay, so before I feel too guilty making light of awful events where brave people did the best they could in difficult circumstances, I want to highlight the Mediterranean Campaign in WWII. There the Royal Navy capital ships were victorious in a number of individual actions, but also made a major contribution to the overall strategic success in the region. The North Africa campaign was won because men and equipment could be supplied to the Allies, and prevented from reaching the Axis forces. Keeping Malta as a viable base and preventing the Italian Navy from being a credible threat cost ships (Barham, Ark Royal, Eagle) but was probably an example where a bold strategy paid off over a more cautious one.

      • bean says:

        The British Royal Navy had some great battleship designs, notably the Dreadnought which ‘made all previous battleships obsolete’,

        I’m not sure that Dreadnought was all that great of a design overall. The wing turrets were just awkward. South Carolina, designed at the same time, had a much better armament layout. The building of Dreadnought, on the other hand, was heroic. It was revolutionary, but that revolution would have happened elsewhere if Fisher had never been First Sea Lord.

        and the Queen Elizabeth class, which replaced inefficient and labor-intensive coal with oil as a fuel source. With 15in guns and 24kt speed the QE class seems years ahead of its time (being described as the first “fast battleship”).

        The QE was a great design, but I feel compelled to point out that we are again seeing better British build rates. The Germans, for instance, went to 15″ guns before they knew the British would. Also, the Nevada class was designed for oil-firing well before the QEs were converted. (Common claims that they were designed that way are untrue.)

        Unfortunately, they fought in few true battleship-on-battleship encounters, so it’s a bit hard to determine their true strength in comparison with other designs.

        I can’t think of another class we have better information on. And they held up very well at Jutland.

        There are arguments that this wasn’t an issue with the ships, but how they were used – in theory the battlecruiser was not meant to be a line-of-battle ship used against other capital ships, instead against cruisers and smaller warships.

        This is a common theory, but wrong. Friedman clearly shows that Fisher clearly intended them to take part in the line of battle. That’s why they got 12″ guns instead of the 9.2″ or 10″ originally proposed.

        In reality, they got used as a kind of scout ‘vanguard’ for the main fleet which put them in range of other battleships and battlecruisers, with unhappy results.

        The unhappy results at Jutland were the result of suicidal magazine practices intended to increase rate of fire, not a flaw with the ship or concept. Lion’s gunner didn’t like said practices (taking the powder out of the storage cans ahead of time so they didn’t waste time doing it in battle), and she survived a turret hit without the magazine going up. Hood was lost to a freak hit that I don’t think it is fair to judge the concept on.

        – 25 years, almost to the day, after discovering that sending battlecruisers against capital ships was a bad idea, that leads to ships exploding with tremendous loss of life (Jutland, May 1916), an essentially identical battlecruiser was sent against the most modern ship in the German navy (Hood vs. Bismarck, May 1941). The result: the ship exploded with tremendous loss of life.

        Calling Hood ‘essentially identical’ to Queen Mary (much less Invincible and Indefatigable) is nonsense. Hood was essentially a 30-kt Queen Elizabeth, and her design had been extensively modified after Jutland. When she was sunk, the QEs and Rs were told that they were just as vulnerable to whatever had killed her.

        I’m with you on the RN’s performance in the Med. From what little I know of it (I have a copy of Roskill I need to read when I get time) they did a fantastic job.

        • wavey davey says:

          I’m not sure that Dreadnought was all that great of a design overall. The wing turrets were just awkward. South Carolina, designed at the same time, had a much better armament layout.

          The wing turrets are unfortunate I agree. I credit the USN with introducing super-firing turrets early (and later on for standardizing on the triple-turret rather than the quad-turrets in the KGV class).

          Calling Hood ‘essentially identical’ to Queen Mary (much less Invincible and Indefatigable) is nonsense. Hood was essentially a 30-kt Queen Elizabeth, and her design had been extensively modified after Jutland.

          Ah, somewhere along the way I assumed the Hood was basically an improved version of earlier RN battlecruisers with some more deck armor added to the design after Jutland. I didn’t realize how different the design was.
          I still think it’s odd that British ships with “battlecruiser” in the class name suffered from ammunition explosions after gunfire hits so disastrously. I recall (it’s either in “Castles of Steel” or “How to lose a war at sea” – another good read) that at Jutland German hits penetrated British armor in circumstances where the reverse wasn’t true, and that the Germans generally put more emphasis on the survivability of their designs. Although agree the ammunition-handling problems were also a big part of the disaster at Jutland.

          • bean says:

            Ah, somewhere along the way I assumed the Hood was basically an improved version of earlier RN battlecruisers with some more deck armor added to the design after Jutland. I didn’t realize how different the design was.

            The British battlecruiser went through several iterations, each quite different from the previous one.

            I still think it’s odd that British ships with “battlecruiser” in the class name suffered from ammunition explosions after gunfire hits so disastrously.

            It is, but such things happen in history. If Hood hadn’t exploded, I doubt that the battlecruiser name would be nearly as beaten up as it is. The situation is made particularly bad by the fact that most ascribe all the battlecruiser losses to inadequate deck protection, which is almost certainly untrue for every single one of them. The big lesson of Hood is probably ‘don’t use cordite’.

            I recall (it’s either in “Castles of Steel” or “How to lose a war at sea” – another good read) that at Jutland German hits penetrated British armor in circumstances where the reverse wasn’t true, and that the Germans generally put more emphasis on the survivability of their designs.

            I’d have to check Campbell for details on armor penetration, although the Germans did have somewhat better shells. There was more emphasis on survivability on the German side, and a lot more armor relative to size. That said, the British concept that warships take a lot of killing was proved to work reasonably well in the cases where catastrophic hits didn’t happen.

            Although agree the ammunition-handling problems were also a big part of the disaster at Jutland.

            The only other major contributor was cordite, and that didn’t sink Lion.

      • The original Mr. X says:

        – Sending a bunch of (admittedly obsolete) pre-dreadnoughts into mined Turkish waters during the Gallipoli Campaign, and having three of them sunk and three more damaged by mines in a few hours.

        IIRC what happened there was that the British had already swept that area for mines the previous day, then the Turks snuck out at night and re-mined it. (At least I’ve heard that happened at Gallipoli, and I assume it’s the same incident you’re describing — how many mine-related disasters can one navy have during the course of one battle?)

        • bean says:

          Not quite. Apparently, the Ottomans had expanded their minefield, but it was about 10 days earlier, and the British just failed to spot them.

    • John Schilling says:

      From your mention of the Admiral Scheer, you almost seem to be counting the Panzerschiffen as battleships, which is obviously controversial. Nomenclature aside, what do you think of the concept and the execution – and do you want to give the rest of the audience that part of the history lesson or shall I?

      • bean says:

        I’m not. That was an illustration of the magnitude of the surface raider threat, because S & G did pretty badly. They weren’t battleships, but it highlights the need for battleship escorts on the RN side.
        (On mobile, will explain more on the concept later. Or you can take it if you want.)

        • Protagoras says:

          S&G were ordered not to engage enemy battleships, and so were unable to attack convoys which had battleships protecting them. They did have only 11″ guns, but they were two battleships working together. I’ve encountered arguments that hits above deck by basically any size battleship gun will fairly rapidly cause enough damage to relevant systems to mission kill even a battleship (one person cited 20 hits as if it were a magic number). Any thoughts on how likely it is that the two of them working together could have incapacitated a single enemy battleship without either of the two being crippled? In other words, should they have had more aggressive orders, or was the caution justified?

          And, relatedly, if there’s any merit to the argument that it’s not all about penetrating the main armor belt, because hits to less armored or unarmored areas can add up to a mission kill, should battleship designers have experimented with designs that carried slightly greater numbers of slightly smaller guns? How effective would a final generation dreadnaught with, say, 4 quadruple turrets with a total of 16 11″ or 12″ guns be compared to the 8-9 15-18″ guns that the actual battleships of the final generation had?

          • bean says:

            S&G were ordered not to engage enemy battleships, and so were unable to attack convoys which had battleships protecting them.

            I have a feeling that Scheer had similar orders, and she did a lot better. Although in fairness, some of that may have been a much longer range, due to being smaller (more range from a tanker) and having diesels.

            They did have only 11″ guns, but they were two battleships working together. I’ve encountered arguments that hits above deck by basically any size battleship gun will fairly rapidly cause enough damage to relevant systems to mission kill even a battleship (one person cited 20 hits as if it were a magic number).

            20 isn’t exactly a magic number, but it’s not far off, either. In some cases, it’s less than that. SoDak was beaten up pretty badly at Guadalcanal after 20-something hits, mostly from cruisers. PoW was roughly handled from half a dozen or so hits at Denmark Strait, and Bismarck was mission-killed early on in her final battle.

            Any thoughts on how likely it is that the two of them working together could have incapacitated a single enemy battleship without either of the two being crippled? In other words, should they have had more aggressive orders, or was the caution justified?

            I’m not really sure. When they went up against Repulse Renown off Norway, they didn’t fare particularly well. I’ve heard that this was due to an overcomplicated fire-control system that was guaranteed not to work in action. Dial that out, and I’d take the two of them against a single old British battleship.

            And, relatedly, if there’s any merit to the argument that it’s not all about penetrating the main armor belt, because hits to less armored or unarmored areas can add up to a mission kill, should battleship designers have experimented with designs that carried slightly greater numbers of slightly smaller guns? How effective would a final generation dreadnaught with, say, 4 quadruple turrets with a total of 16 11″ or 12″ guns be compared to the 8-9 15-18″ guns that the actual battleships of the final generation had?

            I’m not sure. Past a certain point, spotting seems to get harder with more guns. Evidence suggests this number is 12. In terms of contemporary evidence, the USN was thrilled to go from 12 14″ guns to 9 16″ on the North Carolinas. In theory, small guns seem like a really good idea, but nobody seems to have acted that way at the time. S&G were scheduled to get twin 15″ guns, and everybody else went bigger. I’m not sure how to reconcile this, but I’m reluctant to declare everyone back then wrong.

        • John Schilling says:

          I’ll leave the panzerschiffen specifically to you for now. On the general concept of heavy surface raiders, I think the Germans weren’t tactically aggressive enough to get fair value for their investment in rather expensive ships. If you are resolved not to attack heavily-escorted convoys, merchant raiders are more cost-effective and maybe just plain more effective.

          With heavy surface warships as raiders, yes, there is a strategic preference for not endangering them in combat with enemy battleships, and there is value to keeping them intact as a Tirpitz-esque fleet in being. But sometimes you don’t really have a choice. Near as I can tell, every time the Germans had a choice they ran from the fight, and did it ever work well for them?

          Battle of the Falklands, WWI. Scharnhorst and Gneisnau (the first ones) encounter a superior force at anchor, run for the sea, are chased down and sunk in a humiliating defeat. If they had pressed the attack, the worst possible outcome is that they are sunk in a glorious defeat against overwhelming odds, and it’s possible that they could have sunk the British battlecruisers before they were ready for battle.

          River Platte, WWII. Graf Spee inflicts heavy damage on three British cruisers and then runs away, only to scuttle herself and have her captain commit suicide. Worst case if she presses the engagement is the same outcome with better propaganda and the Exeter at least being sunk. Quite possibly she prevails and forces the British to detach a battlecruiser or two for a very prolonged and public chase.

          Denmark Strait: Bismark had the upper hand over Prince of Wales after the loss of the Hood, and would probably have sunk her had she pressed the engagement. And then returned to Norway, too damaged to continue her sortie, where she’d have served as a mythic uber-Tirpitz, a fleet in being that had “proven” it would take an entire squadron of British battleships to sink. And maybe sortied in company with the Tirpitz later in the war, and wouldn’t that have been fun?

          Operation Berlin: Scharnhorst and Gneisnau (the new ones) twice declined to engage large convoys protected by a single old battleship. Here at least there is the real possibility of the Germans suffering a more severe short-term loss if they had pressed the engagement, but both of those ships combined never accomplished much in the entire rest of their careers and ultimately the Scharnhorst alone would die trying to engage a modern battleship to prove her relevance.

          KMS Tirpitz, entire career thereof: Yes, a fleet in being has value, but the value of tying down British battleships in 1942 and thereafter is extremely limited.

          And yes, pressing the fight against capital ships means that even a victorious surface raider is going to be too badly damaged and too low on ammunition to continue her raiding. In the age of radio and aircraft, running away from an enemy naval force also likely ends a raider’s career. Did any of the German heavy raiders, after running from a fight with their peers, ever accomplish anything besides being chased down and sunk or being hounded into a port from which they would never really emerge?

          • bean says:

            That’s a very good point. I’m not 100% sure about Denmark Strait (PoW was fought rather well there, and deserves a better reputation than she has), but other than that, you’re correct. I think this is just a symptom of the fact that ‘the German is not an aquatic mammal’. Their lack of understanding of the nature of sea power in WWI is shocking, and I’ve seen nothing to indicate that they made sense of it for WWII.

          • Protagoras says:

            Sounds right to me, and similar to my criticism of the IJN’s battleship use in the early part of WWII. Aggressive use of battleships means suffering losses, but sometimes inflicting heavy damage on the enemy. Cautious use of battleships to avoid losing them does not reliably keep them safe, but is pretty reliable in guaranteeing that having them doesn’t do you much good.

          • cassander says:


            A significant reason for the japanese reluctance to commit battleships might have been desire to conserve fuel. For an elaboration of this idea, see here. Of course, this strategy eventually proved to be pennywise but pound foolish, but ultimately, that’s sort of the problem with the whole IJN effort in the first couple years of the war, a doctrine that (rightly, IMHO) calls for preserving forces to fight decisive battle combined with a culture that insisted on “not one step back” thinking that led to continual half-assed efforts that they should have either whole-assed or no-assed that bled away their initial advantages.

          • John Schilling says:

            (PoW was fought rather well there, and deserves a better reputation than she has)

            Her crew fought well, the ship functioned poorly. At the critical moment, PoW had five guns working, all forward, the Bismarck’s damage was tactically negligible, and the Prinz Eugen was at extreme torpedo range and closing. That’s a no-win situation for the British. Run, and accept bombardment from a battleship’s forward guns at 16,000 yards without reply. Turn to engage the Bismarck, and be outgunned two to one until the torpedoes hit. Engage her fellow Prince, and take full broadsides from a battleship.

            Leach was right to disengage, and Lutjens was wrong to do so. Particularly in hindsight, but I don’t think any other outcomes were credible.

          • beleester says:

            Somewhat related question – in most contexts, fighting something means to attack it. Why is it that, in a naval context, “fighting a ship” means commanding it?

      • cassander says:

        oh, good, this gives me an opportunity to sling one of my hobby horses. What you wanted for commerce raiding wasn’t the Panzerschiffen, but a usually reviled concept, the flight deck cruiser.

        The cruiser is usually reviled because it was a worse carrier than an actual carrier and worse cruiser than an actual cruiser. And this is true. And that’s why it’s perfect for commerce raiding. The biggest problem a commerce raider faces is finding convoys to hit and avoiding enemy surface and air units that can sink him. Having a very powerful gun armament isn’t very useful, because, as happened to the ~~scheer~~ Graf Spee, the enemy can just send two or three cruisers after you and you’re toast. You don’t need a very powerful gun armament to wreck convoys, so being a worse cruiser isn’t a huge handicap as long as you can avoid the enemy’s heavier units, which is where the flight deck comes in.

        A couple dozen long range fighters are exactly what you need for commerce raiding. They aren’t enough to sink heavy surface units, but your targets are escorts and merchantmen so that’s no issue. They can shoot down enemy scout aircraft sent to find you, particularly the float planes most cruisers of the time carried. You can use them to locate targets and convoys, then strafe or bomb the ones you can’t easily reach, or close on the ones you can. to attack with guns.

        It’s everything you need for commerce raiding and nothing you don’t, a vessel powerful enough to wreck convoys that takes vastly more resources to hunt down and kill than it costs.

        • bean says:

          I’m actually with you on this one. Flight-deck cruisers seem to be one of the concepts that never really got a fair chance.
          The Panzerschiffe never made much sense. I think it was just the Germans having no strategic concept at sea again.
          I should point out that it was Spee and not Scheer that was hunted down and sunk at the River Platte.
          (Did you, by chance, run across this concept in Tarrantry?)

          • cassander says:

            I’m actually with you on this one. Flight-deck cruisers seem to be one of the concepts that never really got a fair chance.

            It does seem they would have helped a lot with the CAP problem that plagued early carrier ops, and didn’t have all that much less capacity than the Independence class carriers. Might have had issues with larger late war planes, but so did a lot of earlier designs.

            I should point out that it was Spee and not Scheer that was hunted down and sunk at the River Platte.


            (Did you, by chance, run across this concept in Tarrantry?)

            I did not. I had no idea that existed until now. What a wonderfully odd corner of the internet.

          • bean says:

            It does seem they would have helped a lot with the CAP problem that plagued early carrier ops.

            This, I sort of have to disagree with. I’d think that the thing to do if you’re trying to provide CAP is to take off the turrets and turn it into a CVL. That’s basically what the US did with the Independence-class. A flight-deck cruiser’s guns would get in the way in that role, and I’m not sure the air group would be big enough. I’ll check Friedman on the concept when I get home.
            (I got a copy of US cruisers Monday!)

          • cassander says:


            I didn’t mean to imply that the independence weren’t better, I mean that having a few flight deck cruisers instead of normal cruisers would have helped with the cap problem compared to the situation that actually existed in 1942. The US navy almost built the flight decks, probably would have if admiral Moffett hadn’t died.

      • bean says:

        On Panzerschiffe:
        The Panzerschiffe were one of the oddities to come out of Germany in the interwar years. The treaty of Versailles limited Germany to ships of 10,000 tons and 11″ guns, which was fairly standard for a coastal defense ship. However, the Germans decided that they were going to build something else, and produced ships capable of 28 knots, with 2 triple 11″ turrets, and armored against something between 6″ and 8″ shellfire. The extensive use of welding saved weight, and the diesel engines were revolutionary, giving amazing range. Their performance was aided by lying about their actual displacement. Deutschland/Lutzow (the first) was 10,600 tons standard, Scheer was 11,550 tons and Graff Spee was 12,340 tons. The British dubbed them ‘pocket battleships’ and they caused panic among the Western navies. Supposedly, they were able to outfight any cruiser and outrun almost any battleship or battlecruiser. It’s rather difficult to figure out what the actual strategic concept behind them was, but they seem to have been optimized for commerce warfare against France. (I have seen claims that they were intended against the Russians, but they have way too much range for that to be plausible.) France lacked battlecruisers, and their cruisers were poorly armored and few in number.
        In practice, they didn’t work very well. Admiral Graf Spee was hunted down and sunk in the opening months of the war by three British cruisers, exactly the scenario that had been supposed to work in their favor.
        Lutzow never did very much, usually ending up damaged right on the eve of a battle. Scheer was the most successful commerce raider, but spent most of the war in Norway with the rest of the German fleet, occasionally threatening convoys to Murmansk and Archangel.
        Overall, it was an interesting idea, but not a successful one. A conventional 8″ gun cruiser probably would have been more useful, and a commerce raider should have been faster and probably had a better air complement. They might have been useful against the French in the early 30s, but in the war that actually happened, they were next to useless.

        • Protagoras says:

          I thought the reliability of the diesel engines was one of the big problems with the design. Certainly the Germans didn’t build any more diesel heavy cruisers or try to build diesel battleships, so they must have had some reason for abandoning the approach. Though perhaps it was harder to supply enough diesel fuel than to supply the less refined marine fuel oil used by the turbine ships; I’m not entirely sure what all influenced the decisions about adopting or (usually) not adopting diesel engines for warships. I know the Japanese discussed diesel for Yamato. And perhaps they should have gone for it; Yamato had what turned out to be inadequate speed and arguably inadequate range, both of which might have been fixed if (probably big if) they managed to make the diesel work.

          • bean says:

            The Germans kept trying to make diesels work, and there was significant confusion over how much they’d done. Jane’s Fighting Ships of WWII lists S&G as having CODAS propulsion, IIRC, when they were pure steam. The H-class was supposed to be diesel-powered, and they built really big diesels as prototypes. D&G has pictures. I’ll have to check books for more details on their machinery plans.

          • bean says:

            Re diesels, Breyer (who is probably most trustworthy on German ships) suggests that the main reason for selecting them was that they minimized length. Apparently, there was talk of completing Graff Spee with a superheated steam plant, due to improvements in that technology. There was talk of S&G getting diesels, but it would have been a step too far, and steam was better for that power range. Bismarck was intended to go turbo-electric before that got turned into conventional steam, too.
            D&G agree that diesel wasn’t ready for S&G, and give some details of the propulsion plant planned for H. They would have had 12 diesels, 4 for each shaft. Each would have 9 cylinders, with a 950 mm stroke and a diameter of 650 mm. Yes, that big. Oh, and it’s double-acting.
            Eclipse of the Big Gun (D.K. Brown) suggests that the vibration off of the diesels badly hampered their gunnery. D&G describe the foundations built for the H class’s diesels in some detail, suggesting that this is true.

            The Japanese planned to use diesels on the Yamato, and had the installation designed, but scrapped it because the engines were unreliable and it would have been too hard to change them through 8″ of armor. (D&G)

        • John Schilling says:

          Overall, it was an interesting idea, but not a successful one. A conventional 8″ gun cruiser probably would have been more useful, and a commerce raider should have been faster and probably had a better air complement.

          The best commerce raider is almost by definition a cruiser; that’s inherent in the latter term’s etymology. But I’m not sure the Panzerschiffen weren’t as good as it gets, in the heavy raider role.

          8″ gun cruisers were locked in by the treaties; it isn’t clear that they were really the tactical optimum by 1939. Speed, I think you are more concerned with sustained cruising speed than flat-out maximum, unless you plan to chase down enemy cruisers (let them run) or run away yourself (as noted above, that trick never works). And the air wing is of limited value if you haven’t learned to operate seaplanes above Sea State 2 or 3, which as far as I know only the Japanese were comfortable with. The Panzerschiffen had their pair of Arados for calm weather; that was enough.

          And commerce raiding isn’t just about tactical efficiency and tonnage sunk; you also get real gains from ships hiding in port, convoys inefficiently routed, and battleships scattered to the four corners of the Earth to guard against them. You want your raiders to sink ships, yes, but you also want them to strike terror in the hearts of admirals.

          Eight-inch treaty cruisers just like everyone else’s, don’t do that. Panzerschiffen, for a time, did. Until they revealed their timidity in actual battle. If the supposed advantage of your heavy raider is that they can sink anyone else’s cruisers, you really need to prove that by sinking a few cruisers at the outset. If the Germans weren’t willing to do that, the Panzerschiffen were I agree overpriced for the job.

          • bean says:

            Annoying as it is for my reputation as a naval sage, I think you’re right. I will point out that my reference to aviation facilities was about the flight-deck cruiser, although I wouldn’t use that against hard targets.

    • skef says:

      For me, your posts have had way too much information to go through looking for the only interesting topic about battleships, or indeed all of war: dazzle camouflage!

      But again, that’s just me.

    • Salem says:

      I am fascinated by the series and hope you continue.

    • gbdub says:

      One more request to continue. I’m giving up SSC at work for Lent so I stopped commenting on your posts, but didn’t stop reading and enjoying them.

      Totally useless aside: I get that ships are always given feminine pronouns in English but it seems odd in cases where the ship is explicitly named after a man (e.g. George H.W. Bush, Admiral Scheer – actually, aren’t ships properly masculine in German?)

      • bean says:

        Totally useless aside: I get that ships are always given feminine pronouns in English but it seems odd in cases where the ship is explicitly named after a man (e.g. George H.W. Bush, Admiral Scheer – actually, aren’t ships properly masculine in German?)

        It doesn’t seem weird to me, but I’m deeply weird myself. How could something so pretty be masculine, anyway? I’m not sure about the genders in German, but I know the Russians use masculine pronouns. That’s probably why they’re so bad at being nautical.
        Google suggests that the Germans do often use ‘she’, but it’s not universal. Which fits well with their seagoing record. (As further check of this hypothesis, the Dutch use feminine genders, and the French seem to use masculine.)

      • FacelessCraven says:

        Rule 63 is older than the internet.

    • Eltargrim says:

      I’m afraid I don’t have much to contribute, but I’m finding these to be a fascinating read and I appreciate all the effort you’ve put in so far!

    • bean says:

      Thanks for all of the responses from people who are enjoying the series. I may check occasionally, but for now it’s safe.
      I think the next installment will be on battlecruisers. I have many thoughts on the subject, and need to write them down.

    • Nornagest says:

      Just wanted to say I’m enjoying these, although I often don’t have enough domain knowledge to comment meaningfully.

    • FacelessCraven says:

      I’m in lurking mode at the moment for a variety of reasons, but this entire series has been extremely fascinating and directly relevant to my interests.

    • Vermillion says:

      Also wanted to toss in another vote for continuing these. I look forward to reading this series every OT.

      Possible topic, how will the fleets of major powers change in the next 10 years? 30? 100?

    • ADifferentAnonymous says:

      Please continue! Literally the first thing I did on opening this open thread was ctrl-f ‘bean’ to read the latest battleship thread.

      And I’m sure you’re not short on topics. I was going to say I at least need your thoughts on battlecruisers, and a skim through this thread indeed shows that’s next in queue.

      I also saw this sentence: “Their [the Germans’] lack of understanding of the nature of sea power in WWI is shocking”. That’s gotta be a post’s worth of material right there.

      And we’ve heard a lot about the World Wars, but how about steel battleships’ debut in the Russo-Japanese?

      And given how often magazines seem to explode, surely munitions storage and handling is a pretty big topic in its own right?

      • bean says:

        I’m trying to avoid the top-level posts spiraling off into every naval thing ever, so the stuff on German strategy may or may not happen. It does seem to come up a lot. I don’t know much about the Russo-Japanese war, and research for these has to compete with other reading, so that’s not likely to happen soon.