Open Thread 118.25

This is the twice-weekly hidden open thread. You can also talk at the SSC subreddit or the SSC Discord server.

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

612 Responses to Open Thread 118.25

  1. CatCube says:

    Did the comments get reordered, or am I having a stroke?

  2. Le Maistre Chat says:

    Here’s a mystery for the blog that coined the term Lizardman’s Constant: where did the trope of serpent men originate?
    In August 1929, both Robert Howard’s “The Shadow Kingdom” and Edmond Hamilton’s “Outside the Universe” featured serpent men as the antagonists. The next month in Blue Book, Edgar Rice Burroughs’s Tarzan at the Earth’s Core began serial publication, and it first introduced the “lizardman” or “snakeman” Horibs partway through. It’s just possible, but very improbable, that Burroughs was able to read either of the earlier stories to be influenced before finishing his manuscript, but the other two appeared practically simultaneously. Is there an earlier source I’m not aware of?

    • Nornagest says:

      H.P. Lovecraft’s “The Nameless City” (1921) features a race of sapient reptilians who built the eponymous city, but they’re only vaguely humanoid:

      To convey any idea of these monstrosities is impossible. They were of the reptile kind, with body lines suggesting sometimes the crocodile, sometimes the seal, but more often nothing of which either the naturalist or the palaeontologist ever heard. In size they approximated a small man, and their fore-legs bore delicate and evident feet curiously like human hands and fingers. But strangest of all were their heads, which presented a contour violating all known biological principles. To nothing can such things be well compared—in one flash I thought of comparisons as varied as the cat, the bullfrog, the mythic Satyr, and the human being.

      Wikipedia also points to Helena Blavatsky’s Theosophical ideas as an example, but I’m having a hard time nailing down a quote from Google. Theosophy got everywhere in the late 1800s and early 1900s, though, so that’s probably your source.

      • Le Maistre Chat says:

        Madame Blavatsky was attempting to synthesize the Western esoteric tradition with Hinduism and what she’d accept of modern science (hence Lemuria, which wasn’t esoterica but Haeckel’s hypothesis for how lemurs were found in Madagascar and India in the absence of continental drift theory), so she could have just been talking about the nagas.
        I guess “The Nameless City” could be the missing link, but it’s hard to be sure when they’re typical Lovecraft “I thought of comparisons to X, Y ans Z” and not straight serpent men.

  3. johan_larson says:

    In another thread we were discussing how the designers of D&D seem to have omitted a useful ability from low-level mages: making fire.

    But let’s ramp up the difficulty here. Suppose you had a classic AD&Dv1 party who arrived in the wilderness with nothing: no spellbooks, no backpacks with flint&steel, nothing. They have the clothes on their backs, if that. Maybe they survived a shipwreck. How could they make fire?

    Two things come to mind. A cleric could ask for divine assistance through prayer. This is a character close enough to the divine that he is regularly given the power of life and death. Making a special request to get a fire going seems pretty reasonable.

    The other idea is to be very generous with any fire-adjacent spell the magic-user has memorized. Affect Normal Fires isn’t supposed to be able to light fires, but perhaps with effort it could be made to do so. Magic Missile makes no mention of heat, but maybe the missiles could be made hot.

    Other techniques, anyone?

    EDIT: Oh, and let’s assume these are first-level characters.

    • Nornagest says:

      Druids are divine casters and they get a number of fire spells starting at 2nd level: fire trap, heat metal, produce flame, etc. Aside from that, your best bet is probably to lean on wilderness lore. There’s no survival skill in AD&D1 like there is in later editions, but there are some nonweapon proficiencies that might apply, and I’d allow it without one for any druid or ranger.

      • johan_larson says:

        Clerics have a harder time of it, at least RAW: the first spell that looks useful for creating fire is Flame Strike, a 5th level spell. So the local High Priest won’t have a problem.

        • Nornagest says:

          Much more far-fetched but technically not disallowed: use create food and water (cleric 3) to summon a flambé.

      • johan_larson says:

        there are some nonweapon proficiencies that might apply

        What do you mean? It’s been a while, but as I recall in AD&D1 you got your abilities from your race or your class. I don’t remember anything about a skills system.

        • John Schilling says:

          1st Edition AD&D added “Nonweapon Proficiencies” that allowed your character to do a few things completely outside the normal game system, like seamanship or carpentry. Almost no detail or mechanics, just a bit of flavor text, but some of them were sufficiently wilderness-survival-like that it would be reasonable to assume they include improvised firemaking.

          By 3rd edition, there was an explicit skills system that replaced all of these anmany of the class-specific special abilities; each class had favored skills and thieves “rogues” had a particularly high skill allowance to buy something approximating their traditional abilities as discrete skills (climbing, lockpicking, etc). This has the advantage of e.g. allowing Fahfrd to buy at least a modicum of sneakiness so he doesn’t automatically give away the Grey Mouser as they go about their business. And it includes a “Survival” skill, which presumably Rangers are expected to max out, that doesn’t specify firestarting but can “keep yourself and others safe and fed in the wild”. Per London, that had better include starting a fire.

          • johan_larson says:

            1st Edition AD&D added “Nonweapon Proficiencies” that allowed your character to do a few things completely outside the normal game system, like seamanship or carpentry.

            Somewhere in the DMG?

          • Nornagest says:

            They were in the Player’s Handbook in 2E, but I think they were spread across a bunch of splats in 1E.

          • Protagoras says:

            Yes, I’m having dim memories of 1st edition eventually including, I think, 2 “survival guides,” one being the Dungeoneer’s and the other being Wilderness, one or both of which had non-weapon proficiencies.

          • John Schilling says:

            I’m pretty sure there was a table of nonweapon proficiencies in the back of the 1st-edition DMG, but my brother has that one so I can’t check. Obviously, it was viewed favorably enough to grow via splatbook and 2nd edition to the eventual skills system of 3/3.5e

    • Fire bow or fire drill, constructed from materials on site.

      • johan_larson says:

        That’s a thought. Not sure how to play it out though. Should we assume that the PCs know what a fire drill is because the players know? Also, making one of these things with just your teeth and fingernails with is going to be a challenge. And they are hard to use.

        • The Nybbler says:

          Making a fire drill without a knife will be obnoxiously difficult. Hopefully our PCs have skills in flint-knapping as well as knowing about fire drills.

          • Another Throw says:

            There have been a couple times in my life where just smashing a rock was the most expedient means to accomplishing something. I generally don’t carry a knife, the sharp shards were sufficient for the purpose at hand, and much faster than going and getting a knife.

            I’m not sure knapping is really required. It certainly provides more control in the resulting shape, with many advantages flowing from that, but I don’t know if it is really needed for this purpose. In a crunch you may not have time for it, anyway.

        • beleester says:

          Anyone with the Craft or Survival skills should be able to do it easily.

          (Survival doesn’t explicitly say it lets you make fires, but it helps you live off the land and survive harsh weather, so presumably there’s some cooking and heating abstracted into those skills)

          Wait, they would be “Nonweapon Proficiencies” before 3e. Did 1e have a Survival equivalent?

        • Lambert says:

          Knowing about fire drills shouldn’t be unreasonable for someone who has been or expects to be in the wilderness.
          Not sure about folks in a vaguely medieval setting who are city-dwellers (or even a reasonably dense rural area, where you could alway count on there being a homestead within an hour or two’s walk) unexpectedly stuck in the wild.

          In day-to-day life, it’s much easier to keep a fire going, or to borrow or buy fire from somebody else.
          Indeed, there are said to be tribes who do not know how to make fire, but rely on their ability to keep a fire burning.

    • Ninety-Three says:

      We don’t need to generously interpret anything to get an arcane caster lighting fires: Burning Hands is a level 1 spell that makes a cone of fire damage.

      • johan_larson says:

        No spell-books, right? So let’s hope the mage has the spell memorized.

        • Ninety-Three says:

          Sorcerer casting means you don’t have to cross your fingers on memorization, though Burning Hands is a weird spell to put on your first level sorc.

          • Nornagest says:

            No sorcerer in AD&D 1E (or 2E, unless you’re playing Baldur’s Gate). And IIRC, sorcerers do have spellbooks in at least 3E, for… some reason. (That class often struck me as poorly thought out.)

          • bullseye says:

            Sorcerers do not have spellbooks in 3E. 3E sorcerers use the wizard spell list, but they have a smaller number of spells known (which don’t change from one day to the next) and a larger number of spells per day.

          • Nornagest says:

            Now that I’ve looked it up, you are correct. Don’t know why I remembered that.

  4. Le Maistre Chat says:

    So I’ve inexpensively acquired a lot of fantasy miniatures that include women riding lions, orcs riding tigers, and knights on bears. Rationalists, help me rationalize using them!
    Real male lions and the larger subspecies of tigers are too short to ride, but not by much (the smallest horses the Mongols rode were 12 hands, 1.2 meters). That small difference can be handwaved. The bigger problems would be their spine not supporting weight, the logistical problem of supplying obligate carnivores, and that they’re ambush hunters not endurance hunters.
    Grizzly bears are capable of endurance running for miles at 28 mph, and can eat the same kinds of plants as humans. If you could get past the behavioral issues that keep people from successfully taming them IRL, they could function like super-horses in battle. However, the largest male brown bears consume something like 17 pounds of salmon a day, so meat being as energy-dense as it is, how much would that be in standard human rations…?!

    • Nornagest says:

      However, the largest male brown bears consume something like 17 pounds of salmon a day, so meat being as energy-dense as it is, how much would that be in standard human rations…?!

      Not sure about tigers, but I can answer this: I expect it works out to its weight in rations or a little bit more. Meat’s energy-dense, but so is bread, and that 17 pounds probably includes plenty of bones and offal. And brown bears are omnivores, so they’d probably handle human food fairly well.

      • Le Maistre Chat says:

        Nifty; a 1,200 pound horse doing light exercise needs 5-7 pounds of feed, so being able to feed a kodiak-size bear for the same logistics as ~3 horses is good news. Bears are famous for still fighting with mortal wounds, while cavalry horses being peaceful/skittish is why phalanxes were so effective against cavalry.
        So it’s possible the only reason there were never brown bear cavalry IRL is that they can’t be tamed, and I’m handwaving behavior.

    • sfoil says:

      Women are lighter than men, so it makes sense that they would be riding lions whose spines couldn’t support an armored man. Ditto with the orcs, if they’re either more the goblin type or just smaller guys who aren’t suited to being the typical hulking orc warrior. Neither would be heavily armored. Probably light lancers in general equipment.

      Since neither animal is much suited to endurance, they’re used as “interceptors”. Typically, they simply guard important locations; offensively, rather than riding long distances as cavalry do they must be moved on transports (think circus animals). They only attack targets identified (and ideally fixed tactically) by reconnaissance forces in order to avoid wasting valuable operating hours on patrolling. Their riders spend more time hunting things to feed their mounts than riding.

      • bullseye says:

        It doesn’t have to be the riders doing the hunting. These could be expensive super-weapons with a lot of support personnel.

        • sfoil says:

          I don’t think they’re super enough to to justify modern ratios of support personnel. A tiger rider might be more than a match for a horseman but it’s not like an attack plane that can annihilate arbitrary numbers of enemies with virtual impunity. Also, the dietary requirements are a difference of degree rather than kind with the logistical problems posed by horses anyway; it’s definitely an operator-level problem in my eyes.

          • bullseye says:

            They might be super-effective at panicking enemy horses, though that would not require a person riding the tigers and bears.

            Or maybe not. The Romans had access to more or less tamed bears and tigers for the gladiatorial games, but to the best of my knowledge never used them to frighten enemy cavalry in real war.

          • Le Maistre Chat says:


            Or maybe not. The Romans had access to more or less tamed bears and tigers for the gladiatorial games, but to the best of my knowledge never used them to frighten enemy cavalry in real war.

            It’s actually unclear to me why we don’t have tame bears. Jared Diamond in his GG&S chapter on domestication claims that brown bears are utterly non-tamable, citing an Ainu tradition of raising them from cubs and then killing and eating them before they’re big enough to do the same to humans. Yet there are instances of tame bears working alongside humans. Obviously Diamond could just be full of crap, but then what’s the deal? Are bears just not useful when we have dogs and horses?

          • Another Throw says:

            I think we don’t see domesticated bears like we do, e.g., dogs is because:

            A. Offhand it seems domesticatability is associated with being a social species. Bears aren’t particularly social.

            B. Even well domesticated animals occasionally get annoyed and bite/kick/claw. Having an annoyed dog bite you really kind of sucks, but pales in comparison to having an annoyed bear bite you.

            C. Aside from small animals that decide to follow you around and can’t actually hurt you, domestication is deliberate breeding program across centuries. For most of these centuries, “well domesticated” is probably not an apt description. The risk of them deciding that just eating you is way easier than waiting for you to bring dinner is non trivial. (Consider the number of zookeepers that still get mauled.)

            D. In order to come up with undomesticated cubs to start the process, you pretty much have to fight the mother first. Mother bears are famously really pissed off. Fighting said really pissed off mother without good weapons is likely to result in a few casualties. After half your hunting party has been killed I don’t know how forgiving you are likely to be to the cubs afterwards.

            E. Even if you are feeling forgiving, the rest of the tribe almost certainly isn’t, and is very likely to kill you for just bringing that monster here.

            F. And, while you MIGHT find the cubs in the short period between the mother dying from other causes and the cubs also dying, they are siblings and breeding them isn’t going to work well. To make a breeding pair you need to either do something that unlikely twice, or trying stealing some from a live mother.

            G. And that only gets you one breeding pair. For a real domestication program, you are going to need periodic infusions of bears that aren’t cousins.

            H. Also, domestication is going to require culling the ones that are too aggressive. Either mommy isn’t going to like that, or you will have to wait until baby isn’t a baby anymore and isn’t going to like it either.

            If you come across lonely cubs, it is way easier (and more satisfying) to just sell them to the coliseum so you can watch those bastards you captured in the last war fight them. No matter who wins, you get to watch a bastard you hate die!

            ETA: Re: Wojtek

            After a quick glance at Wikipedia, if he was unable to eat and had to be nursed when they found him, about the time they were giving him to the zoo he would have been 5 or 6 years old.

            Female brown bears appear to reach sexual maturity at around 5.5 years old, and males first attempt to mate at around 6.5 years old. This would imply that he probably did not reach sexual maturity while living with the soldiers. Since I think sexual maturity is associated with a change in temperament across a wide range of species, he probably isn’t a very good argument in favor of domesticatability.

            Cool, though!

          • theredsheep says:

            I’m not totally clear on where the line is between taming individual animals and domesticating a strain of the species. There are lots of weird counterexamples. Nobody ever domesticated foxes–yet the Russian farm fox project has produced fine specimens. It just took a lot of ruthless work. Where does one draw the line? Likewise African elephants are supposed to be untameable, but Hannibal famously used war elephants, and it’s hard to believe he was importing them from India.

            EDIT: looked it up; he used a now-extinct northern subspecies. But Wiki also notes that even the main variant were sometimes tamed; it was just hella hard. So it may be a matter of relative difficulty.

          • albatross11 says:

            Cats aren’t social animals, and I don’t think pigs or chickens are, either. The birds used in falconry aren’t social, either. (I’ve heard the claim that the birds don’t ever get close to/fond of their handlers, which may be because they’re not social.)

          • Another Throw says:

            I tend to think that animals that are large and dangerous enough that temperament matters, and/or work closely with humans are extensively domesticated. Dogs and horses are both. (You do not want to get kicked.) Dogs, certainly, (I am less clear on horses) have been breed for an extremely long time to be compatible. They are also both social animals, which certainly helps with the programs. In addition to breeding, we also near universally neuter the males to control their behavior.

            Cats and chickens are neither, and are more tame than domesticated. When the cute kittens turn out to be jerks as adults, you just throw them outside so they can eat the outside mice instead of the inside mice. The breeding for them has been much less extensive and, such as it is, focused on other areas. They’re not social and we don’t need them to be for what they do. (We macerate all of the extra male chicks rather than bothering trying to neuter them.)

            Pigs and cows are an example that is kind of a half and half. They are big enough that they can hurt you, but clipping horns and tusks helps and is easy. We also don’t try to do any work with them. They’re food. We tend to eat them before their adult temperament becomes too much of an issue. Males that make it to adulthood before we eat them are a huge pain in the ass. The people that are stupid enough to try raising them as pets find out the hard way how large and prone to being temperamental they become as adults.

            Falcons and elephants are an exercise for the reader.

          • Nornagest says:

            We also don’t try to do any work with [pigs and cows].

            Pigs, yes, outside a few niches like truffle finding. Oxen are common draft animals, though, and were more common before the invention of the horse collar sometime in the early Middle Ages. They’re slower, but stronger and less neurotic, than horses, which makes them well-suited to niches like plowing where you don’t need much speed.

          • Controls Freak says:

            Cats aren’t social animals

            Cats are barely-domesticated, solitary desert ambush predators. I’m usually just thankful that mine can’t hurt me that much.

          • The original Mr. X says:

            Cats tend to hunt alone, but it’s misleading to say that they aren’t social. If you look at feral cats, for example, they tend to live together in big colonies, rather than going off on their own. (Though maybe that’s just something that was bred into them by humans?)

          • Name hidden for obvious reasons says:

            Cats are ambush predators, but they are not solitary. When feral (alley cats) or semi-feral (barn cats), they naturally form mutually assisting colonies of up to a few dozen members, who hunt cooperatively, protect each other, teach and protect each other’s kittens, and sleep together in piles for shared warmth.

            Just because the species of wild cat that most closely resembles the domestic cat is solitary, does not mean that domestic cats are.

          • bullseye says:

            I don’t think cats were bred by us to live in groups. They bring home food to the group, which is only useful for a group living in the wild and irritates human owners.

            I think cats, like dogs and horses, are descended from a wild population that no longer exists.

          • bullseye says:

            There are dancing bears, and I guess they don’t eat their handlers too often. Maybe a bear aggressive enough to be a war bear would be too aggressive to control.

            Guns Germs and Steel mentions that, if rhinos were tamable, they’d make awesome cavalry. But part of their theoretical effectiveness as cavalry is their willingness to kill, which they might lose as tame animals.

          • Protagoras says:

            Guns Germs and Steel mentions that, if rhinos were tamable, they’d make awesome cavalry. But part of their theoretical effectiveness as cavalry is their willingness to kill, which they might lose as tame animals.

            I don’t see why. Elephants can be tamed and still be trained to fight effectively. If rhinos could be bred to be more similar to elephants in personality, presumably they could be war animals. But as I recall, elephants have not particularly been domesticated by breeding; it was more common historically to capture them in the wild as adults and train them than to breed and raise them, probably because of their slow maturation. It’s just that for whatever reason, the base elephant personalities make it possible to tame a wild elephant, while that’s not the case for rhinos. Rhinos don’t take as long to mature as elephants, but it’s still a lot longer than horses, so anybody who wanted to start a breeding project to produce a better behaved rhino would have to be planning for the extremely long term. It’s not surprised nobody ever bothered with that (or at least nobody ever got anywhere trying).

      • Le Maistre Chat says:

        Women are lighter than men, so it makes sense that they would be riding lions whose spines couldn’t support an armored man. Ditto with the orcs, if they’re either more the goblin type or just smaller guys who aren’t suited to being the typical hulking orc warrior. Neither would be heavily armored. Probably light lancers in general equipment.

        And that’s how the miniatures look: women with just torso armor and woman-sized orcs in in clothes, equipped as chargers.

        Being moved on circus-like wagons and dropped off at prepared positions as interceptors is an interesting, plausible idea, thanks.

    • Basil Elton says:

      Actually I doubt any of those would be any more battle effective than a horse when used as a horse. Heavy cavalry mostly relied on its sheer mass and speed to break through the enemy lines, and horse is both heavier and faster than any predator. And cats themselves attack by jumping on a prey – which is much harder to do with a rider on your back, and I imagine rider won’t be in a good position to strike in this moment, thus not adding to offensive capabilities. What I’m trying to say is that such a unit would attack either in a “predator mode” or in a “knight mode”, each being much weaker than an unmounted predator or a knight on a normal horse respectively.

      That being said, I can imagine a setting where those are useful. First of all, it’s probably a level of technology where you don’t have much armor, otherwise predators would just not be very useful in a battle, period. There’s little fangs and claws can do to an armor made to withstand a strike of a sword. Second, I’d suggest that those miniatures actually represent something like a parade, or ritual, or demonstration – not how they would actually go into battle. Rather the “riders” (who probably should be called trainers) would fight alongside with their pets guiding and directing them, while mostly relying on them to do the actual fighting. (Yes that does imply fantasy druid-ish level of control above pets, but we already presume we have it if we’re going to use solitary predators as mounts in fight) That lifts the concern with low endurance and a weak back – animals can carry they trainers short distances purely to show off, but on march or in battle they don’t.

      Under these limitations, I think such units can be basically the ultimate power in a close quarter combat. In a dense enough forest, on a very rough terrain with lots of rocks and gorges or if you somehow ended up having street fights in what is basically a bronze age technology level. In these conditions range weapons and traditional cavalry are almost useless, so are infantry formations, and animals capable of moving silently, sensing the enemy behind an obstacle, and having much faster reactions, gain a tremendous advantage over humans.

      As of logistical problems, I think those can be solved pretty easily. First, you naturally can feed your animals on the corpses of your killed enemies. In fact the predators probably won’t really ask you before doing so anyway. When those are not available, it doesn’t seem terribly hard to have some amount of livestock (preferably capable of feeding off of grass so you don’t have to carry fodder for them) with your supply train, especially given that such elite and specialized units are expected to always operate as a part of a large army. This would’ve limited ability of these units to make long fast marches but as was mentioned being ambush predators they probably never had it in the first place. Plus you can always buy/seize stock from locals if there’re any, as armies did with all the other food.

      Having just returned from my first visit to Mexico, I propose to put these in the alternative history where smilodons survived until Neolithic and then were (magically I guess) tamed by the Maya people and used for wars in jungle. Helps too that the Mayas were rather short, and I doubt they’d have any concerns about feeding their pets with their enemies; also apparently smilodons were somewhat more stocky than lions. And when conquistadors arrived Mayas fucked them badly, without of course mistaking them for gods even for a second (“Your mount eats what? Grass?? Get off my way child!”) And yes Wikipedia says there used to be grizzlies in Mexico too.

      • John Schilling says:

        Heavy cavalry mostly relied on its sheer mass and speed to break through the enemy lines, and horse is both heavier and faster than any predator.

        Heavy cavalry mostly relied on scariness to break through enemy lines. No horse is large enough or fast enough to survive an encounter with a firmly-held Mark I Pointy Stick, really speed and size work against you in that context, but big horses carrying big armored men with pointy sticks of their own are scary enough that the enemy will often turn and run rather than killing them. Cavalry charges against a prepared enemy are a bluff.

        Bear cavalry, or psycho bitch-lady on tiger cavalry, just might have an edge here. If your ladies aren’t actually psychotic bitches, you’ll want to fake that. Costumes will help, and should be reflected in the miniatures. Use your imagination, and encourage the enemy to use his.

        • Basil Elton says:

          Ok, not claiming any kind of expertise in medieval warfare here, but. What you seem to be saying is that the only thing heavy cavalry can do with an infantry trained, equipped and prepared to fight off a cavalry charge (think of Landsknechts, facing the cavalry and already in formation) is to scare them. I agree with that, but from what I know massive use of such infantry was basically one of the reasons why chivalry fell out of use on the battlefield. And while they coexisted deploying such infantry was basically a standard countermeasure against heavy cavalry. Even though a predator may have slightly better chances at scaring, I doubt that would be enough to really change these dynamics.

          What I was talking about was a cavalry charge against pretty much anything else: archers, other cavalry, light infantry, heavy infantry without pikes, heavy infantry with pikes but already having broken the formation and/or facing the other way.

          Also, if we’re presuming that technologies and tactics in this universe developed in roughly the same patterns as in ours, then by the time you have something similar to pike squares weapon and armor are already developed enough to render any predator almost useless per se against a trained and equipped soldier, thus basically leaving you with a scary, small, low endurance and extremely expensive horse. I was mostly thinking about much lower technological level where soldiers don’t have most of their body covered in armor, and halberds and two-handed swords don’t exist yet, like Ancient Greece, or even Egypt, or as I mentioned pre-columbian America.

          • John Schilling says:

            Ok, not claiming any kind of expertise in medieval warfare here, but. What you seem to be saying is that the only thing heavy cavalry can do with an infantry trained, equipped and prepared to fight off a cavalry charge (think of Landsknechts, facing the cavalry and already in formation) is to scare them.

            No, pretty much the only thing heavy cavalry can do to any sort of infantry in a fair fight is to scare them, and then kill them when they turn and run. It’s just that infantry that isn’t trained and in formation, scares easy.

            To find a material exception to this, you’d need to to the extreme of javelin skirmishers vs. cataphracts with armored horses. Otherwise the horse takes a pointy stick (javelin, spear, arrow, or bolt, almost all pre-gunpowder infantry had at least one of these) before the rider comes within reach, and the conflict is reduced to guy who just fell off a horse vs. guy who didn’t just fall off a horse.

            Harald’s infantry at Hastings wasn’t trained or equipped to fight off a heavy cavalry charge; there had been no such thing as heavy cavalry in England for about five hundred years. But William’s proto-knights made no progress until several feigned retreats and the fortuitous death of Harald resulted in a breakdown of English discipline.

          • theredsheep says:

            This agrees with what I’ve read in military histories; since infantry can concentrate force better, and no horse is stupid enough to charge headlong into a bunch of guys just standing there with spears pointing at its eyeballs, melee infantry that stands its ground is basically a brick wall against melee cavalry. Pikes make them even more effective, but ordinary spears will do.

            (In Byzantium, at least, pikemen–menavlitoi–made their appearance at almost precisely the same time heavy cavalry were reintroduced; both were ninth-century inventions)

      • Nornagest says:

        horse is both heavier and faster than any predator.

        Not in sprints, they aren’t. Bears are much faster than they look: grizzlies can reach 35 mph over open ground, as fast as a racehorse, and tigers are in the same ballpark or better. They’d probably do better still if they’ve been selectively bred as battle mounts. Horses have far better endurance, but we’ve covered that.

        A big bear is about the same weight as a small horse, but intimidation and pointiness might make up for that.

        • Basil Elton says:

          Fair point about speed. But it still seems very doubtful that it’d make sense to trade off endurance, mount’s own weight, much more of carried weight, and complicate logistics greatly – all for just intimidation and some increase in sprint speed.

          • Le Maistre Chat says:

            I would think the trade-off would be as much or more willingness to charge pointy sticks, which men can be trained to have but horses can’t.

          • theredsheep says:

            I imagine bears are also rather intimidating to horses. Might make an interesting counter to mounted archers; they get close and unleash a volley, and the bear rushes them. Horse freaks out and throws rider, bear gets meal. Something like that?

  5. Uribe says:


    As I mention above, I don’t read the fantasy or sci-fi which seems to be the popular genres on this forum, but I can tell many readers here are also well read in traditional literature, which is more my speed.

    What mainstream literary fiction are you into? My tastes in such are mostly mainstream: Tolstoy, Hemingway, Melville, Kafka, Fitzgerald, Nabokov, Roth, Kundera, Pynchon… More recently: Knausgaard and Ferrante’s novellas. Is anyone much into that shit? I like classic literature because I feel like it shines the light on human existence as it is — something I’m still desperate to understand. I also love the poetry in much of the prose in those writers. I’m not much into pure poetry, because I like characters and narrative.

    I was too late to SSC to respond to the On the Road review. I disagreed with that take vehemently. For me, a novel is a work of art in the same way as is a symphony. It’s entertainment. High-level entertainment, but entertainment, not to be confused with social commentary or didactic fiction. So my disagreement with all the negative takes on On the Road boils down to: You aren’t reading this as a work of art; there is no social commentary; there is no message; there is nothing to learn from this about how you are supposed to live. You are just supposed to go along for the ride and enjoy it, like you might a symphony or a superhero movie. It’s an adventure story, and that is all. Same goes with all great works of literature. And symphonies. And Breaking Bad episodes. These things are beautiful in the same way a mathematical proof which may have no practical value may be beautiful.

    • WashedOut says:

      I think I share your literary taste. I’ve probably written over a thousand words on SSC alone on Crime and Punishment, which I claim is as near to a perfect work of art as can be possible. Dostoevsky and Kafka are my bread and butter; Asimov my after-dinner mint.

      I tried reading Tolstoy but it didn’t vibe with me. Shortly after, I heard Jordan Peterson argue along the lines that “Tolstoy was a sociologist whilst Dostoevsky was more concerned with the individual” and I think this helped explain my preference. The English-made War and Peace TV miniseries was fantastic though.

      I recently read the short stories by Akutagawa (Rashomon et al.) and loved them; would recommend to any Kafka fan. I’m currently reading Killing Commendatore by Murakami – so far so good, although I worry that it’s going to get lost in it’s own pensive melancholia and I’ve read enough Proust (which is to say any Proust) to know better.

      Has anyone read any Platonov? I started reading The Foundation Pit because I wanted more civil engineering in my Russian lit., but it hasn’t captivated me so far.

      • J.R. says:

        Have you read previous Murakami novels? If so, how do you feel about Commendatore relative to his other work? I’ve read 5 of his novels (Kafka on the Shore, 1Q84, Wind-Up Bird Chronicle, Hard-Boiled Wonderland, Colorless Tsukuru Tazaki) and don’t think I need to ever revisit him. Hell, even Wind-Up Bird alone encapsulates all his pet themes quite nicely.

    • dndnrsn says:

      Thinking about it, I’ve probably read more serious fiction than genre stuff over the past year, year and a half, though I haven’t read much fiction in general. The 3 most recent pieces of highbrow fiction I read were:

      1. The entire Sea of Fertility series by Yukio Mishima. Really torn on these. Probably the weirdest stuff I’ve read that, in English at least, would get filed under “proper Literature”. The first two impressed me much more than the last two, with the second really standing out. The third and fourth are interesting in their picture of Japan during and after the war, but I didn’t enjoy reading them as much. The first two are much more immediate; I think he cared more about the subject matter. I read it in English translation; no idea how it reads in Japanese.

      2. The Sea, the Sea by Iris Murdoch. Fabulous writing. A lot going on, as far as the plot goes. I read this one a little more sporadically than usual, so I probably didn’t get everything I could have from it.

      3. The Regeneration trilogy by Pat Barker. Historical fiction so good they gave it a Booker and let it call itself Real Literature. I recommend this highly. Reads very quickly; her descriptions of people are wonderful. Really well drawn characters.

    • J.R. says:

      I share much of your taste, particularly Kafka, Nabokov, Roth, and Pynchon. Add Gaddis (see username), DeLillo, D.F. Wallace, C. McCarthy, Dostoevsky, and Salter and you have a good cross-section of my favorite authors.

      I haven’t read On the Road or Scott’s review of it, and I have resisted learning anything about it at all by cultural osmosis. Generally, I am most interested in the quality of the prose and how different narrative techniques are used to give insight into the characters. I’d argue that having something interesting to say about culture or human nature — themes — is a prerequisite for my enjoyment as well, unless the prose is so outstanding that I could listen in rapt attention if the writer was describing paint dry (see: James Salter). Some of my favorite novels certainly have something to say about how to live, or at least comment on the writer’s society: JR is a satire aimed at American capitalism’s commodification of art and what happens in a society that is saturated in media consumption; Infinite Jest is about our addiction to distraction and renunciation of same; The Human Stain is about the myth of the self-made American.

      One of the things I wanted to do this year was to start a blog with very detailed reviews — think SSC-level posts — of my favorite novels. I am/was not an English major, so my aim is to give a layman’s analysis of theme and prose, and make the case for why it should be read. Would there by any interest in this?

      • Deiseach says:

        Would there by any interest in this?

        Sure, I’d be interested! Always up for a discussion about books 🙂

    • dodrian says:

      Some of the mainstream literary fiction I’ve enjoyed included Treasure Island, Great Expectations, The Phantom of the Opera, Les Miserables, Moby Dick and To Kill a Mockingbird. I do enjoy picking up classics, especially late 19th/early 20th century. I have read a number of short stories by Russian novelists, and enjoyed those.

      I don’t tend to read current mainstream literature (I mostly stick to Sci-Fi). I think the most recent non sci-fi/fantasy book I’ve read and enjoyed is A Confederacy of Dunces (pub. 1980).

      • Plumber says:

        I read A Confederacy of Dunces back in the late 1980’s or early ’90’s (I girl/women I had kissed recommended it), and it didn’t really stick with me.

    • Le Maistre Chat says:

      Hmm, glances at bookcase

      Swift, Defoe, Austen, the Brontes, Walter Scott, Poe, Balzac, Dumas, Borges, Tolstoy, Dostoevsky.

      • Plumber says:

        Borges counts?

        I think of gmail as a fantasy writer.

        I have a couple of short story collections of his (and Poe as well).

        Those are good.

        • Nornagest says:

          I think of gmail as a fantasy writer.

          That’s my favorite autocorrect mishap in quite a while.

          Borges doesn’t fit cleanly into Anglosphere literary genres. He wrote some stuff that’s closest to fantasy (smarmy lit majors might call it “magic realism”, but it has the rules that MR usually lacks), but he also wrote a lot of stuff that’s closer to litfic, and a lot of stuff that’s closer to the Western than to anything else in English. Thank gauchos getting into knife fights.

    • Well... says:

      Sort of along those lines is why I think “The Wire” is fundamentally different from “Breaking Bad”: the former teaches the viewer something about us and our society; the latter does not.

      Anyway, I haven’t read a whole lot of classic literature. I read a fair number of Hemingway’s novels and a few of Steinbeck’s in my late teens/early 20s and liked those. I was pretty into those two writers back then.

      I thought Moby Dick was sorta fun but sorta stupid.

      I’ve read at least one or two volumes of Poe stories and liked a lot of them, but thought others (e.g. “The Sphinx”) were friggin retarded.

      I really liked both “White Fang” and “Call of the Wild” as well as some of Jack London’s short stories, if those count.

      How about Dante? I thought “Inferno” (Mark Musa translation) was amazing, but “Purgatorio” was so dull I quit a few dozen pages in.

      “Dr. Jekyll and Mr. Hyde” was great, despite its verbosity.

      I very much like the two or three Conrad novellas I’ve read. Same goes for Kafka (a bunch of his short stories) and Camus (a few novellas).

      • Deiseach says:

        How about Dante? I thought “Inferno” (Mark Musa translation) was amazing, but “Purgatorio” was so dull I quit a few dozen pages in.

        Oh, them’s fightin’ words! 😀

        It really depends on the translation. If you find the start boring, then maybe skip it until
        Dante and Virgil start climbing the mountain, on the first Terrace of Purgatory where the proud purge their debts. I agree, if you’re not in theological agreement/sympathy with the concept of Purgatory, it will seem very dull next to Hell.

        The whole point of the opening of the Purgatorio is the sense of calm after the action of Inferno. Previously, Dante and Virgil have been climbing, walking, running, flying, in boats – constant motion downwards and downwards until they get out.

        Now at the start, it’s the pale cool colours of dawn, unlike the fires and smokes of Hell. It’s the trembling moment of repose, balanced between stasis and movement, the rest and recovery before they start climbing the mountain. Under the instruction of Cato, Virgil takes Dante to wash his face in the morning dew – to cleanse himself of the smuts and stains of Hell – and to gird himself with the reeds. It’s a delicate, tender and beautiful moment that is referenced by other poets, the one I know best is Seamus Heaney who used that twice; first in a poem about the sectarian murder of his cousin, then years later as part of another poem where he reflects upon it and if he maybe understated that murder, turned it into a poetic image rather than the reality of what it was:

        (The poem was written in memory of Heaney’s second cousin Colum McCartney, a young Catholic who was shot by Protestant militants in the Fews Forest, south-west of Armagh, Ireland, in 1975.)

        The Strand at Lough Beg
        In Memory of Colum McCartney

        All round this little island, on the strand
        Far down below there, where the breakers strive
        Grow the tall rushes from the oozy sand.
        –Dante, Purgatorio, I, 100-3

        Leaving the white glow of filling stations
        And a few lonely streetlamps among fields
        You climbed the hills toward Newtownhamilton
        Past the Fews Forest, out beneath the stars–
        Along the road, a high, bare pilgrim’s track
        Where Sweeney fled before the bloodied heads,
        Goat-beards and dogs’ eyes in a demon pack
        Blazing out of the ground, snapping and squealing.
        What blazed ahead of you? A faked road block?
        The red lamp swung, the sudden brakes and stalling
        Engine, voices, heads hooded and the cold-nosed gun?
        Or in your driving mirror, tailing headlights
        That pulled out suddenly and flagged you down
        Where you weren’t known and far from what you knew:
        The lowland clays and waters of Lough Beg,
        Church Island’s spire, its soft treeline of yew.

        There you used hear guns fired behind the house
        Long before rising time, when duck shooters
        Haunted the marigolds and bulrushes,
        But still were scared to find spent cartridges,
        Acrid, brassy, genital, ejected,
        On your way across the strand to fetch the cows.
        For you and yours and yours and mine fought the shy,
        Spoke an old language of conspirators
        And could not crack the whip or seize the day:
        Big-voiced scullions, herders, feelers round
        Haycocks and hindquarters, talkers in byres,
        Slow arbitrators of the burial ground.

        Across that strand of ours the cattle graze
        Up to their bellies in an early mist
        And now they turn their unbewildered gaze
        To where we work our way through squeaking sedge
        Drowning in dew. Like a dull blade with its edge
        Honed bright, Lough Beg half shines under the haze.
        I turn because the sweeping of your feet
        Has stopped behind me, to find you on your knees
        With blood and roadside muck in your hair and eyes,
        Then kneel in front of you in brimming grass
        And gather up cold handfuls of the dew
        To wash you, cousin. I dab you clean with moss
        Fine as the drizzle out of a low cloud.
        I lift you under the arms and lay you flat.
        With rushes that shoot green again, I plait
        Green scapulars to wear over your shroud.

        • sfoil says:

          I wasn’t unimpressed with the poesy of Purgatorio, but what really struck me was Purgatorio’s theme of repentance compared to the Inferno’s narrative of punishment. It’s easy to come away from Inferno thinking that these people are in Hell because they’re bad; but no, it’s because they refused to admit their sin and ask forgiveness.

          The paradisaical pageant at the end Purgatorio was stilted and tedious, though.

      • sfoil says:

        I’ve read at least one or two volumes of Poe stories and liked a lot of them, but thought others (e.g. “The Sphinx”) were friggin retarded.

        Poe was supposedly the first American to support himself solely by writing fiction. I don’t know whether that’s true, but he did have to write to feed himself and as a result seems to have published pretty much whatever he wrote. This led to a pretty uneven catalogue, but on the other hand it means you can really see him play around with ideas in works another writer would have trashed before he really gets it right.

    • sfoil says:

      I do read and enjoy literary fiction, mostly 19th – early 20th century. Still, I don’t read it as much as nonfiction or SF. My tastes are very mainstream — standard with classic literature since “canonical” works tend to in fact be very good. I used to be more into the better-known literature from Classical Antiquity but I got away from it and my Latin has faded from long disuse.

      As far as newer literature, the most recently published litfic I’ve read is Denis Johnson’s Vietnam War novel Tree of Smoke from 2007. It’s written from a spook/intel perspective and rises well above the genre standards of the “spy novel” but I wasn’t ultimately that impressed. Like other novels about rear-area hijinks, I couldn’t help comparing it unfavorably to the admittedly far less accessible Gravity’s Rainbow.

      You aren’t reading this as a work of art; there is no social commentary; there is no message; there is nothing to learn from this about how you are supposed to live. You are just supposed to go along for the ride and enjoy it, like you might a symphony or a superhero movie. It’s an adventure story, and that is all. Same goes with all great works of literature. And symphonies. And Breaking Bad episodes. These things are beautiful in the same way a mathematical proof which may have no practical value may be beautiful.

      I haven’t read On The Road, but I strongly disagree with this. I suppose you can argue that one ought “to go along for the ride” but that just isn’t what happens. Literature (and other forms of art) simultaneously reflects and shapes the culture that produces it (and the individual artist, but that’s another matter). The artifice involved is generally morally neutral, but the use to which it is put is not. And the “use” of On The Road, viz. glorifying “defecting all over the place” against a highly functional high-trust society by appealing to trumped-up victim mentality, is very bad.

    • brad says:

      I’ve participated in book recommendation threads before, but I don’t think I’ve recommended these yet: Sometimes a Great Notion (Kesey), East of Eden (Steinbeck), and Man In Full (Wolfe)

    • Basil Elton says:

      Nice to see someone defending On The Road here! I do agree with your view on it, and I also liked it a lot.

      To the main question, I like both sci-fi and classical literature, though I’ve done most of my classic literature reading in the student years (it has nothing to do with curriculum, I was studying CS). Hemingway, Remarque, Steinbeck are the favorite, also Palahniuk’s Haunted if you’re willing to count that as a “traditional literature”

    • I don’t know if you count Kipling as literature–he did get a Nobel prize–but he is probably my favorite long-dead author. Kim is his one good novel, but he wrote a lot of good short stories and poems. There are other older poets I enjoy, but I can’t think of any English prose writers earlier than Kipling whom I have read much of.

    • Tarpitz says:

      In my teens, I read almost exclusively fantasy, sci fi and historical fiction. Over the years, my tastes have moved far more into literary fiction. Of your list, I very much enjoyed what I’ve read of Hemingway and Nabokov, and intensely dislike Fitzgerald. Others I like include Laclos, Austen (though not Mansfield Park – Fanny’s insufferable), Dickens (earlier work generally better than later), Hardy, Forster, Galsworthy (honestly a little marginal on the literary front, but tremendous fun nonetheless), Waugh, Capote and most of all Graham Greene (my favourites are The End of the Affair and The Heart of the Matter). Among living writers, Mantel and Tartt are excellent, and while Le Carre can be a little verbose at times I think his best work (The Spy who came in from the Cold, the Karla trilogy, The Constant Gardner) is legitimately good. And having recently re-read Pullman’s Dark Materials trilogy, well, they may be children’s fantasy, but they’re also just honest-to-god excellent.

    • sorrento says:

      “On the Road” was never intended as “high-level entertainment.” It’s not a ride in Disneyland. It was one of the founding texts of the Beat movement in postwar America. It was certainly intended as social commentary and a message about how you are supposed to live. A huge number of influential writers and artists read it and took away a strong message from it.

      Of course, you can read it any way you want. Hell, you can analyze whether there’s cryptograms hidden in the text like a crazy numerologist. But I think you do understand what Scott is getting at. There’s a strong undercurrent of nihilism and irresponsibility in the book. Most of us wouldn’t want to be friends with these people. And the other implied question is whether maybe some aspects of the past were better, and the Beats and the Hippies made it worse. Scott has talked about this before (usually from an optimistic perspective…)

      Breaking Bad also wasn’t intended as pure entertainment, but as a description of a one man’s fall from grace. The tone just gets darker and darker as the series goes on. By the last season, it’s amazingly heavy-handed. Walt even starts a conversation with “I think you know that we’re not going to heaven… we’re going to the other place” (or something like that, I forget the exact quote). Walt’s actions have really bad consequences for his family and his community, and the show points that out at every possible turn. This isn’t the Simpsons, where Homer gets his job and his car back at the end of the episode.

    • Plumber says:


      The only “non-genre” (fantasy, historical, mystery; and science fiction) fiction that I can remember reading after leaving high school and finding “page turning” is Donna Tartt’s A Secret History, Kerouac’s On The Road, John Steinbeck’s In Dubious Battle, and The Grapes of Wrath.

      I simply don’t remember well enough anything else “mainstream” that I’ve read to name, and all that has been short stories in The Atlantic, Harper’s and The New Yorker, and the like.

  6. What’s the largest “dark matter” subculture that exists in your area but doesn’t interact with you at all? I can think of a few, including Protestantism (broadly defined) and Basic Millenial Woman culture, but maybe there are even bigger groups that are even more invisible to me.

    • The Nybbler says:

      All of them, that’s what happens when you’re not social. There’s all the synagogues and churches. Then there’s the various groups of kids, groups of parents, and groups of parents and kids; groups based around high school sports may be the largest, though some groups based at the Jewish Community Center might be larger. There’s a couple of nightclubs in town, so there’s some groups there. There are senior citizen groups. I expect the Spanish-speaking population has something.

    • Uribe says:

      SSC is dark matter culture to me except when I’m on it. Don’t know anyone in real life who resembles anyone on this forum. I did enjoy some of the D&D origins talk up-discussion, only because I played D&D some as a kid and knew nothing about the backstories. I’ve never been much of a reader of fantasy or sci-fi, so much of the fiction talk on this forum is beyond me.

      On the D&D note — yeah I could Google it — but I’ll just ask here: who came up with the multi-sided dice? Did early D&Ders invent and make them? I can remember how as a 12-year-old those translucent 8 and 12 sided dice seemed like magical jewels in my possession… It was part of the appeal for me, as the local DM for the neighborhood.

      I interact with a lot of “normal Americans” because I work at unexceptional jobs and hang out at bars where there’s a mixture of only moderately ambitious professional types, who may be Republican or Democrat, who may be druggies or lawyers but often both. Some young, more old. Religious freaks and atheists. Feminists and libertarians. None too intellectual, though. Have older, more intellectual friends I rarely speak with much anymore, maybe once a month. But they aren’t SSC, 160 IQ intellectuals, just people who are well-read enough to hold a conversation about Dostoyevsky.

      In general, I feel like there is a huge disconnect between what I read on the webz and those I interact with in real life. Mainly in the sense that it seems like all the bloggers I read are super-successful and hyper-intelligent, whereas my cohort is only moderately ambitious at best. I sometimes get sick of reading online about all the intellectual and business superstars (I read Marginal Revolution a lot.)

      So I feel like a meet a decent cross-section of American society, if only transactionally at bars.

      • Le Maistre Chat says:

        On the D&D note — yeah I could Google it — but I’ll just ask here: who came up with the multi-sided dice? Did early D&Ders invent and make them? I can remember how as a 12-year-old those translucent 8 and 12 sided dice seemed like magical jewels in my possession… It was part of the appeal for me, as the local DM for the neighborhood.

        In the early 1970s, Gary Gygax wanted to implement percentage-based tables made by real military scientists in war gaming. By 1972, he found a teacher supply catalog (apparently Creative Publications of California) selling plastic Platonic solids (d4, d6, d8, d12 & d20) and figured increments of 5% were good enough. The d20 found its way into the original 1974 edition of Dungeons & Dragons, which still used common six-sided dice for damage, hit points, etc. (The prior Blackmoor campaign that Dave Arneson ran with a house ruled version of Gygax’s wargame Chainmail, used a bell curve of six-siders). Since his supplier refused to just sell him d20s, he changed D&D to have weapons do different dice of damage, and the hit points of different classes became “roll a d8 for Fighters, d6 for Clerics, or d4 for everyone else when you level up.”
        The d10 has yet another source, which I don’t know about. TSR sold the plastic Platonic solids separately and as part of the 1977 Basic Set, and I guess they started up their own manufacturing by the time the game was changed to use the d10 (Fighter hit points in the AD&D Player’s Handbook)?

  7. Statismagician says:

    As discussed in the pre-registration thread, I’m going to be doing some casual analysis in the spirit of Scott’s birth-order article using a neat public dataset (UNC’s Add Health. I want to see how weird SSC is compared to a larger more randomly-selected sample, and also to investigate some thoughts Deiseach had on potential modulation of birth order effects from of half-siblings and large age gaps.

    Is there anything else you’d all like to know about while I’m setting things up? There are literally thousands of variables covering basically very nearly every facet of life for many thousands of people across ~14 years; within the limits of the Don’t Be A Nazi* form I had to sign to get the data I’m willing to look into whatever. I’m going to do this probably in a couple of weeks when I have more time, and I’ll try to re-post this in a couple more open threads as well as do a proper pre-registration once I finalize the analysis plan. If something suitably weird comes up I may withhold chunks of the results and try to get them in print, but that seems pretty unlikely.

    This widget searches the codebook if you want to if they’ve got specific variables.

    *No, I won’t try and figure out who any of the respondents were. Yes, I do realize that very small cell counts can be individually-identifying and won’t publish them.

  8. Uribe says:

    China Debt

    I keep reading how China’s debt keeps mounting. Who’s buying it? My guess is it’s Chinese citizens, but if so, why?

    • John Schilling says:

      Chinese culture has long, and even after several generations of communism and quasi-communism still does, make a cardinal virtue out of Saving Your Money to Provide For Your Family in Times Of Adversity. Given the weakness of other safeguards against adversity in China, this is quite sensible. Chinese households have a savings rate of over 40%

      Chinese culture has generally not, and even after a generation or so of quasi-capitalism still mostly doesn’t, make a virtue of ordinary people “gambling” on the stock market, and is not well-provided with things like discount brokerages and mutual funds. So there’s no good place for all that savings to go except cash in mattresses, government bonds, or local savings banks that put it into government bonds.

      Consequently, in addition to all the real prosperity that has been generated since 1984, there is a similar level of virtual prosperity which China’s citizens have agreed to forgo realizing so long as things are going well. What happens if the Times of Adversity come and a few hundred million Chinese demand to realize that virtual prosperity to safeguard their family’s foothold in the new middle class, is an interesting question.

      • Uribe says:

        It’s one of the questions I’m most interested in right now, as a spectator. Perhaps I should say disinterested in.

        Saw recently where despite the abandonment of the one child policy, Chinese fertility is down even further. Demographics loom large.

        Maybe it’s just who I’m following on Twitter, but I sure have turned against Xi-lead China recently. I’d call myself a globalist neoliberal who has been excited about China’s rise, but now that I’m reading about the increasing authoritarianism — the seeming backsliding into more authoritarianism — I’m growing very sour on China. I hope Trump hasn’t influenced my perspective. I’m not a Trump fan and disagree with his approach to China, but I do agree China needs to be challenged. I don’t want them exporting their political culture.

        As they revert to more socialism, in the form of more government vs. private investment, I hope their economic experiment fails and that there is a (relatively bloodless) revolution in China. Their system strikes me as evil and moving ever quickly in the wrong direction.

        Not that I claim to understand China at all. Would be interested in hearing more perspectives on the current direction of China on this forum.

      • @Uribe:

        You might be interested in my parents’ experiences of China, reported in their autobiography (Two Lucky People). My father met with both Zhai Ziyang, then general secretary of the Communist Party, and on a later visit with Jiang Zemin, then general secretary. His impression of Zhai was very positive; he appeared to be someone with a good intuitive feel for economics. His impression of Jiang was very negative–someone who didn’t understand economics, imagined that the success of Japan and Singapore was due not to the market but to control from the center, and that that was the model China should imitate.

        Nothing on Xi, who wasn’t yet in power at the time, however.

  9. theredsheep says:

    So, an unusual request for book recommendations. My seven-year-old son and I just finished the kids’ version of The Boy Who Harnessed The Wind, which is about a Malawian teenager who builds an electric windmill out of scrap to power his family’s hut. It was our bedtime story. My boy loved it. But I don’t know where to go from here. He’s sufficiently fascinated by stories about people doing technical stuff that I think he’d sit there while I read him The Martian and/or Artemis. Sadly, y’know, profanity, sexual content, weird references to seventies pop culture … anybody know some good kid-friendly books about people, preferably kids, doing/making tech stuff? He also really enjoyed The Door in the Wall, in large part because Robin spends a lot of time on crafts like woodcarving.

    • Statismagician says:

      Does it have to be high technology, or just kid-friendly inventive problem-solving? My Side of the Mountain could work.

      • dick says:

        Yeah, second that. And maybe check out the “Primitive Technology” youtuber guy (and his many, many imitators, many of whose videos are also great).

      • TakatoGuil says:

        In a similar vein, A Series of Unfortunate Events features a character who MacGuyvers several solutions to the protagonists’ problems a book, though the latter half of the series may be a bit advanced for a seven-year old.

    • albatross11 says:

      Heinlein’s _Have Spacesuit, Will Travel_ is probably rather dated, but it is a kid-appropriate story involving a kid who’s spending a lot of time tinkering with a space suit.

      • Evan Þ says:

        Second the recommendation! That was one of my favorite books when I was eight or so.

        Heinlein’s other “juveniles” are good, too. For particularly inventive protagonists, I’d recommend Have Spacesuit, Will Travel, Farmer in the Sky, and Tunnel in the Sky. The Rolling Stones doesn’t quite fit the bill, but I also really enjoyed it as a kid.

    • johan_larson says:

      There was some tech stuff discussed in Robinson Crusoe when the protagonist was trying to figure out how to survive on his island. You might need to edit some of Friday’s behavior on the fly though.

    • Swiss Family Robinson. Robinson Crusoe. The Mysterious Island (Verne). They aren’t mainly about kids, but meet your other requirements.

      How to books, such as How to Stay Alive in the Woods.

      Heinlein’s Tunnel in the Sky.

    • Hoopyfreud says:

      Hatchet is good, but possibly not age-appropriate just yet. Up to you.

    • SamChevre says:

      Suggestions-from my wife, who’s the kid’s book expert here:

      Landmark books about inventors, including Young Man in a Hurry and the one about the transatlantic cable.

      The 1911 Boy Scout handbook–a how-to book rather than a storybook.

      Ralph Moody’s books (Little Britches and its sequels through The Fields of Home.)

      And from me: a lot of books from the Amish-Mennonite world (Rod and Staff, Pathway)–these are books for and about farm kids mostly. I particularly like the Benji books from Pathway.

      • Evan Þ says:

        Why only through Fields of Home? The other three books set after Ralph Moody grew up (Shaking the Nickel Bush, The Dry Divide, and Horse of a Different Color) are different, but they show just as much of his inventiveness, and I enjoyed them too when my mom read them to me as a kid.

    • Incurian says:


    • theredsheep says:

      Thank you guys for all the suggestions. Alas, even my son is not nerdy enough to enjoy HPMOR. He might spend tremendous amounts of time gluing popsicle sticks into model bridges, but he has his limits.

    • Elephant says:

      Another vote for My Side of the Mountain. Also, my son of a similar age likes the “Classic Starts” series of simplified classic literature. There’s Robinson Crusoe there, which is quite good (as are others in the series that I’ve read).

    • theredsheep says:

      I remember reading The Gammage Cup (though not any survival-ish aspects). But my local library’s entry for it has an … interesting summary: “A sensation when published in Moscow and a bestseller in Europe, the memoirs of this remarkable woman—the widow of the charismatic Bolshevik leader Nikolai I. Bukharin—offer a new dimension to our understanding of Soviet history.”

      I guess it was a very, very deep allegory. The Mushrooms are the Trotskyites, I assume.

    • Paul Brinkley says:

      Castle by David MacCauley. In fact, just about anything by MacCauley, especially Cathedral, City, Pyramid, Mill, Ship, and Mosque. (By now, you should detect the pattern.)

  10. Edward Scizorhands says:


    I have given up on taking 0.3 mg at around 3pm. My sleep was really bad and I could not fall asleep, even with lots of exercise.

    One day I took 1.5 mg at 10pm and was sleeping pretty well.

    There are other possible explanations but this has gone on enough. I declare my experiment over. I am just going to take 3 or so of my 0.3 mg pills each night.

    • Evan Þ says:

      I might’ve missed previous posts about your experiment, but did you perhaps try taking 0.3 mg around 10 PM?

    • T82 says:

      You’re taking melatonin at 3 PM? Is that about when you go to sleep?

      Also, isn’t melatonin destroyed by light hitting your skin? If you take it at 3 PM and you’re not in a bedroom with heavy curtains aluminum foil over the windows, that could be a problem.

      As for me, I just take 50mg diphenhydramine (benadryl) if I want to get sleepy.

      • Statismagician says:

        Not quite; natural melatonin production is suppressed by light, , and the 3pm timing is based on metabolic rates, as I recall.

      • Edward Scizorhands says:

        You’re taking melatonin at 3 PM? Is that about when you go to sleep?


        So if you want to go to sleep (and wake up) earlier, you want to take melatonin early in the day. How early? Van Geijlswijk et al sums up the research as saying it is most effective “5 hours prior to both the traditionally determined [dim light melatonin onset] (circadian time 9)”. If you don’t know your own melatonin cycle, your best bet is to take it 9 hours after you wake up (which is presumably about seven hours before you go to sleep).

        I was aiming for a 10pm sleep time.

    • Controls Freak says:

      Different issue here. Taking 0.3mg of melatonin around 3p puts me to sleep wonderfully on schedule. However, what I’m not sure about is use over time. Scott’s article says that you need to persistently use it to correct disorders, but doesn’t really speak to otherwise.

      It seems like if I take melatonin regularly in the afternoon, I have no problem falling asleep on time. Lately, however, I’ve been waking up earlier and earlier. I’m not sure whether this is due to specific other factors going on in my life (I’m aware of the factors and how they might impact my sleep) or whether continuing to take melatonin at the same time in the afternoon will continue to “drag my sleep cycle closer to it”. If I don’t take it, it’s sometimes a crapshoot as to whether I’ll fall asleep easily. Any advice?

    • Dack says:

      An afternoon dose is for trying to hack your circadian rhythm. (When you sleep.)

      If you are using melatonin as a hypnotic (to get to sleep in the first place), you should take it about 30 minutes before bed.

      The smallest dose I’ve been able to find over the counter is 3mg, so I’ve been cutting those into quarters (~.75mg) and that dose works very well helping me get to sleep.

      I have tried taking the 3mg dose, and I find that it works too well, as in I am still very groggy getting out of bed 8-10 hours later, and I am physically aching from sleeping too long/deeply on a particular body part. Also when I tried to get up after only 6-7 hours I found I was about to doze off behind the wheel.

  11. AlphaGamma says:

    Phrases not words:

    There’s the old joke about the HRE being neither Holy, Roman nor an Empire.

    The fisher cat isn’t a cat (it’s a mustelid) and doesn’t catch or eat fish (its main prey are porcupines and snowshoe hares).

    • theredsheep says:

      I wish the article included info about the whole trauma-workshop thing the officers go to; Emily Yoffe did an extended piece some time ago about how a lot of that was supposedly based on the work of one woman who didn’t back up her words with evidence.

    • Aapje says:


      To add to what theredsheep linked to, this seems to me to be mainly about introducing a bias to police investigations and convictions. Take the start of the story:

      Detective Justin Boardman had a reliable way of clearing many of the rape cases that crossed his desk. When the only witness was the victim, he would call her, warn that it was a “he said, she said” case that would be tough to investigate, and hope that she would drop it.

      Usually, she did.

      Then one day, Boardman ducked into a class on the trauma caused by rape. He heard scientific explanations for why rape victims could not scream or fight back, and why they often initially struggled to remember details of the crime.

      Soon he realized that he had closed dozens of cases in which the victim likely was telling the truth. It shook him to the point of tears.

      What I see here is a lazy, shitty detective who refused to investigate the claims of the accused and accuser using far more objective information. Or what I like to call it: detective work.

      And later in the story for a prosecutor who became a trainer for detectives:

      Kelly realized that the same behaviors that she had cited to dismiss many of the cases that crossed her desk could instead be used as evidence that the rape happened. And if police truly understood the effects of trauma, they would arrest and charge more rapists.

      The story in the Atlantic that theredsheep refers to shows that this unscientific theory of how trauma works, results in any inconsistencies in accusation to not just be explained away as trauma, but to actually be considered evidence that the person was victimized.

      It seems to me that this fundamentally violates the assumption of innocence, where mere “he said, she said” should not be taken as evidence that the accused is most likely guilty.

      Furthermore, even if it is the case that poor testimony is evidence that a traumatic event happened, the big elephant in the room is that this still doesn’t mean that the traumatic event was exactly as alleged or that the accused was the perpetrator.

      I see many red flags in the story suggesting that both Boardman and the entire police department were and are awful at their work.

      Just to name one, Boardman went to a trial and there gave mental support to the accuser. This is the job of the lawyer/prosecutor who is supposed to be biased for the accuser and/or those who give aid to victims. In contrast, Boardman’s job is to find the evidence against AND in favor of the accused, which is incompatible with picking the side of the accuser and interpreting things in her favor.

      • albatross11 says:

        The “he said, she said case” thing seems like a useful heuristic.

        Everything I’ve ever heard/read on the topic says that going through the investigation of a rape case is pretty horrible for the victim. (People I know personally who have gone through it have referred to the investigation/trial as being worse than the rape/sexual assault being investigated.) The investigation is also expensive in terms of police resources, which are not unlimited.

        So if you can look at a case and quickly determine that, because of the lack of any evidence other than the testimony of the alleged victim of the crime, there’s basically no way you’re going to be able to bring the case to trial (or if you do, the defendant will almost certainly be found not guilty), why would it be bad to set that out up front, so the victim knows that if they investigate this fully, she’s probably going to endure a long awful ordeal for nothing?

        Suppose you change the police’s behavior so they always carry out the investigation, even in cases where it’s clear they’ll never be able to prove anything. The net effect is approximately zero additional rapists taken off the street, but many more victims traumatized and many more police hours spent doing unproductive investigations instead of investigations likely to end up actually putting some criminal in jail. Offsetting that, there’s the benefit that the rapist has to go through an ordeal, too–that might be some deterrent to the rapist. But it sure seems like this is one of those things where you make a great show of your concern for an issue, in a way that ends up hurting the people you’re supposed to be helping.

        • Statismagician says:

          I think I broadly agree with this. The first-order thing to do would be to try and modulate legal proceedings for this class of crime such that they’re less awful for victims/less wasteful of police and public legal resources, but I’d want to be really sure we were going to get it right on the first try and then stop fiddling with the underpinnings of our legal system as soon as we were done, which I’m not sure I trust anybody to do right now (certainly not both parts).

        • Thomas Jorgensen says:

          The overwhelming majority of rapes are serial, false charges are really rare. The obvious thing to do (note, not a cop) is to go through her social circle and drop hints that if anyone had a rape happen they never reported, now would be an excellent time (without, obviously, naming names) Odds are fairly good you will get another three witnesses, and now it is a slam dunk.

          • Statismagician says:

            Not a lawyer; does this really work? I’d have thought that would generate 3 more unwinnable cases, but I don’t know how several of those coming out simultaneously plays with judges/juries usually.

          • Randy M says:

            I was going to ask this as well. Of course your Beyesian process does give you much better odds now with three accusations, but whether that makes a enough of a difference in court, idk.

          • Statismagician says:

            Yeah. I mean, court of public opinion, absolutely and trivially proven, but that only works for celebrities.

          • Deiseach says:

            It also gives you a great chance of anyone with a grudge, or attention-seeking, or who knows because Lisa told her what Greg did before she went to the cops, deciding to make a claim in order to get on the accusation train.

            I think if you get three witnesses before you go on a fishing expedition, great. If you go asking “Hey, anyone here been raped/assaulted and want to help us out?” in a social circle where there’s a good chance people already know about Lisa, or they know you asked Greg to come in for a little chat, then I’m not so sanguine.

          • Walter says:

            I’ve always wondered where the ‘false charges are really rare’ thing comes from.

            Like, the false accusations that fail show up as cops not prosecuting, or perps getting off. The ones that succeed show up as rapes.

            How would you know how many false charges there are? Like, absent a thought reading machine it seems mad tuff to figure out.

          • theredsheep says:

            Last I heard, the “percentage of false claims” thing was based exclusively on proven false claims, such as when we can verify that the accused was somewhere else at the time, or when the rape kit doesn’t match his DNA, etc.

            Now, I’ve also read that there’s a certain pattern to false rape claims, as far as anyone can tell: it tends to be done by women who feel they would get in trouble otherwise, e.g. if there’s a pregnancy, or if a kid’s out way past curfew. In fact, a lot of false accusers are minors, because they’re bad at evaluating the probable consequences of their actions. But I don’t know where I read that.

          • albatross11 says:


            I think this is a very general problem: how many people in our prisons are innocent of the crime for which they’re serving time? I’ve seen plausible estimates around half a percent to a percent, but also some estimates that are much higher for particular times and places (DNA testing old rape cases in a Southern state).

            ISTM that it’s difficult to estimate the real rate of false accusations, because we can imagine errors in both directions. Some of the cases where the police are convinced the accuser is lying are probably real rapes. Some of the cases where the police and prosecutor and judge and jury are all convinced the accuser is telling the truth are probably false accusations. How would we untangle that?

          • @albatross11:

            Estimating the rate of false positives is hard. Various people have attempted it, generally for some nonrandom subset of cases. Here are two examples, conveniently available in the footnotes to a book I am about to publish:

            Gross et. al. 2014 give 4.1% as a conservative estimate for the fraction of defendants sentenced to death who were innocent. Their conclusion is based on data on exonerations. A defendant may be exonerated on the grounds that the evidence on which he was convicted was not strong enough to satisfy the “beyond a reasonable doubt” requirement for criminal proof yet still be guilty, but the authors believed that the bias from counting all exonerations as implying innocence was more than outweighed by other biases in the opposite direction.

            Roman et. al. 2012 made use of a cohort of 634 Virginia cases dating from 1973 to 1987 for which tissue evidence happened to have survived. Their results suggest that in about 16% of the cases the evidence, if analyzed, would have reduced the probability of conviction (my analysis of their results).

          • Mark Atwood says:

            false charges are really rare

            Have you all forgotten that Scott has already analyzed this exact issue?


            tldr: being credibly falsely accused is about as likely as being killed in a car accident. Do you wear seatbelts?

          • Aapje says:

            I want to point out that ‘falsely accused’ is often incorrectly interpreted (including by Scott in his article that Mark links to) as requiring intent to deceive. This is not the case, just like a falsehood can be, but does not have to be, a lie.

            The distinction between a malicious false accusation and a unintentionally false accusation is generally not that significant to the accused person once the accusation is reported to the police or other people with the power to punish, because both can result in negative judicial and extrajudicial consequences.

            It’s not the job of the justice system to figure out whether accusers genuinely believe that their accusation is correct, but rather, to figure out if the accusation is (reasonably) correct.

            This is especially relevant since self-reported data suggests that a very high percentage of people who believe that they were raped where under the influence of mind-altering drugs (mainly alcohol) at the time (and presumably the alleged rapist often as well).

            Secondly, the matter under debate is not just what the false accusation percentage was in the past, but especially also what the consequences will be of changes to the legal system, the creation of extrajudicial forms of judgment and punishment, etc.

            I personally suspect that the cases that were (fully) investigated and/or prosecuted in the past were far less likely to be false accusations due to stricter standards (not necessarily correct ones, as we saw in the story, with our upwards failing detective), than cases that were never reported or quickly shut down.

            Many seem to have as an ideal that every woman who thinks that she was raped goes to the police, gets her accusation investigated, often with a very strong assumption of both good faith and of her infallibility and ends up with having the accused prosecuted.

            This can potentially result in many more false accusations resulting in very negative consequences to those who are accused, but who are innocent (or (far) less guilty than alleged).

            It’s especially worrisome that this potential problem seems to often be ignored, including in the article that Nancy showed us. If we could trust the media and government agencies to ask critical questions and investigate potential problems, it is far less likely that things are allowed to get out of hand first and that this is only noticed very late, when many people have been hurt (this is a common pattern in general).

            Note that the article that theredsheep responded with is by the only journalist that I’m aware of who does half-decent investigative work into Title IX excesses, Emily Yoffe. She is an unlikely person to do so, as she seemed to have given up on serious journalism, preferring to write advice columns and writing shallow ‘Human Guinea Pig’ stories. Yet apparently she cares about this issue.

            It is rather serious that only one person seems to investigate this. Proper attention by journalists is often important to keep organizations honest and it seems likely that one mere journalist is not good enough and can merely enlighten those who truly pay attention, but not enough people to truly embarrass delinquent organizations. Again, this makes it quite likely that problems keep getting ignored until the injustices pile up so high that they can no longer be ignored, resulting in a scandal and backlash.

          • tldr: being credibly falsely accused is about as likely as being killed in a car accident.

            That (.3%) was the result Scott got on his hyperconservative estimate–ignoring all of the things wrong with the article he critiqued except that one that was provably false. His moderately realistic result was 3% and he suggested that it might plausibly be 2.5 times that high, i.e. 7.5%.

          • Thomas Jorgensen says:

            Do not see why it would fail in court – the standard is “reasonable” doubt, and if you manage to find four victims willing to testify, no sane jury is going to question that. (and yes, the person worrying about *coordinated* perjury is expressing insane doubt. Sorry Deisarch, but that shit is nuts).

            The most likely ways for this investigative strategy to fail is:

            1: The perp has more victims, but you straight up just fail to convince any of them to come forward.

            2: Rape is bi-modal – Surveys of perpetrators (People will tell an anon survey everything, apparently) tell us there are two types of rapist – The one-time offender who, well, was likely off his head, and does not do it again.

            And the serial offender, who averages 8-10 victims. 80-90 percent of people who get raped had a run-in with someone in the second group, but this “investigation technique” will obviously fail against first time offenders, regardless of which group they belong to, so that is some 15-20% (First time serial, one time offenders added together) chance you are barking up an empty tree.

            It also completely fails against stranger rape – you can only flush out more victims if the rapist is targeting a group you can effectively contact (This is why Meetoo wreaked such havoc on the famous. Once someone goes “Je-Accuse!” at a famous rapist, every single one of his victims will hear. That is really hard to arrange for Bob the traveling salesman who has raped 42 women across 32 states but is otherwise a non-entity)

            This approach should also be fairly proof against honest cases of mistaken identity – Avril, Sandra and Tessa are not going to all finger the same (wrong) guy because he happened to bear a striking resemblance to their rapist. Eyewitness testimony is bad, but not that bad.

          • John Schilling says:

            Do not see why it would fail in court – the standard is “reasonable” doubt, and if you manage to find four victims willing to testify,

            Testify to what, exactly? Because the law is big on exact. “Having been a rapist” is not a crime, and properly so. “Raping Alice on Friday, December 28, 2018” is a crime. If you’ve got Alice testifying that Bob raped her on 12/28/18, that’s one witness to one crime, he said she said. If you’ve got Carol claiming that Bob raped her on 12/30/18, that’s a different crime, and I don’t think she will even be allowed to testify at Bob’s trial for raping Alice. Nor Alice at any trial for Bob raping Carol.

            IANAL, but only if Bob is convicted of raping Carol, and if in his trial for raping Alice he himself goes beyond denying the rape and testifies “…because I’m just not that kind of guy”, is any reference to his rape of Carol allowable in his trial for raping Alice.

          • The Nybbler says:

            and yes, the person worrying about *coordinated* perjury is expressing insane doubt.

            Off the top of my head, there was this “mean girls” case. There was the Jian Ghomeshi case. The idea that people would coordinate lies to harm someone they disliked is not “insane”.

            Testify to what, exactly. “Having been a rapist” is not a crime, and properly so. “Raping Alice on Friday, December 28, 2018” is a crime. If you’ve got Alice testifying that Bob raped her on 12/28/18, that’s one witness to one crime, he said she said. If you’ve got Carol claiming that Bob raped her on 12/30/18, that’s a different crime, and I don’t think she will even be allowed to testify at Bob’s trial for raping Alice. Nor Alice at any trial for Bob raping Carol.

            They allowed five such third parties to testify at Cosby’s second trial.

          • brad says:

            IANAL, but only if Bob is convicted of raping Carol, and if in his trial for raping Alice he himself goes beyond denying the rape and testifies “…because I’m just not that kind of guy”, is any reference to his rape of Carol allowable in his trial for raping Alice.

            This is for federal but it’s about the same in most states, FRE 404:

            (b) Crimes, Wrongs, or Other Acts.

            (1) Prohibited Uses. Evidence of a crime, wrong, or other act is not admissible to prove a person’s character in order to show that on a particular occasion the person acted in accordance with the character.

            (2) Permitted Uses; Notice in a Criminal Case. This evidence may be admissible for another purpose, such as proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident. …


            Subdivision (b) deals with a specialized but important application of the general rule excluding circumstantial use of character evidence. Consistently with that rule, evidence of other crimes, wrongs, or acts is not admissible to prove character as a basis for suggesting the inference that conduct on a particular occasion was in conformity with it. However, the evidence may be offered for another purpose, such as proof of motive, opportunity, and so on, which does not fall within the prohibition. In this situation the rule does not require that the evidence be excluded. No mechanical solution is offered. The determination must be made whether the danger of undue prejudice outweighs the probative value of the evidence in view of the availability of other means of proof and other factors appropriate for making decisions of this kind under Rule 403.

            If the defendent chooses to testify, rule 608 applies:

            (a) Reputation or Opinion Evidence. A witness’s credibility may be attacked or supported by testimony about the witness’s reputation for having a character for truthfulness or untruthfulness, or by testimony in the form of an opinion about that character. But evidence of truthful character is admissible only after the witness’s character for truthfulness has been attacked.

            (b) Specific Instances of Conduct. Except for a criminal conviction under Rule 609, extrinsic evidence is not admissible to prove specific instances of a witness’s conduct in order to attack or support the witness’s character for truthfulness. But the court may, on cross-examination, allow them to be inquired into if they are probative of the character for truthfulness or untruthfulness of:

            (1) the witness; or

            (2) another witness whose character the witness being cross-examined has testified about.

            By testifying on another matter, a witness does not waive any privilege against self-incrimination for testimony that relates only to the witness’s character for truthfulness.

          • 10240 says:

            If you’ve got Alice testifying that Bob raped her on 12/28/18, that’s one witness to one crime, he said she said. If you’ve got Carol claiming that Bob raped her on 12/30/18, that’s a different crime, and I don’t think she will even be allowed to testify at Bob’s trial for raping Alice. Nor Alice at any trial for Bob raping Carol.

            @John Schilling : If that’s true, I don’t think it’s right. Assuming that, a priori, the events “Bob has raped Alice” and “Bob has raped Carol” are highly correlated, but “Alice falsely accuses Bob” and “Carol falsely accuses Bob” are only slightly correlated, then “Alice accuses Bob” increases the a posteriori probability that Bob has raped Carol, and vice versa.

          • John Schilling says:

            I’m not sure the false accusations are as weakly correlated with each other as you imagine. And you’re neglecting the fact that false accusations will also be correlated with true accusations, and vice versa. There’s no point in making a false rape accusation that won’t be believed, and here we have statistically literate rationalist saying that an accusation should be judged more believable if they are preceded by other accusations. Every rape accusation, true or false, increases the probability of false accusations against the target.

            And it increases the probability of true accusations for the same reason. I suspect the correlation is stronger for false accusations, because a true accuser can also expect that physical evidence will support their claim but a false accuser depends entirely on the integrated credibility of all accusers. But that’s speculation.

            True accusations will be positively correlated with prior true accusations, false accusations with prior false accusations, true with false and false with true and I’d be surprised if you could quantify any of those correlations within a factor of two. And then I will laugh when you tell me that juries are composed of statistically literate rationalists who will use this information to improve the reliability of their verdicts, and I will agree with the law’s great skepticism of such “evidence”.

          • Aapje says:

            An issue with human cognition is that it is highly interpretive, where people fit their experiences into narratives. The exact same experiences can support quite different narratives.

            For example, imagine a man who chatted with a woman during a party and after some time kissed her on the mouth, even though she was not into him. Then after he kissed her, she pushed him away and ran off. This can support these three different narratives on the part of the woman:
            1. I probably sent signals of interest by accident and the guy can’t be blamed for the unwanted kiss. I overreacted and should have just told him, so we could have handled it like adults.
            2. The guy was too aggressive and made a mistake, perhaps because he was drunk, but no harm done. That he didn’t come after me, shows that he isn’t a bad person.
            3. The guy sexually assaulted me and if I hadn’t ran off, he probably would have raped me.

            What interpretation people choose seem to depend heavily on things that have little to do with the exact details of the event, like the reputation of the person. A high-status person can get an extreme benefit of the doubt, while someone who is commonly believed to be a bad person can be assumed to have ill motives for things that would be shrugged off if done by a person without this reputation.

            So these kind of reputation effects cause there to be unwitting coordination between people, in how they interpret the actions by a person.

            Once a narrative has been picked, an issue is further than evidence often gets interpreted in the framing of the narrative. The behavior that can be considered nice and romantic in one narrative can be interpreted as manipulative and psychopathic in another.

        • John Schilling says:

          There were some parts of the article that suggested Utah was able to improve their interrogation procedure to obtain more actionable information from rape victims using the insights provided by Valentine et al. That’s probably the biggest gain here, and it’s a disappointment that it is a secondary focus to the article. Because, yes, the narrative that goes “Step 1: Policeman believes traumatically victimized rape victim. Step 2: ??? Step 3: Justice!!!”, is almost as annoying as the “Step 1: Collect underpants Step 2: ??? Step 3: Profit!!!” one.

          More generally, for almost every type of crime other than murder and kidnapping, the most common police response is “Yes, we agree that this crime happened, and we’re going to put that in our files for future reference. We suggest you move on as well”. There is no reason to expect rape to be any different, even if all bias and prejudice are removed from the system. What we need are good heuristics for deciding which crimes to investigate and prosecute, and ways to minimize the harm to the victims whose attackers are going to get away with it.

        • 10240 says:

          Are “he said, she said” cases generally hopeless? I may have too romantic an idea about the effectiveness of police questioning, but I would assume police ask both the accuser and the accused detailed, tricky questions, and the one who is lying will often end up getting confused and contradicting themselves. Whereas if (e.g.) the accuser is telling an entirely coherent story, then she is likely to be telling the truth, especially if the account of the accused is on shaky grounds.

          However, this implies that police need to ask the accuser awkward, detailed questions if her testimony is to provide strong evidence, and that contradictions in the accuser’s account should invalidate it as evidence. To further decrease the likelihood of a false testimony, it should be made risky by prosecuting false accusations if self-contradictions or other evidence makes it clear that the accusation is false — though, once again, with a strong presumption of innocence, not prosecuting based on the sort of confusion that might happen even during a truthful accusation.

          • The Nybbler says:

            but I would assume police ask both the accuser and the accused detailed, tricky questions, and the one who is lying will often end up getting confused and contradicting themselves

            Most likely they’d both get confused and contradict themselves, regardless of who is lying. This is how the FBI builds perjury traps, after all. Made worse by alcohol being involved in so many cases.

          • John Schilling says:

            I would assume police ask both the accuser and the accused detailed, tricky questions, and the one who is lying will often end up getting confused and contradicting themselves.

            Unfortunately, so does the one who is telling the truth. Memory is fallible, suggestion is powerful, and truth is usually embellished. Smart policemen know this, and know the limitations of tricky questioning. Stupid policemen just wind up arresting whoever is foolish enough to talk to them without a lawyer, because the inevitable contradictions and confusions “prove” they are lying guilty scumbags. And really, smart policemen sometimes do this as well, because they are smart enough to know that the incentive is to make an arrest that won’t get laughed out of court and actual guilt or innocence is secondary.

            ETA: Ninja’d by Nybbler.

          • Aapje says:


            A common mistake is that people think that humans have good perception and memory. We don’t. We have shitty perception and memory and very advanced ways to fill in the gaps to get workable narratives, where we use informed speculation.

            The result is that testimony commonly sucks, including by victims who try to accurately recount what happened to them. That said, some people are way more accurate than others.

            The way to best deal with this is to treat testimony as indications of what might have happened and then find more objective evidence which provide hard facts and can also be used to figure out how reliable each testimony is; find the commonalities between various testimonies, etc.

            Your suggestion mainly works as a sort of IQ-test, where people who are bad at interpreting reality get called liars.

        • theredsheep says:

          I don’t know if Rod actually verified that the guy worked the scene he says he did, but it seems relevant.

      • Nabil ad Dajjal says:

        This may seem like a bit of a nitpick, but the presumption of innocence doesn’t and shouldn’t apply to the police and prosecutors. The court should presume that you’re innocent until proven guilty beyond a reasonable doubt, but it’s not the prosecution’s job to argue your side of the case nor the police’s jobs to go looking for exculpatory evidence.

        It’s still a bad idea to teach cops unfalsifiable theories about how victims behave, because police and prosecutorial resources are limited and there’s always a risk of wrongful conviction even with a fair trial.

        • but it’s not the prosecution’s job to argue your side of the case nor the police’s jobs to go looking for exculpatory evidence.

          I agree with the first but not the second. The police are supposed to be finding out who committed the crime, and evidence that someone didn’t commit it is relevant to doing so.

        • The original Mr. X says:

          Maybe the police shouldn’t be looking specifically for exculpatory evidence, but they need to keep an open mind, otherwise you open the system up for all kinds of abuse. For example, there was a rape case here in the UK a few years ago where it turned out that the police had been deliberately withholding exculpatory evidence from the defence and prosecution lawyers. (Specifically, texts from the plaintiff to the accused which completely contradicted the girl’s story; the police had confiscated the accused’s phone, and refused to release it, until the prosecution lawyer, who thought that there might be some evidence he could use on it, got a court injunction ordering the police to hand it over.)

          • Nabil ad Dajjal says:

            I didn’t mean to imply that police or prosecutors should hide evidence, that’s absolutely not how the system is supposed to work.

            My point was more that there’s a reason why it’s called the adversarial system. The process as a whole is supposed to be neutral but the police aren’t neutral fact-finders. They should use effective techniques and not waste time on suspects who probably didn’t do anything. But we shouldn’t be relying on their objectivity.

          • Aapje says:

            @Nabil ad Dajjal

            The prosecutor is adversarial, but the police is not supposed to be. Otherwise no justice can be expected, because without reasonably objective fact-finding, no fair case can be presented to the jury/judge.

            Are you not aware of how you can make pretty much any claim look reasonable by cherry picking the evidence? A system where the police looks for evidence against the (initially) accused, but not evidence that is exculpatory and/or against others, is one where the deck is stacked heavily against the accused. So the idea of innocent until proven guilty then becomes a farce.

            Right now it already seems a pretty big issue that those who are accused in the US and have money to pay for their own investigators and such, are much more likely to be exonerated than those who lack that money and have to depend on what evidence and avenues of investigation the police and prosecution deem to be relevant.

            Ultimately, the job of the police is not to make a case against the accused. Their job is to figure out whether (serious enough) crimes happened and who perpetrated them. The person(s) who the police believe to be guilty of a (serious enough) crime doesn’t (only) have to be the accused and the crime that gets identified doesn’t have to be similar to the original accusation.

            If the police has the goal to merely find evidence for the original accusation, that is called tunnel vision and thus bad policing.

        • brad says:

          This may seem like a bit of a nitpick, but the presumption of innocence doesn’t and shouldn’t apply to the police and prosecutors. The court should presume that you’re innocent until proven guilty beyond a reasonable doubt, but it’s not the prosecution’s job to argue your side of the case nor the police’s jobs to go looking for exculpatory evidence.

          That’s not quite right. While we do have an adversarial system, prosecutors have a duty to see that justice is done–not just to secure the highest possible penalties for the most number of people.

          From the ABA standards:

          The primary duty of the prosecutor is to seek justice within the bounds of the law, not merely to convict. The prosecutor serves the public interest and should act with integrity and balanced judgment to increase public safety both by pursuing appropriate criminal charges of appropriate severity, and by exercising discretion to not pursue criminal charges in appropriate circumstances.

          I’m not as familial with any universally applicable codes of professional conduct for police officers, but I’d think their role is to build the strongest case possible against the correct person. That search for truth is best accomplished by looking for all evidence–culpatory and exculpatory.

          • Statismagician says:

            Specific incentives probably matter more than boilerplate-y codes of conduct; if the cop has an arrest quota or the prosecutor needs to get x% convictions to be considered for promotion, that will plausibly do something weird to the proper functioning of the judicial system. I don’t if either of those general types of things are common, though.

          • brad says:


            I think you are conflating is and ought.

          • Statismagician says:


            Could you unpack that? I thought you were conflating is and ought, at least potentially; as mentioned I don’t know a lot about the relevant incentive schemes.

            EDIT: added the @~.

          • brad says:


            I guess I was implicitly echoing Nabil’s “don’t and shouldn’t” and so wasn’t being entirely clear on is / ought.

            To clarify:
            1) I think prosecutors ought to seek justice.
            2) Prosecutors have a formal obligations to seek justice.
            3) I have little to no confidence that most prosecutors act in full compliance with their formal ethical obligations including #2.

          • Statismagician says:


            I agree with all of those.

      • dick says:

        This is an article about someone who has worked in a field for many years, and who thought they were good at it, who then went to a seminar, learned a new and possibly better way to do it, put it in to practice, saw better results, championed it to their boss, got it more widely adopted, and then got an academic to study the results, who agreed that it was indeed better. It takes a lot of chutzpah to be confident that they’re wrong based on a newspaper article summarizing their experience.

        • Aapje says:

          It’s a guy who as a police officer was known for using excessive violence.

          Who admits to have been a lazy police detective who cared not about finding out the truth, but to get cases cleared.

          Who seems to now feel bad about how shitty he acted and now appears to want to make up for it by being unprofessional in the other direction. I did not make up this unprofessional behavior, it’s right there in the article.

          Any more red flags and you are in a Chinese parade.

          and then got an academic to study the results, who agreed that it was indeed better

          I explicitly said that the old situation was awful. Your criticism only makes sense if I would have argued that the old situation would be better.

          It is perfectly possible that the old situation was extremely shit and the new situation is slightly less shit. That doesn’t necessarily make it acceptable.

          Note that the study only investigated certain things and not things like the percentage of accused or convicts who report being mistreated, falsely accused or falsely convicted. So it can very easily make things seem better than they are, by cherry picking what things they studied.

          The study also had methodological limitations, as the article itself notes.

      • BBA says:

        It seems to me that this fundamentally violates the assumption of innocence, where mere “he said, she said” should not be taken as evidence that the accused is most likely guilty.

        I see this as a potential flaw in the concept of presuming innocence.

        • Statismagician says:

          It’s a feature, not a bug. Our justice system is very explicitly supposed to* let more guilty people go than it convicts innocent people. We can fiddle with that principle if you can get enough people to sign off on it, but I’d at least like to see some kind of cost/benefit analysis done.

          *I am well aware ought isn’t is (even though ought ought to be what is is; sorry, I’ll let myself ought).

        • The Nybbler says:

          I see this as a potential flaw in the concept of presuming innocence.

          The traditional response to this is to capriciously accuse you of rape. If the bare word of an accuser, put up against the bare word of an accused, is sufficient for conviction, we’re going to have a whole lot of state-imposed injustice.

          • BBA says:

            As opposed to the injustice of having a large class of serious crimes that are impossible to prosecute?

            Granted, on balance I come out against weakening the presumption of innocence, mostly because it would end up hurting already underprivileged people worse.

          • The traditional view is that it is worse to convict an innocent person than to fail to convict a guilty one. Do you agree or disagree?

            Do you want to give protecting the underprivileged lexicographic priority over protecting other people? If a lower standard of proof results in one innocent poor man being convicted of rape and fifty fewer rapes of not-poor women, is it clearly a bad thing?

          • dick says:

            The traditional response to this is to capriciously accuse you of rape. If the bare word of an accuser, put up against the bare word of an accused, is sufficient for conviction, we’re going to have a whole lot of state-imposed injustice.

            I invite you to waltz down to your local police station and capriciously accuse someone of rape and see how far you get.

        • broblawsky says:

          The judge and jury are supposed to presume innocence, not the rest of the justice system. The prosecutor (and the police) have no such requirement, and should probably assume guilt on the part of anyone they suspect is guilty, since they’re the ones who usually bring charges in the first place.

  12. Nancy Lebovitz says:

    I’m so behind I don’t know whether anyone has posted this.

    Still, I think it’s a really interesting example of consenquentialism (I think her existence was a net gain) vs. deontology (a whole lot of lying involved, and some losses as a result).

    Also, there’s a lot of talk in morality about the obligation to tell the truth, but little or none about an obligation to make it easy for other people to tell the truth.

    • Nabil ad Dajjal says:

      Thanks for the interesting link.

      “Confused artificial insemination” is one of the weirder concepts I’ve ever heard of. It reminds me of the old idea that one member of a firing squad would randomly be given a blank so that each member could have plausible deniability about whether they had actually shot the condemned man. The deniability is much less plausible here though: if you’re so infertile that mixing in donor sperm sounds like a good idea, the chance that the kid is yours is basically zero.

      Anyway, putting aside normal questions of the morality of lying this seems like a big violation of medical ethics. Sperm donor are screened, to a ridiculous degree in fact, but there was always a real chance that something in her real father’s family history would have been medically relevant. Even just the fact that she was only half Ashkenazi Jewish is important for her to know, because that changes her risk of having or passing on certain genetic disorders. Plus there’s the risk that she could have unknowingly had children with her half-brother, which is rare but has happened before with the children of sperm donors.

      It’s understandable why her parents lied to her. They were given a flimsy excuse to believe that their daughter was theirs and almost certainly didn’t understand the risks they were taking. But the doctors who encouraged that deception were gambling with her health and any of her potential children’s health.

      • John Schilling says:

        Meh. Object level, it has almost always been the case that any family-history medical data carries the caveat, “…as best we know, but of course the father could have been the pizza delivery guy”, and doctors should have always been taking that into account. Almost always it makes no difference, because how often is a treatment decision really changed on the basis of paternal medical history even when we’re sure about paternity? The sort of genetic diseases where that would be dispositive, are fairly rare.

        Here, we have a set of related cases where doctors and parents could have exploited the circumstances of assisted reproduction to provide children with the benefit of a more-reliable-than-normal medical history, but chose instead to provide the psychological benefit of a link between biological and environmental paternity. That may be a marginal benefit or a marginal loss if we do the consequential math perfectly (spoiler: we can’t), but it’s not a qualitative game-changer on either side of the equation.

        • Nabil ad Dajjal says:

          The best estimate I’ve seen for the rate of false paternity in the general population is 2%. One in fifty isn’t a small number, and doctors need to expect to run into it fairly frequently given how many people walk through their doors. But by the same token that means that 98% of the time family history is going to accurately reflect the patient’s parentage.

          In this case, the doctors knew the father with near-certainty and because he’s a sperm donor they had his complete family history on file. If they shrug and say “not my problem” figuring that other doctors will sort it out later, as you suggest, then the patient is in the same bucket as the unlucky 2%. If they take their oath to do no harm seriously and give the patient access to their own accurate medical history, the patient is in the same bucket as the other 98% of the population.

      • albatross11 says:

        It sure seems like the probability of accidentally marrying your half-sibling via sperm donation would be really low, unless you had a very small pool of donors relative to the set of babies. Wouldn’t the same risk exist w.r.t. adoption, in cases where you don’t know your biological family?

        One place where you could imagine this happening would be in those cases where the fertility doctor used his own sperm for all his patients, so that there ended up being hundreds of his kids living in the sdame city.

        • Deiseach says:

          It sure seems like the probability of accidentally marrying your half-sibling via sperm donation would be really low

          Try reading this story, here’s the ending to it:

          We’ve even joined a Girl Scout troop with another of Aaron’s bio-kids who is the age of my youngest daughter and lives about an hour away. …Who knows how many more of Aaron’s bio-kids there are – he’s estimated there could be as many as 67.

          They seem to be easily locatable once anyone starts looking, so they must be in the same general area?

          (I started off this thinking the ex-wife was being unreasonable but by the time I got half-way through she sounded like the only sane person in the entire thing, I just wish the sisters could be kept in contact though I see why the ex-wife wouldn’t want the younger daughter dragged into this nest of crazy).

      • AlphaGamma says:

        It reminds me of the old idea that one member of a firing squad would randomly be given a blank so that each member could have plausible deniability about whether they had actually shot the condemned man

        I have heard that Japan uses a similar system for executions, where multiple people press identical buttons simultaneously, one of which opens the trapdoor below the gallows.

  13. What do you make of Taleb (Black Swan man)’s commentary on IQ? Is he right? Wrong? I know too little about statistics to really grasp what he’s saying here.

    EDIT: It seems like he’s saying that noise can completely obscure whether IQ correlates with performance. He makes a graph showing it correlating only with negative peformance, but then adds noise until it becomes fuzzy enough that it’s similar to an IQ vs SAT scores chart, which are supposedly well correlated with positive performance. In other words, he’s arguing that a test created to show deficits (left tail) is being used to extrapolate higher intelligence (right tail).

    • WashedOut says:

      1. As much as I like his fundamental style, I don’t think the way he conducted himself on twitter was conducive to people understanding his points. Probably should have linked to his medium article and a few other peripheral pieces and left it at that.

      2. Not an expert but I tend to agree with his points of criticism about the statistical basis of IQ.

      3. I think his invocation of “racism/eugenics” as a motivating basis for IQ as a concept was incredibly poor, both strategically (for making his argument) and scientifically.

      • Statismagician says:

        +1. I don’t necessarily agree with specifically the ‘and scientifically’ part of your #3, having not looked into it sufficiently and knowing that the past was often pretty bad about this sort of thing, but except for that minor thing I’m on board.

      • dick says:

        That was similar to my reaction, the article seemed more like a snooty dismissal of IQ than a principled objection. I’m broadly on his side already, but I can’t imagine someone who’s not being persuaded by this mish-mash. Lots of unsourced statistics and graphs, flitting around from point to point, sometimes not even bothering with complete sentences. I think the nicest thing you can say about this post is that it could turn in to a good essay with some more work and an editor (and since he mentions a forthcoming book, I assume that’s what will happen).

    • The Nybbler says:

      It’s mostly nonsense, substituting snarling for argument. It’s true that Binet’s original IQ test was intended to determine students who needed remedial education. But Terman’s version — the “Stanford” in Stanford-Binet — was intended to measure both ends. And modern IQ tests (including the current Stanford-Binet test) do the same. Note that Taleb talks about IQ as if there was a single test for it, which betrays either ignorance or intent to confuse.

    • brad says:

      I think there are at least a few strong points in there that should cause updates until and unless they are convincingly rebutted. The strongest to my mind is the critique of the age, power, lack of replications, and statistical sloppiness of seminal IQ studies (including validation studies).

      It’s curious that IQ enthusiasts often overlap with bitter critics of other types of psychology research but don’t ever seem to approach the papers in their area of enthusiasm with the same level of skepticism.

      • Statismagician says:

        If there were good enough metrics, looking at the correlation or IQ enthusiasm, measured IQ, and actual intelligence would be really interesting.

        • brad says:

          My sense, built in part on prior SSC surveys, is that among people that report having high IQ test scores, after excluding the pure liars and the online test reporters, the vast majority of the remainder took childhood IQ tests and aren’t clear on the distinction.

          • theredsheep says:

            I reported the IQ test I took in high school, at age 15. I think it was called the WEIS-III or something. I don’t know whether it was child or adult.

          • Plumber says:

            I just put down “100” as I don’t know how to get an “official” test, and I have little inclination to expend much effort find out.

            As I never took the SAT I left that blank, but if I knew what the average was I’d have put that.

            If our host doesn’t want to administer a test himself, or link to one that’s what he gets.

          • brad says:

            We are getting off topic, but why not just leave it blank? That’s what I did for the IQ test questions since I never took an adult IQ test (and have no interest in doing so).

            It looks like that test starts at 16 and is called “Wechsler Adult Intelligence Scale”, so I’d say it counts as an adult test.

            Out of curiosity, do you know why you took it? I had one in elementary school for the purpose of admission into a gifted and talented program, but by the time high school came around tracking was more about achievement than potential.

          • theredsheep says:

            Well, that’s a funny story. In sophomore chemistry, the teacher asked if anybody knew the composition of gunpowder. I raised my hand and said saltpeter-charcoal-sulfur, fifteen to three to two (quoting almost verbatim from a Jerry Pournelle novel I’d read multiple times). The teacher was impressed, and a smart-alec classmate turned to me and said, “how do you know that? Are you maybe planning something?” I rolled my eyes and said (this was several months before Columbine), “that’s right, genius, I’ve got me some black powder and a muzzle-loader, and I’m gonna climb the bell tower and blow away all the little ants down there.”

            I’ll give you three guesses as to how that led me to a situation where I received an IQ test.

          • I reported the figure on the list of student IQ’s that one of my fellow high school students obtained. I was probably 15 at the time, but the test could have been taken at any age from that down to about five, since I was in the same school through most of K-12.

            I was aware of the problem with IQ tests at young ages but that was the only score I had and it wasn’t an online test or the like, which I think is what the question warned against.

            The question in the future should contain something about the distinction between child and adult IQ scores.

          • Plumber says:

            “….why not just leave it blank?..”

            It was my best guess (and an employer had administered one when I was in my very early 20’s and told me the result was “about average”), I suspect that my number will be drowned out by over-inflated self reporting anyway.

          • I suspect that my number will be drowned out by over-inflated self reporting anyway.

            That would be relevant if the question was the average IQ of responders. But the more interesting question is the correlation between IQ and other things. If everyone else is inflating his IQ, you should inflate it too by a similar amount, in order to keep the correlations more nearly correct.

            Besides which, 100 is an implausibly low estimate of your IQ.

          • Plumber says:


            “….100 is an implausibly low estimate of your IQ.”

            That’s very kind of you but, given my forgetfulness and that I can’t do arithmetic sums and drive at the same time, I think around 100 (if not less) is about right, if not then I would assume some sort of “grade inflation” is effecting IQ scores.

  14. Sniffnoy says:

    So, saw this on Reddit, this could be big — according this paper, some biologists figured out how to (via genetic engineering) improve the famously inefficient type of photosynthesis (C3) used by many plants (a grouping including many food crops), yielding pretty large increases in the plants’ (in the study, tobacco) biomass…

    • Brett says:

      That’s fantastic news, if they can duplicate it. I know they’ve been trying to create staple crops that use C4 photosynthesis instead of C3 for a while now without success, but it’s also a huge boon if they can greatly improve C3 Photosynthesis in crops.

  15. johan_larson says:

    Have you played Pandemic, the board game? Did you like it? Want more?

    Pandemic Legacy: Season 1 expands the game into an interconnected series of 12-24 games, with player decisions in earlier rounds having consequences in later rounds. Pandemic Legacy: Season 1 is currently the second most highly rated game on

    I am assembling a group of players to play this highly praised game. The plan is to play through the whole thing, start to finish. If we play once a week, it might take 12 weeks.

    I’m located in Toronto. Ideally, we could play the game here in town. Travelling some distance might be possible.

    Contact me if you’d like to take part in this grand campaign. My rot-13-ed email address is wbuna.t.ynefba@tznvy.pbz .

    • jgr314 says:

      If you decide to allow remote players/some form of PBeM, I would be interested.

    • John Schilling says:

      We played that in real time across 2016, and I envy you the experience that awaits. I highly recommend role-playing it, and keeping a fictionalized campaign log.

    • LesHapablap says:

      Would this game be suitable for a group that never plays board games? Would you want to have played Pandemic first?

      • johan_larson says:

        Would this game be suitable for a group that never plays board games?

        Pandemic is pretty straightforward as board-games designed for adults go. I expect virtually anyone could pick it up with an explanation and a couple of games to understand how things work. Pandemic Legacy builds on the original game, but things ramp up gradually.

        Would you want to have played Pandemic first?

        It would make sense to play at least a couple of games of Pandemic first.

        • Doctor Mist says:

          It would make sense to play at least a couple of games of Pandemic first.


          You don’t really need this to master the mechanics, which as you say are pretty simple. (Matt Leacock deserves every bit of royalties from every single one of the growing family of Pandemic games, just for inventing that mechanism.)

          But play the base game a few times to make sure you would enjoy the experience, since it would be a shame to get stuck in a campaign if it just creeped you out.

          My wife and I play Pandemic a lot, and we are eagerly awaiting Season Three; like John Schilling I wish I could do the first two again (and Season Two is even better than Season One). But there was a while, during the Ebola crisis, when I just couldn’t play it at all.

          Additional tip: BoardGameGeek has a terrific site with spoiler-free clarification of things that come up each month. It’s safe to click any month if you are about to start playing that month. Keep it handy.

          Have a terrific time!

        • LesHapablap says:

          I purchased the original and Legacy: Season 1 and will give it a go.

  16. Deiseach says:

    This just proves nobody reads the terms and conditions 🙂

    A firm awarded a government contract to provide extra ferry services has used website terms and conditions apparently intended for a takeaway food firm.

    Seaborne Freight was given the £13.8m contract to run a freight service between Ramsgate and Ostend in the event of a no-deal Brexit.

    Its original terms and conditions advised customers to check goods before “agreeing to pay for any meal/order”.

    …The government has been criticised for awarding the contract to a firm which has never run a ferry service and has no ships.

    • Erusian says:

      There are service companies whose sole trade is generating those. Their biggest competition isn’t people hiring lawyers to make custom ones. It’s people copying generic terms.

  17. rahien.din says:

    A different take on metis-versus-modern : The Termite and the Architect

    • Statismagician says:

      Very interesting article, thanks for sharing.

      Is ‘expert tries to understand something natural, fails’ really a different take on metis-vs.-episteme just because it didn’t happen to explode horribly this time, though?

      • rahien.din says:

        I think it’s a different take because : rather than imposing a sterile quasi-mathematical order, the architect’s intent is to emulate an organic system… but they still end up impose an alien idea thereupon.

  18. brad says:

    Somehow with the switchover to https my old login stopped working and I had to reregister. Did that happen to anyone else?

    • Randy M says:

      Yes, and I couldn’t figure out for a day how to reset it, since the e-mail with the link appended a couple extra characters, making the link invalid without using copy/paste (rather than clicking).

      I was going to take it as a sign that I should do more productive things, but I couldn’t leave your reply here unanswered. So, thanks for killing that resolution. (joking)

  19. Statismagician says:

    Has anybody else read this report on racial differences in home valuation from the Brookings Institution?

    The authors do a regression and find that homes in majority-black neighborhoods are undervalued by ~$48,000/home as a national average controlling for home quality, neighborhood amenities, crime rates, anti-black sentiment, and various other things; they conclude this is because of systemic racism.

    On the one hand, the Brookings Institution is much smarter than I am, and generally knows what they’re talking about in my opinion.

    On the other hand, without detailed methodological information a lot of what they did seems very questionable. Of particular note: I think they model public transit access incorrectly (they do it as always positive everywhere, I think, where I’d do it only for a subset of specific cities and treat light rail as positive + bus as negative); they don’t seem to include any kind of metric for occupancy rates; their crime variables only include violent and property crimes (and the definitions are kind of weird; I’d call carjacking a violent crime intuitively but I may just be wrong there), and not the nuisance/general quality-of-life crimes* I very strongly suspect are what actually matter for home prices; their measures of house and neighborhood quality are so crude as to be essentially meaningless**; and they generally take counts of things (businesses, restaurants, etc.) when it seems very clear to me that those numbers are useless without some kind of quality weighting (i.e. the shabby burger joint is not as important to home prices as the nice Mediterranean place).

    My read of this is that what they actually did was take a bunch of things which are strongly associated with concentrated poverty, split them apart, and found that they are indeed also correlated with each other and that hey, guess what, concentrated poverty is bad. I may be misreading the report (clearly not really meant for an expert audience and so not including lots of details that I want), and/or I may be unfairly biased against it because it does the very common social science thing where the language of the report is super obviously committed to a particular interpretation of the data but never actually makes an actionable suggestion, and I cannot stand that thing.

    My initial position was that as a matter of statistical fact majority-black neighborhoods are disproportionately bad places to live and so houses in such neighborhoods costing less made perfect sense. This report was not sufficient to convince me otherwise.

    Anybody have thoughts?

    *This is a very serious methodological problem, in my view; violent crime rates really ought to be assessed at the neighborhood level rather than the census tract due to low absolute rates, and drug activity/vandalism/petty larceny are going to be way more strongly associated with people not wanting to move into a neighborhood and also the sort of thing you might notice/care about less after living with it for a while, and therefore decrease pressure to move out of the area from what it would otherwise have been. I don’t have data to support this, to be fair.

    **For example, ‘does the home have a kitchen’ and ‘does the home have gas/electric heating’ are included; I think you have to get way out of the standard range of available home quality before these start being an issue, but could be wrong.

    • The Nybbler says:

      We matched schools to census tracts based on the latitude and longitude coordinates, which are available via the Department of Education. Our approach was to take a 5-mile radius around each census tract and consider every school in that radius as a potential school for that neighborhood. The nearest schools to the tract—including all those in the tract—were assigned to the tract until the cumulative school population in grades 4 to 8 equaled the population of 10-to-14-year-olds in the tract.

      Uh, yeah, too bad that’s not how schools are actually assigned. And why exclude high schools?

      In general, it looks like they took a grab bag of features they claim should fully explain housing cost differences between neighborhoods, found out they didn’t, and claimed racism must explain the difference. This is not very convincing. Consider their “structural characteristics”. They include number of bedrooms, but not size of home or lot. They include year built, but not condition. They include “share of homes with gas or electric heating”, which makes no sense at all to me (is oil supposed to be worse than both?). They don’t include air conditioning, which is significant in many climates.

      They completely failed to control for crime

      Exposure to crime is an important neighborhood characteristic that likely affects home values. Unfortunately, comprehensive data on crime is only available at the county-level, and our analysis did not find that neighborhoods located in counties with higher crime rates had lower property values. We do, however, control for the median age of residents in the neighborhood and the percent of families that are single-mothers with children under 18 living in the home. Both are correlated with crime rates (-.28 and .47 respectively), suggesting that we are likely capturing crime effects in our analysis.

      • Statismagician says:

        It’s done on a subset of cities in an appendix; they claim it shows no significant impact and generalize back up to the main body of the text. As noted, I disagree with their definitions and choice of data.

      • albatross11 says:

        I wonder how many choices they had for which things to include in their analysis, and how they accounted for that in their statistics….

      • Reasoner says:

        Both are correlated with crime rates (-.28 and .47 respectively), suggesting that we are likely capturing crime effects in our analysis.

        Yes, to some degree, but you’re not fully capturing them. Those numbers don’t even add up to 1 (and even if they did, you wouldn’t be fully capturing crime unless the two factors were themselves entirely uncorrelated with each other). If majority black neighborhoods tend to have higher crime, knowing that a neighborhood is majority black will likely provide information above and beyond those predictors which you’ve included. So its predictive coefficient will therefore be nonzero.

    • Hoopyfreud says:

      In related news, quality indices suck.

      Look, there’s no good way to do this study. You can’t establish equivalence between homes in black neighborhoods and homes in white neighborhoods; careful reasoning about the data is much more valuable here than any regression you can do, because a regression necessarily collapses a strongly multivariate issue into a fewer variables with no real justification for doing so. So much information is lost that there’s no version of this study that isn’t subject to valid methodological criticism.

      There are three valid responses to this study. One is to assess the informational content, which is probably more than nil, update accordingly, and move on. Another is to demand access to the original data and assemble your own index, which is idiotic because you only understand how you make decisions, not how other people do. Lastly, you can demand that people not do science this way, but that’s a fool’s errand when everyone and their mother is screaming for “proof” to substantiate their beliefs about how other people behave.

      There is no law that says that causal relationships have to be obvious from the data that you can collect. It’s almost like information theory – the noise (other variables in the decision making process) in the restricted communication channel (observed behavior) is so great that there’s a real limit on information we can extract about the signal (motivation). Quality indices necessarily restrict the channel further.

      • Statismagician says:

        Fair. I really wish we did a better job of teaching statistical thinking earlier/more broadly; possibly then policymakers wouldn’t demand and people wouldn’t accept this sort of thing quite so much.

      • albatross11 says:

        If the average cost of housing is unjustifiably lower in majority black neighborhoods (in ways not explicable by crime/quality of schools/quality of life issues), then wouldn’t that imply that there is a substantial premium being paid by racists to avoid these much cheaper but otherwise perfectly fine houses? And that blacks, who presumably are the least likely people to be prejudiced against blacks, are getting a huge benefit in terms of cheaper housing at the same quality?

        If you believed these results, wouldn’t an obvious response be to look to move into a predominantly black neighborhood, since relative to quality houses in these neighborhoods are a big bargain?

        • Hoopyfreud says:

          I mean, kind of? Because black people tend to live in poorer living conditions anyway, a black neighborhood may not be available for you to move into at a given quality level. That said, poor white families living in black ghettos because it’s cheaper there is a recognizable pattern.

        • The Nybbler says:

          If you believed these results, wouldn’t an obvious response be to look to move into a predominantly black neighborhood, since relative to quality houses in these neighborhoods are a big bargain?

          This sometimes happens (though I would argue usually due to a change in the neighborhood’s desirability due to reduction in crime or a rise in nearby businesses rather than a drop in racism). The result is called “gentrification”.

        • Statismagician says:

          Only true if we’re dealing with undervalued houses in poorer neighborhoods which really are of comparable quality to those in richer ones; if they’re undervalued but also of (sufficiently) lower actual quality, they can still sit empty even at deflated prices. And/or there’s a hump effect; if you try and sell a house at ~80% of its true value but the only person you can find who wants to live in it can only afford ~75% of its true value, you’re out of luck.

      • SamChevre says:

        This does rather remind me of the house I built, which when I looked at a Walkscore (I think–one of those) showed as having a strong positive effect from “easy access to local government facilities.”

        Yeah, right: the nearby “local government facilities” for which it got credit were:
        1) The county landfill
        2) The town transfer station (at the dump)
        3) The pound

        • Statismagician says:

          The walkability metrics are trying so hard to capture a really important concept, and I’m always sad when I think about how bad a job they do.

    • Mark Atwood says:

      It looks to me like they brainstormed lots and lots and lots of possible variables, did an analysis on all them, and then file-drawered away the ones that didn’t have good enough fake significance metrics, classic pre-“Why Most Published Research Findings Are False” social faux-science, good for generating headlines and demands for more funding and demands for social justice.

    • edmundgennings says:

      There are a host of fairly obvious challenges with such an endeavor, many of which have already been observed. While some of these were addressed to some extent, they exerted nowhere near the massive amount of effort to actually overcome these methodological difficulties. If one group tends to have different average than another and the same variation, then unless one measures all variables with no error(which is basically impossible to do with house values) then one would expect omitted variables and errors in variables to result in the group that is higher being reported as being higher even after attempting to control for all variables. This was a functionally impossible task and I think a good and unmotivated social scientist should have known that.
      The main impact of this report is for me to trust the judgement of the Brookings Institute somewhat less.

      • Hoopyfreud says:

        The problem is the universal demand for these metrics. “We paid a bunch of people to become intimately familiar with housing markets in [places], and they’re pretty sure the differences in pricing between places in black and white neighborhoods are partly due to ‘living with a bunch of black people’ being a negative contribution to that home’s value” doesn’t persuade people. At least (in my opinion) not as much as it should.

        • idontknow131647093 says:

          The thing I find incdedibly suspicious is two things:

          1. Handwaving away crime. A 12 year old could tell you that this is such a massive factor in home choice that, if fully taken into account, would likely overcompensate for the differences, and make it seem like black housing is overpriced.

          2. No direct comparables for neighborhoods. Where is the test case for their regression? The White neighborhood with the same schools, SQfootage, crime, etc where the housing costs are $40k more than the equal black neighborhood? In their massive database could they not find such a comparable? Is there no 90% white neighborhood that has a 90% black neighborhood with similar stats? If so, isn’t the whole model basically tarot cards?

        • edmundgennings says:

          There is certainly a general problem of the public buying into these kinds of things but the role of social scientists is to resist this temptation. Now every one wants to publish and the temptation to sacrifice best practice for publishablity is a real one even in the best of circumstances.
          That would be an interesting style of evidence and I like the idea. That highly subjective form of evidence would have substantial weight if it included people from a variety of prior positions of whom a good number changed their beliefs in one direction after intensive study. if people stick with their initial position or merely refine it then that is much less convincing that that position is true. But proving this style of inquiry to have actually been well done is very hard.

    • Reasoner says:

      anti-black sentiment

      How is this different than systemic racism, exactly? At the very least we’d expect anti-black sentiment to be very strongly correlated with systemic racism, right? The authors write that “we are likely capturing crime effects in our analysis” by measuring single motherhood and fraction of young people (!) So why shouldn’t anti-black sentiment (commonly known as “racism”) count as “likely capturing” systemic racism effects?

      Does the Schelling segregation model count as systemic racism?

      Seems like systemic racism is a sort of “god of the gaps”. If there’s some interesting race-related phenomenon which we don’t know how to explain otherwise… it’s probably a result of “systemic racism”! Too bad no one has a model of what “systemic racism” actually is that would allow us to measure it as its own construct.

      Also, how do the authors explain the gentrification of black neighborhoods such as Harlem? Seems like if rich people DON’T want to buy a home in your neighborhood, and home prices are low, that is systemic racism. But others say if rich people DO want to buy a home in your neighborhood, and home prices are rising, that is another form of systemic racism! There’s no way to win.

  20. Statismagician says:

    What’s our collective take on modern art? Feel free to use whatever definition makes the most sense to you, but please note what that happens to be.

    I would start, but the catalyst for this is that I realized I couldn’t actually articulate what I meant by modern art beyond ‘I know it when I see it, and don’t generally enjoy the sight.’

    • The Nybbler says:

      IMO, modern art usually refers to either conceptual art or other non-representational art. And it pretty much deserves its bad reputation.

    • j1000000 says:

      Well, if we’re talking about modern art like Van Gogh and Matisse and Monet, then I like it. (Not that I spend much time at museums but I do enjoy it when I go.)

      If we’re talking Duchamp’s Fountain and its ilk (and then all the weird postmodern/contemporary stuff today) — it’s fun when you first hear of it and get to have the “well what IS art, really” conversation, but I don’t actually like it.

    • Le Maistre Chat says:

      Art that exists in an arms race of abstractness.
      My thought: it should be illegal to spend tax revenue on this trash. If the rich want to pay lazy artists, that’s their business (and whether the commons want to Kill the Rich for being so corrupt is theirs).

    • John Schilling says:

      Art is generally some combination of depictive, decorative, or manipulative and particularly in modern usage “Art” is taken to exclude all that depictive/decorative crap as crass commercialism. In which case, I have little use for purely manipulative art, laugh at people who use it to extract status points from one another, and vehemently reject demands that I pay for any of it in money or status.

      I am willing to pay for, respect, and admire aesthetically pleasing decorative works, and there’s a fair bit of that being produced in the 21st century even if it isn’t usually labeled “modern art”.

    • bean says:

      I’ll defend some modern art as probably not totally pointless. There’s a lot of interesting thoughts implicit in moving past pure representation, and if you can find someone who likes it and is willing to communicate about it in a reasonably clear matter, it’s a fun way to pass an hour or two.

      But at the same time, the field as a whole reminds me of Chip Morningstar’s thoughts on deconstruction:

      Deconstruction is an idea that would make a worthy topic for some bright graduate student’s Ph.D. dissertation but has instead spawned an entire subfield. Ideas that would merit a good solid evening or afternoon of argument and debate and perhaps a paper or two instead become the focus of entire careers.

      The same thing (including the evolutionary race to greater heights of weirdness) has taken place in the modern art world. (I’m sure that there are loads of books on the parallels between modern art and deconstructionism in the parts of university libraries I never go to, now that I think about it.)

      • Björn says:

        Interestingly enough, Deconstruction is not a philosophical equivalent of (radical) Modern art. Deconstructing the Mona Lisa means asking yourself what the painting shows or what it is, but also playfully looking for counterarguments. If you think that the Mona Lisa is a portrait of Lisa del Giocondo, like some art historian say, then you can also imagine (for fun) that the Mona Lisa is the portrait of no one in particular. This would make all the guessing about who it is quite futile.

        Now you can find such thought experiments silly, but they engage with the actual things quite much. (Radical) Modern art is about creating artworks that do not relate to anything. (Radical) Modern philosophically is much more related to someone like Adorno, who says no poems after Auschwitz and who thinks that 12 tone music is the only acceptable music in the 20th century.

    • Björn says:

      Modernism, which is the art historical epoch to which Modern art belongs, is characterized by the rejection of traditions. How much traditions are rejected actually varies. Some art styles that can be called Modern try out new things, but in a traditional framework:
      Cubism distorts the form of all things that are painted, but the things that are painted are in theory very traditional. 12 tone music wants to create the ultimate antiharmony, but the composers still write operas. Fascist art has a very stereotypical “strong” look, but it relies strongly on symbols from the past.

      Others are much more radical in their rejection of tradition: John Cage composes random music for things that are not musical instruments. The Abstract Expressionists start with painting pure abstract forms and end with Jackson Pollock making random paint splatters. Bauhaus means turning everything into cubic shapes. Dadaism is meme culture 1910 AD.

      Because of quite complicated reasons, literature, music and cinema always kept the tradition in some way, while in the visual arts tradition was destroyed almost completely. This is the reason why visual art (or at least the art that is respected in the “art scene”) nowadays consists mostly of bizarre stunts and anti-art. This is why it has become strangely marginalized, because many people simply can not care about it.

      My personal opinion on Modernism is twofold. I like the aspects that keep some connection with tradition, but at the same time explore territory that Western art from 1500 to 1900 ignored, which was mostly about showing things realistically. Abstract or stilized art was always part of what humans can do, just look at Ancient Egyptian paintings. But I hate the aspects that are all about being “in” while hiding that the artist has no actual skills.

      • Aapje says:


        To add to that: quite a bit of traditional 2D art tried to be beautiful in commonly accepted ways, like by mimicking reality, by telling stories, by showing creative alternative interpretations, etc.

        Postmodern art often tries to make a single point, which makes it fairly shallow and repetitive. Ceci n’est pas une pipe by Magritte made a decent point that the 2D image is not the object itself, but merely a partial representation where the painter chose what to present. However, once one gets that point, the painting itself is boring and bland, with no additional messages. Furthermore, repeating the same trick with different bland pictures adds little enjoyment.

        John Cage’s 4′33″ literally has nothing beyond its postmodern point.

        But I hate the aspects that are all about being “in” while hiding that the artist has no actual skills.

        Indeed. I think that one of the main reasons why so much modern 2D art sucks is because talentless hacks are rewarded and thus proliferate. If you focus on the top 1% and forget that the rest exists, there is some nice stuff.

        Of course, there was a lot of crap in the past as well. Although much of it was probably not even intended as art, despite us now seeing it that way.

      • The Nybbler says:

        You could just as easily write on a note card, “Hey, what about a blue square? That’s art now, right?

        Ah, you’re familiar with the work of Sol LeWitt then.

        This is a bit unfair to LeWitt, but he did write instructions for his wall drawings, which are re-created at every installation. He also did a work consisting of a blue square, but it was a blue square and not just a description of it.

    • SamChevre says:

      Modern visual art: I’ll define it as “focused on visual interest rather than representation.”

      I like some of it: geometric shapes in bright colors, interesting color work, abstract sculpture…it doesn’t have to look like something to be interesting to look at. I don’t enjoy art that requires an explanation to be interesting.

    • BBA says:

      At what point does “Modern” cease to be modern? The Armory Show was more than 100 years ago…

      I like Andy Warhol, and the simultaneous sincerity and irony of the work he did. Most contemporary art can’t manage that, it’s either got a million layers of irony with nothing underneath or it’s just blunt and unclever. But I guess not everyone can be a Warhol.

    • Another Throw says:

      Modern are is a highly specialized industry to create intrinsically unappraisable vehicles to be used as tax deductions. When dealing in art, if you don’t know who the sucker in the room is it is you.

    • tossrock says:

      Surprised no one has mentioned the CIA connection yet:

    • Mr. Doolittle says:

      I thought most art, modern certainly included, was a way for the ultra-wealthy to store value? In that case, it is essentially analog Bitcoin for the richest people. I don’t recall any details, but I feel like I heard there were positive tax implications to buying art as well?

  21. Edward Scizorhands says:

    Is there still time to take the SSC survey?

    • RavenclawPrefect says:

      Should be, so long as it hasn’t closed; IIRC, it’s usually left open to collect responses for a few weeks. You can find it here, and if you haven’t had enough after that, the supplemental survey here.

  22. Walter says:

    Here is a super dumb question.

    So, if you are on facebook or other social media, you will receive the occasional friend request from someone you don’t know. This person joined just a few minutes ago, put an attractive woman’s picture as their profile, maybe uploaded some other pictures and then started shooting out friend requests. They have mutual friends if and only if you know a lot of really thirsty people.

    So this is obviously a bot, right? My question is…what’s the end goal?

    Like, say one of these bots gathers a few thousand friends, what is next? What is the goal of whoever operates it?

    • dodrian says:

      Data-mining (how many people have an email/phone number on their profile?), phishing, getting you to click on virus payload links, getting you to send compromising photos then blackmail for bitcoin – I don’t know which of these strategies they go for because I’ve never accepted, but there’s plenty of scams that can benefit from a mineable facebook profile.

    • Kyle A Johansen says:

      I don’t know about this issue, but they might simply want friends to seem genuine. I think sock puppets sometimes work like that. Or they might not want a thousand friends, they just want the dupe who will help her out with her frozen Nigerian bank account and the mass-span is obviously a necessary step..

    • Statismagician says:

      The probably-correct answers are already taken, so I’m going with ‘crazy six-degrees-of-separation scheme to connect international spy rings to each other through friend-of-fried-visible social media posting.’

    • My father is a probation officer who runs one of these operations to catch people posting probation violations on Facebook. Of course, it’s vanishingly unlikely that any of them read Slate Star Codex.

  23. ahenobarbi says:

    I think the Less Wrong link is a bit too lively for the Embalmed Ones section.

  24. Brett says:

    1. I’ve been trying to find the perfect ball point and gel pens. I’ve bought a bunch, but what do people prefer that won’t totally break the bank (i.e. the pens don’t cost more than $20/pen)?

    2. Thinking about the “landfills running out” non-crisis got me thinking about density and population size. If the entire world population (rounded up to 8 billion) lived in LA-level density, we’d occupy a space about 2.443 million square kilometers, or about 50,000 square kilometers larger than Algeria. That’s LA, too – boost it up to Brooklyn level density, and the Planet-City would cover an area slightly larger than european France.

    But we do tend to spread out as much as growing denser. It makes me wonder if space colonies will largely just spread out once the technology is there for self-sufficiency, rather than concentrating so they do massive mega-projects.

    3. If climate change smashes a bunch of crops, we might be eating a lot more lettuce, tomatoes, and potatoes. Those three grow pretty dense and can be grown in greenhouses.

    • T82 says:

      1. I’m not a pen expert by any means, but my favorite pen is probably the UniBall Vision “Fine” 0.7mm pen. The UniBall Vision Elite “Bold” 0.8mm is also good, maybe a little better, and airplane-safe. Neither has a padded grip, which I actually like, since I have a firm grip and a padded grip tends to sqwoosh around and feel awkward to me. Both pens have a lot of flow and lay down thick lines, so I can write quickly, without much pressure on the paper, and with a sort of accent on my lettering, if that makes sense.

      Both will leave fantastic ink smears on the side of your hand if you’re left-handed.

      Apologies if you’ve already tried them, since they’re fairly common. But I like ’em, and you could probably get 5, 10, maybe 15 of them for $20 about anywhere. I was in a situation once where I had to rely on cheap thin bic ball point pens for a month, and I kept wishing I had something like a Uniball vision. They can be a pleasure to write with, even if you’re just filling out forms.

      3. Can zucchini squash be grown with some density in a greenhouse? I’m not sure I could ride out near-apocalyptic climate change conditions without those.

    • Mark Atwood says:

      For “the perfect ballpoint”, someone I know goes through the boxes of BIC Cristals in the supply cabinet at her work, and picks out the ~3 out of each box of 20 with a smooth perfect action, and hoards them.

      Another option that works well is to go to a Kinokuniya, and start sampling the Japanese ceramic ball pens for ones with a smooth perfect action. My experience is that about 4 out 5 of Japanese “disposable” pens are perfect. If you don’t have a Kinokuniya nearby, or if Tokyo is not on your travel plans, is buy boxes of Japanese ceramic ball ballpoint pens from, or else just buy German LAMYs.

    • TDB says:

      Costco sells (used to?) Pilot G-2 in packs of 15 or 20 for cheap.I like ’em. I have a couple of fancy pens that by luck accept the same cartridge as the G-2. I’m not really a connoisseur, so YMMV.

    • rahien.din says:

      Pen Addict’s Top 5 Pens post is a great start for knowing which pens to try first.

      You can find a ton of great information and buying options on JetPens.

      My favorites :

      Cheap and ubiquitous : Pilot G-2, Zebra G-301

      A little more expensive : Pilot Precise V5/V7, Uniball Jetstream, Zebra Sarasa

      More expensive : Uniball Style-Fit series is my go-to. I’ve also used the Pilot Hi-Tec C and it’s very good, too.

  25. Mark Atwood says:

    Why is the location of the Jan3 meetup being rendered as “Тиёда” instead of as “Tokyo” or as “東京”?

  26. sfoil says:

    I would be. His book on the Korean War is IMO the best book I’ve ever read on that war on a page-for-page basis.

  27. Humans are a social species. The ability to talk to people and get along with them is incredibly important, evolutionarily speaking. So why are so many people so bad at it? You might think it’s just an arms race. Relatively, there are many people we think of as being unsocial but we are all better at it than our ancestors. Maybe, but look at the rate of autism, where those with it are often unable to communicate effectively with others. Shouldn’t it be as rare as something like the inability to use grammar?

    • TDB says:

      The hunter-gatherer tribe was the prime environment for the development of social skills. Maybe agriculture and civilization started creating lots of sub-niches for the socially inept? Armies need engineers.

    • Statismagician says:

      My immediate thought is that autism is more common than you’d think it ought to be because it’s an extreme position on a population-net-beneficial spectrum roughly corresponding to good-at-tools, but on second thought I don’t actually recall why I think this and so may very well be completely off-base,

      • I thought about this but it still seems wrong. For one thing, don’t autistic people have lower IQ than average? With IQ being a significant predictor of success, it doesn’t seem like they would have thrived.

        But even putting that aside, a smart person with autism should generally lose out to people without it. I don’t see how they would have had any special advantage compared to anyone else but they do have a significant disadvantage.

        • TDB says:

          Success and thriving are not necessarily the same.

          a smart person with autism should generally lose out to people without it.

          Emphasis on “generally.” A specialist can flourish in a niche. E.G., before computers (as we know them), people did calculations and computations. What sort of person is good at doing math all day long?

          • There might be a statistically significant link between autism and useful skills. That’s quite different than saying that they flourish at it more than other people do. It seems to me that any slight benefit autism gives would be drowned out by the debilitating nature of lack of social skills.

          • abystander says:

            More like the genes associated with autism can also be beneficial in certain circumstances. I got a book this Christmas claiming prodigy’s often have autistic family members.

            It’s like having one copy of the sickle cell gene provides resistance to malaria, but two copies leads to a lot health problems.

          • Maybe. But I think it goes beyond autism. Look st something like language. Pretty much everyone is an expert in the ability to use grammar in their native language. Most people don’t make the mistake of saying something like “two thing” instead of “two things”, even though most language learning is not explicitly taught.

            Compare that to social learning where if you don’t practice a lot, you’re probably going to be really bad at it. Teenagers spend so much of their time uncomfortable because they’re trying to figure out how to socialize. Based on how evolutionary important socializing is, it just seems like the kind of thing that should now be picked up naturally rather than a skill that needs to be extensively practiced.

        • Statismagician says:

          I think you’re looking at this on an individual level, and I was thinking about population trends – i.e. for x/100,000 cases of autism, we get 10x/100,000 people good enough at math to make a competent engineer, or something like that. This is not a factual claim, just something I picked up from somewhere as mentioned above.

        • Aapje says:

          @Wrong Species

          People with autism don’t have to thrive or have an advantage for autism to exist and be common. Cancer also exists and is common, yet seems to never be an advantage.

          All that is necessary is that it is hard to get things right, for things to often go wrong. Given the many manifestations of autism, it may just be that the human cognitive system is fragile and easily disturbed.

    • rahien.din says:

      Algernon’s Law.

    • Levantine says:

      Humans are a social species. The ability to talk to people and get along with them is incredibly important, evolutionarily speaking. So why are so many people so bad at it? You might think it’s just an arms race.

      I’m puzzling over the arms race reference.

      Toward answering the question why are so many people so bad at it: perhaps there are certain societies, social circles, geographical places and historical periods in which such people cluster.

      My first thoughts are that

      > bad schooling typically acts actively against the development and retention of such skills. I think so on the basis of my personal observation, Keith Johnson’s observations as outlined in Intro, John Gatto’s, etc. A simple example of how it’s being done is students being interrupted in the middle of the sentence, in order to be “corrected.” It’s based on no good reason. It’s an aggression, pure and simple.

      > Looking at modern societies, it looks they are ultimately indifferent to the well-being of the majority of people, in the generations currently living or the future ones. Quality communication is neglected in too many places, too often. What I just said is wrong insofar groups of people around us gently help the introverts to join social circles, insofar we have friends when we need them… It is my understanding that in many, many schools in the US, those things have started to happen on a regular basis, that the culture of social relationships has markedly improved in the US schools in the past 15 years.
      That still leaves out many people who spend their lives without contact with such environments.

      • Ketil says:

        I’m puzzling over the arms race reference.

        Maybe: Smooth talkers get the mates, so it makes sense for potential mates to try to sabotage the smooth talking in order to select the smoothest of the smooth?

  28. Le Maistre Chat says:

    New D&D thread: how would you balance heroes with spellcasters?
    I’ve seen some suggestions in the first thread about making it so a character can only cast a large handful of spells in their career (Call of Cthulhu, etc). How about raising the power of heroes to mythic levels?
    Most familiar to Westerners would be Samson, who when the Spirit of God came upon him broke the ropes binding him and slew a thousand Philistines with a convenient donkey jawbone (Judges 15:14-15).
    So at will, a hero of a certain level should be able to do minimum damage equal to a common soldier’s HP and Cleave the 8 soldiers surrounding him — indeed he should be able to high jump over an enemy and jump kick the one behind him to death as part of his move, or he’ll rarely ever be encircled by 8 enemies.
    Note too that a hero should not be forced into “weapon specialization” to be this badass.

    Even at that, this would take 111 combat rounds to play out, so God was probably Hasting Samson to be able to move plus attack at least twice per round.

    Another Western example would be Cu Chulainn, who used a combination of his superhuman power and a choke point (a ford) to act as Ulster’s entire army against the army of Connacht.
    What level of caster would these guys be balanced against?

    • Hoopyfreud says:

      Well, there’s the eternal standby…

      In all seriousness, Exalted and superhero games do this fairly well. D&D Totally not D&D… sort of supports this, but in a way such that the game kind of breaks.

      • Le Maistre Chat says:

        I don’t know anything about Exalted, but it only makes sense that an RPG where D&D-scale magi are balanced with non-magical heroes would be a superhero game.

        • Nornagest says:

          I get the impression that Exalted draws most of its inspiration from wuxia, like Hero or Crouching Tiger, Hidden Dragon, and from action anime like Yu Yu Hakusho or Bleach. But it ends up being its own weird little thing most of the time. It’s not bad — flavor-wise I like it more than most of the WoD lines, and while it’s easy to break the system over your knee with a little bit of effort that’s true for anything White Wolf ever produced — but it comes off less Eastern and more “written by Western weeaboos who don’t understand that the cultural references in the source material were cultural references”.

          • Le Maistre Chat says:

            Yeah, that all sounds meh. It doesn’t help that it has a baked-in world (“Creation”) I would never want to use.

          • bullseye says:

            I heard that “let’s make it Asian” was a last minute decision. Maybe the wuxia elements were initially intended as superpowers.

          • Nornagest says:

            It would make a fair amount of sense refluffed as a medieval superhero game, but that’s a rare and risky enough concept (who wakes up in the morning and says “I wanna play Superman, but on a horse with a sword”?) that I think Occam’s Razor suggests there was always some Eastern or wannabe-Eastern influence.

          • beleester says:

            I think it’s mostly Greek myth, not medieval fantasy. It’s basically got an entire mechanic that imitates Achilles sulking in his tent or Hercules going crazy and killing his family, and a lot of the backstory – the new gods overthrowing the old gods, the lost golden age – is reminiscent of it as well.

            The Charm system is definitely wannabe-wuxia – you have supernatural martial arts techniques with elaborate names like “Dipping Swallow Attack” or “Ready in Eight Directions Stance”, but I could believe the causation went the other way. Did they come up with the idea of “Charms” as supernatural versions of skills, and then start giving elaborate martial arts names to them, or did they decide to give characters elaborate martial arts moves, and then come up with the Charm system to implement that?

    • sfoil says:

      My impulse would be to go the other direction, and make spellcasters, or maybe just magic in general, less powerful.

      Although things got out of hand later on, in the original Dying Earth knowing four spells at once was a big deal. That alone would be enough to shut down D&D-style magic power creep. Combine that with a DM as capricious as Vance the author and your wizard will find himself simply unable to rely on his spells — actually casting a spell in combat will be an absolute last resort.

      Of course, there’s no reason for magic to be useful in personal combat at all. Thinking about it and some of the things posted up thread, D&D-esque “combat magic” is probably a quite recent idea. If magic were limited to astrology, divination, demonology and the like then I doubt any band of ruffians out looking for a fight would bring a wizard along unless, like Gandalf, he also happened to be at least somewhat handy with a sword.

    • broblawsky says:

      Just dig up the Book of Nine Swords.

      • Le Maistre Chat says:

        Tell me more about when Book of Nine Swords PCs turn into Samson or Cu Chulainn.

        • broblawsky says:

          Per your examples:
          – So at will, a hero of a certain level should be able to do minimum damage equal to a common soldier’s HP and Cleave the 8 soldiers surrounding him

          The Iron Heart discipline contains techniques like Adamantine Hurricane (attack everyone within reach twice) and Lightning Throw (kill everyone in a 30′ line).

          – indeed he should be able to high jump over an enemy and jump kick the one behind him to death as part of his move, or he’ll rarely ever be encircled by 8 enemies.

          Tiger Heart discipline contains plenty of maneuvers like this.

          – Note too that a hero should not be forced into “weapon specialization” to be this badass.

          Most of the Bo9S disciplines are associated with ~4-5 weapons, and feats and class features associated with the disciplines tend to boost all of them equally.

        • beleester says:

          If you want to be Cu Chulainn, be a Crusader and take a bunch of Devoted Spirit and Stone Dragon. Battlefield control, damage reduction, and a couple maneuvers specifically designed for planting yourself in one place and not letting anyone through, like Roots of the Mountain or Thicket of Blades. Good stuff.

    • DeWitt says:

      3.5 is a ridiculous system and the odds are stacked against me if I want to make warriors on par with spellcasters in it, so we’ll go with 5e.

      For Samson:

      Assume the Philistine army consists of people with a shield, leather armor, and 12 dex for a fairly generous 14 AC in total. I don’t know that any of the ones he kills are named NPCs with higher levels and such, so we’ll make them all mooks; hobgoblin warriors in 5e have 11 hp, so we’ll again be generous and give the very human Philistines 12 hp each instead.

      Firstly, play with the homebrew rule that nat 20s don’t always hit and nat 1s don’t always miss; hardly a ridiculous ruling for any table, given how much 5e fights are left to chance. Secondly, buff the great weapon master feat, giving it a stipulation where killing an enemy lets you make another attack for free, rather than requiring a bonus action.

      Samson is a biblical persona of incredible strength, so I’m comfortable statting him as a lvl 20 barbarian, the class I feel suits him well. His birth is surrounded in divine shenanigans, but he’s otherwise perfectly human, so we’re giving him variant human stats. At lvl 1, the man starts with 16 STR and CON, 14 DEX, and mental stats I’m not really interested in; his free feat is great weapon master, natch.

      Leveling up, Samson takes ASI’s only: 20 STR, 20 CON, and 16 DEX. At lvl 20, his capstone gives him +4 to both STR and CON, which brings him up to 24 for both. I don’t know if Samson had a shield around, but even without one, he comes out having an AC of 20. The donkey jawbone he uses for a weapon is not at all magical, and we will treat it as an improvised club.

      Samson has 24 strength and a proficiency bonus of 6, giving him a +13 to hit; even on a natural 1, he will always hit a philistine clad in leather and bearing a shield, disadvantage or no. He gets +7 to damage from his strength and +4 for as long as he’s raging, meaning that every hit is an instant kill. If you drop the man in a neat tic-tac-toe grid of philistine soldiers, they all die in one round; that requirement of your post is satisfied very easily. Barbarians get +10 feet of movement per round, and humans have 30 to begin with, so Samson can truck forward a lot: if he starts boxed in by eight and just storms on ahead, that’s 24 more people he can take out just by advancing five feet, killing three Philistines, and doing so until he runs out of movement, all for an impressive tally of 32 dead people per round, or slightly more than five a second.

      At 32 kills a round, it takes Samson about 32 rounds to kill a thousand guys, assuming they advance like mindless automatons instead of running from divine wrath taking two and a half dozen of them down every few moments. Incidentally, barbarians after a point can’t roll lower on STR checks than their strength score in total. Samson’s player could say ‘I break these ropes and there is nothing you can do to stop me, because the break DC is lower than my strength score.’ Similarly, he can do the action movie scene where he vaults over a guy to jump into the middle of enemy soldiers by sheer fiat, because the DC for a jump like so is surely lower than 24, too.

      I don’t know how to do the calculation on whether they could take down a barbarian of his level before he can in fact fulfil those 32 rounds. AC is not the high point of barbarian munchkinry, and 20 AC can be breached even by our mooks even if they don’t have highly impressive bonuses to hit. Still, Samson doesn’t need to attack recklessly, and the bible doesn’t say he hasn’t got a shield, which means he might have 22 AC instead; Samson comes in at a cool 285HP in total, takes half damage from all attacks, and isn’t easily wounded whether his AC is 20 or not. Crossbows weren’t yet invented and I doubt their entire army carried bows, though javelins seem possible; the average damage of a 1d6+1 spear or javelin is a meagre 2 on hit, where most by far will miss, so Samson can weather a good deal of punishment before going down. The man has a chance.

      So, doable.

      The real question, of course, is why any of this is even happening. I had to buff an already powerful feat up a bit, and the no crit fails/misses rule supports Samson a lot more than it takes away from him. The obvious outcome is that Samson is a DMPC who had the rules bent specifically so he could get away with this, something I can see the bible doing very well, so we know at least that power fantasy cool guys stealing everyone’s thunder is an old tradition indeed.

      EDIT: as an added bonus, I realised that Samson dies when he collapses a temple. The man has 24 strength, where a 5e frost giant only comes in at 23; if a giant larger than a house can snap a couple pillars, then surely can a divinely-empowered human somehow stronger than that giant.

      • DeWitt says:

        Okay, so, I read up on Connacht, and apparently what went on there was a series of single combats rather than a chokepoint situation akin to the Chanson de Roland. I’m not sure that this needs anything beyond a fighter of decent level defeating NPCs weaker than he, so this requires a lot less BS than the 20th level PC Samson would need to be.

        (Maybe this is why there are no more pagans: the Abrahamics have way higher leveled guys than they do)

        • Deiseach says:

          What I vaguely remember from the Táin is that the stand-off at the Ford was agreed between Medbh and Cú Chulainn because up to then he’d been conducing a successful hit-and-run guerilla campaign as her unwieldy army marched northwards, so by agreeing to a series of single combats between her champions and him, she hoped that they would eventually wear him down and even kill him so she could continue to march north, and one fight a day (even if he killed the champion) meant only one casualty instead of the many casualties he had been causing up till then (and ruining the morale of the troops). On Cú Chulainn’s part, it was to get Medbh’s army to stay camped in one place long enough, and that he could hold out in a series of single combats long enough, until the curse on the Ulster warriors ran its course and they could come into the battle and meet the advancing army.

          The highpoint of this is the three-day fight between Cú Chulainn and Ferdia; imagine if in the Iliad Achilles and Patroclus ended up on opposite sides and Achilles kills Patroclus?

    • dndnrsn says:

      Balanced in the sense that the wizard contributes no more or less to the party than anyone else, or balanced in the sense that everyone contributes something, where you don’t have people saying “no no don’t play a rogue we need another wizard” or parties consisting of two wizards and two clerics.

      3rd ed tried to beef up martial classes by giving them abilities and feats – but the result was that the “point them at what they need to hit” class for newbies became one with various builds, some of which were much better than others. 4th gave everyone Vancian powers, but this created its own problems – suddenly everyone had a reason to limit how much combat they saw in a day as much as possible, and while Vancian magic represents something tangible in the game world, it’s unclear why your fighter can only do his Mighty Strike or whatever once a day.

      I don’t think it’s possible to avoid the situation where the wizard or cleric is a disproportionately large “chunk” of the party’s power. Their abilities bring something new to the table (going from not shooting fire to shooting fire is a different thing from going from attacking once to twice), and some of their spells are game-changers. Even in 3rd, 4th, 5th where everyone had some sort of cool gimmick, “what spells does our wizard have access to” was important enough that the wizard being 4th or 5th level mattered more than the fighter being 4th or 5th level.

      It is, though, possible to keep the wizard in the role of artillery – can’t take the village without them, but “shells kill more than bullets, so let’s field an army of primarily artillery” isn’t a winning strategy. It’s more about the resources the wizard has than anything else. 3rd, 4th, 5th try to make other classes as good as the wizard by giving them shiny stuff, but they also lift a lot of the limits of spellcasting in earlier editions, probably to try and avoid the bits where the casters have no spells or the wrong spells and get to do nothing fun.

      • woah77 says:

        “shells kill more than bullets, so let’s field an army of primarily artillery” isn’t a winning strategy

        Someone didn’t play enough Supreme Commander Forged Alliance. Primarily having artillery can absolutely be a winning strategy, but you need enough entrenchment and range to make it so.

        Otherwise I entirely agree with your analysis.

      • DeWitt says:

        “shells kill more than bullets, so let’s field an army of primarily artillery” isn’t a winning strategy.

        It is the exact winning strategy in 3.5, and it’s achieved by writing down cleric or druid where another person might do fighter or barbarian.

        • Statismagician says:

          Yeah. The historical strategy of I think I fairly say literally every successful army since the invention of gunpowder has been ‘field absolutely as much artillery* as we can afford/supply/protect,’ it isn’t surprising that a game that developed out of strategy games would keep the same dynamic, and now an artillery piece costs exactly as much as an infantryman while not being any more vulnerable to attack.

          *Or as efficient artillery (precision weapons), or as much armored/air-dropped weaponry as possible; the point is that infantry isn’t central to the actually-accomplishing-war chunk of real wars the way it is to occupation duty. And all of these are also things that clerics/wizards/druids do better than fighters.

          • Le Maistre Chat says:

            These artillery pieces are also sensor platforms and invisible scouts starting at Level 3, immune to ranged infantry weapons at Level 5…

          • dndnrsn says:

            I really messed up the attempt at an analogy, didn’t I. The D&D party is combined arms, of one sort or another, with support coming from some cleric abilities.

            I’m not saying that martial classes and rogues don’t get overpowered in some editions of D&D. That happens and it’s bad.

            (And I’d argue with the statement that infantry isn’t central to winning wars; I’m pretty sure that the combat units of countries with first-rate militaries are primarily infantry of one sort or another, most or all were primarily leg infantry in WWII, etc)

          • idontknow131647093 says:

            Has D & D not developed the idea of armor classes?

            Like, we have the term “glass cannon” use it.

          • Statismagician says:

            @ dndnrsn
            That would be where the ‘as much artillery as we can afford and protect’ bit from above comes into play. I’m going to type up a full version of my thoughts on this when I have time, but basically the gist is that the degree to which military planning revolves around artillery (and things which are more like artillery than infantry/cavalry; e.g. armor and air power) has been increasing since military potential began to decouple from muscle power. This is about relative impact, not necessity or counts; obviously I need to keep the enemy from just walking up and taking my shiny pile of cannon, and because [economics] a lot of them can cost less than a few cannon (thought less so than you might think; I believe the French Army circa 1914 was just under 40% artillery but I need to check my sources). This is not to say armies didn’t try other approaches from time to time, it just tends not to work very well – see e.g. Italy in WW1 or Japan in WWII.

            Limitations; about wars and not occupations, less true in 16** and more true in 2019. Probably boils down to ‘the side which can more effectively extract economic products for military affairs wins, barring weirdness’ which is not controversial but does let me expand the theory backwards into premodern warfare (knights, English archers, etc.).

          • bullseye says:

            To answer idontknow131647093, D&D fighters wear the best armor and wizards wear none at all. But wizards have other defenses, everyone has a huge number of hit points sooner or later so it matter less than you might expect. Also clerics get the best armor too.

          • Statismagician says:

            Yep. If you’re past about level 5 and the wizard isn’t some combination of flying, covered in magical forcefields, hidden with illusions, and/or behind a horde of summoned monsters and a magical obstacle or two, that’s a deliberate choice the wizard made and not a dependable situation.

          • idontknow131647093 says:


            Thats just shitty game design then.

          • theredsheep says:

            I don’t do D&D, but IIRC there’s some rubbish about “some spells require complex gestures and cannot be cast if you’re wearing bulky armor as opposed to still-bulky robes.” I think the console JRPG SaGa Frontier II did it better by having metal actually drain magical power out of you–and intrinsically resist enemy magic–so there was a relatively plausible and consistent tradeoff.

          • Nornagest says:

            Yeah, the “complex gestures” thing doesn’t wash, especially when a lot of spells don’t even have somatic components. I’m not a huge fan of the “iron interferes with magic” approach, either — it does have some folklore behind it, but logically you’d just end up with wizards walking around in bronze, leather, or textile (historically very common!) armor if they’re expecting to go into harm’s way, so it doesn’t do what it’s supposed to.

            If I were designing D&D from scratch, I’d probably just not give casters an armor proficiency by default, but let them pick one up if they want, and spec out magical defenses accordingly.

          • theredsheep says:

            Well, in SF2’s case it added up to a neat set of gameplay choices, because they thought it through pretty well; iron equipment is overwhelmingly superior, but magic is immensely important to gameplay and most non-metal gear boosted your total MP if not your MP recovery rate. Physical tech “magic points” recovered more slowly and at a fixed rate. So on the one hand you could have a guy plated in steel head-to-toe who’d be really hard to kill and who hit like a ton, but he’d run out of steam for advanced skills at a rapid rate, and he’d likely have no magic ability at all. On the other, you could give a person all nonmetal, and they wouldn’t be nearly as tough but they’d be able to use the best spells and hybrid weapon techs (lightning-charged swords, etc) almost for free due to their huge recovery rate. And anywhere in between. It wound up as power versus versatility.

            Also steel had infinite durability, but I think durability is an obnoxious mechanic to inflict on a player.

          • Le Maistre Chat says:


            I’m not a huge fan of the “iron interferes with magic” approach, either — it does have some folklore behind it, but logically you’d just end up with wizards walking around in bronze, leather, or textile (historically very common!) armor if they’re expecting to go into harm’s way, so it doesn’t do what it’s supposed to.

            The Broken Sword, Poul Anderson 1954. Elves are a chaotic warrior race who can’t touch iron, so they all use bronze except for changelings like the protagonist. Changelings are a significant enough part of elven society that they enslave dwarves mostly for the purpose of freeing them in exchange for arms they can only get from dwarf smiths.

          • Protagoras says:

            The no armor for arcane casters thing in D&D never made much sense, especially as magical ways to get decent AC were available quite early on, so it wasn’t even really a matter of balance (the low hit points of casters were much more significant in making them fragile).

          • Nornagest says:

            Elves are a chaotic warrior race who can’t touch iron…

            Oh, that’s fine. If it was elves who couldn’t use iron armor (or weapons?) in D&D, but you could substitute with bronze or jade or dragonbone or whatever, that would be a flavorful drawback on about the right power scale to balance out racial advantages. But the intent of restrictions on armor for arcane magic users is clearly to keep wizards in robes and pointy hats (or witch doctor outfits, or slinky dresses, or mystic tattoos and a live snake — you’ve got to have something to put on the cover), and I don’t think the iron thing logically succeeds there.

          • Protagoras says:

            For the goal of keeping wizards in robes, I note that GURPS mages generally wear little armor. This is because they generally treat strength as a dump stat, and GURPS punishes you for trying to wear heavy armor when you aren’t very strong. And both having wizards be low strength and having that be a reason for them not to wear heavy armor does seem vaguely logical and in keeping with tradition.

          • Protagoras says:

            GURPS doesn’t have charisma as a stat, but in fact GURPS mages tend to treat every stat except IQ as dump stats. If they have a particularly bloodthirsty gamemaster, they may try to have a reasonable health, but they’re almost always both clumsy and weak, and usually not very healthy (GURPS has four stats).

          • theredsheep says:

            Sorry, after posting that I realized I could simply look it up, did, then deleted to avoid clutter. So now I’m re-cluttering to avoid confusion. Yay.

          • Plumber says:

            @Le Maistre Chat

            “…The Broken Sword…”

            Poul Anderson’s The Broken Sword was an AWESOME read!

            Delighted to know of someone else who’s read it!

            Anderson’s Three Hearts and Three Lions also (briefly) features the Elves, and there’s distinctly Elvish (or Faerie including Changelings!) native aliens in his Queen of Air and Darkness, plus in another short story he has Earthling Elves (that are vulnerable to iron) fend off an alien takeover (sorry that I don’t remember the title of it).

            I vastly prefer Anderson’s Elves to Tolkien’s.

          • Le Maistre Chat says:


            Poul Anderson’s The Broken Sword was an AWESOME read!

            Yeah, it was. Anderson wrote maybe half a dozen fantasy novels, and The Broken Sword, along with The Merman’s Children, stand out from the rest for treating the supernatural as part of our Earth rather than a parallel world.
            Besides those two and THaTL, I’ve also read Operation Chaos, about the romance of a werewolf and a witch after 1930s scientists discover how to de-antimagic iron, and A Midsummer’s Tempest, where the conceit is a parallel world where Shakespeare’s plays are accurate historical records and Cavaliers have to save the fay, who are Catholic, from the Puritans and their Industrial Revolution (no, really).


            But the intent of restrictions on armor for arcane magic users is clearly to keep wizards in robes and pointy hats (or witch doctor outfits, or slinky dresses, or mystic tattoos and a live snake — you’ve got to have something to put on the cover), and I don’t think the iron thing logically succeeds there.

            Well no, it doesn’t. While it makes sense that only the warrior class would be proficient in going around armored (remember David when King Saul offered his own armor to fight Goliath in?), any adventurer should be able to pick it up with skill points or whatever. It’s not at all traditional, or good cover art, for magicians, but that ought to be solved with some mechanical trade-off between armor and nudity/a robe rather than an illogical flat Nope.

          • @Plumber:

            My SCA persona was inspired by Carahue of Mauretania, first encountered in Three Hearts and Three Lions (Poul, of course, borrowed him from “Ogier le Danois”).

            Poul’s own persona was Bela of Eastmarch.

          • Le Maistre Chat says:


            (Poul, of course, borrowed him from “Ogier le Danois”).

            … this source is annoyingly hard to track down in English, as it turns out. I eventually resorted to interlibrary loan to find an English translation of La chevalerie Ogier, which turned out to be a c. 1220 chanson de geste about Ogier’s battles with Saracens and rebellion against Charlemagne, without any supernatural elements that I thought were the core of the story. Apparently the version that stars Morgan le Fay comes from a 14th century Roman d’Ogier, which was redacted into prose under the title you gave. Still can’t find either of those in English translation.

          • Deiseach says:

            A Midsummer’s Tempest, where the conceit is a parallel world where Shakespeare’s plays are accurate historical records and Cavaliers have to save the fay, who are Catholic, from the Puritans and their Industrial Revolution (no, really).

            Perhaps he knew the poem The Fairies Farewell by the Protestant Bishop Richard Corbet? It’s a humorous ballad about how the fairies aren’t seen anymore, not since the new religion and the new Protestant monarchs came in, so this ‘proves’ they were all Roman Catholics. Title of Rudyard Kipling’s Rewards and Fairies comes from it.


            Witness those rings and roundelayes
            Of theirs, which yet remaine;
            Were footed in queene Maries dayes
            On many a grassy playne.
            But since of late Elizabeth
            And later James came in;
            They never danc’d on any heath,
            As when the time hath bin.

            By which wee note the fairies
            Were of the old profession:
            Their songs were Ave Maries,
            Their dances were procession.
            But now, alas! they all are dead,
            Or gone beyond the seas,
            Or farther for religion fled,
            Or else they take their ease.

          • Plumber says:


            “My SCA persona was inspired by Carahue of Mauretania, first encountered in Three Hearts and Three Lions (Poul, of course, borrowed him from “Ogier le Danois”).

            Poul’s own persona was Bela of Eastmarch.”

            Yeah, well I saw Poul Anderson’s typewriter when it was at The Other Change of Hobbit bookstore buster!
            (Your SCA tales are amazingly cool, David)

        • dndnrsn says:

          To clarify – I managed to screw up introducing an analogy; artillery-only is a losing strategy IRL (and any wargame that does a good job of modelling reality). In some editions of D&D, the balance is such that the standard fight-sneak-heal-zap party gets abandoned. That’s bad.

      • Le Maistre Chat says:

        Balanced in the sense that the wizard contributes no more or less to the party than anyone else, or balanced in the sense that everyone contributes something, where you don’t have people saying “no no don’t play a rogue we need another wizard” or parties consisting of two wizards and two clerics.

        Where you don’t have the latter, minimally.
        I mean, let’s look at some of the abilities have D&D reserves for the spell system:

        Mage 1: non-fatal Area of Effect attack (Sleep) – pretty thematic & reasonable to be exclusive. Change the world by Charming an important Person – mind control being more powerful than the mundane equivalent is OK, but mundanely charming someone should exist. Light – just eliminates a hireling by consuming a different resource.
        Mage 3: A variety of “sensor platform” spells like Detect Invisible, Detect Evil, ESP, Locate Object. Can also become the best scout by going Invisible. This all seems fair.
        Mage 5: AoE that does 5 dice of damage in a radius of 4 grid squares (Fireball) – Heroes never got to be this destructive until 3rd Edition, at which point casters outstripped them far more than in Old School D&D. Fly for d6+5 rounds – why not let heroes reach a point where they can jump that long? Protection From Normal Missiles.
        Mage 7: Polymorph – perhaps the most classic effect in myths and fairy tales. More sensor/awareness effects (Wizard Eye). Large-scale camouflage ( Hallucinatory Terrain, Massmorph). Really the only effect from the 4th level spell list a hero should have access to is Fear.
        Mage 9: Too many effects to list. I’d like to call out Passwall and Stone Shape as things a powerful enough hero should be able to duplicate with sheer physical force. Also a special mention of Animate Dead for not being broken until Fighters were stripped of their henchmen – try playing a Wizard with the Necromancy specialization in 5E (where Animate Dead has been lowered to a 3rd level spell!) and watch the martial PCs get upset.

        • John Schilling says:

          mind control being more powerful than the mundane equivalent is OK, but mundanely charming someone should exist.

          That’s what Bluff and Diplomacy are for, and something that 3.5e at least tries to get right. But they needed to put more thought and more playtesting into that, because both of them as written are about right at low levels but broken for high-level characters.

          • Le Maistre Chat says:

            Yeah, exactly. I ran 3.5 from Level 1 to 22, and the Bard’s player and I had to hash out an agreement on what the Bard’s Diplomancy bonus would be at higher levels.
            Stupid broken thing that shouldn’t have even required playtesting to notice, though: Diplomacy was basically a skill for two classes that were also spellcasters, because it keyed off Charisma (the single attribute for Sorcerers and Bards).

          • John Schilling says:

            Aggravated by the tradition of every other character class using CHA as their dump stat. Another thing to beat into Gygax et al’s head if I ever get that time machine, is that Clerics would be much more plausible as healers/social engineers rather than healers/fighters. Make CHA a secondary stat for them, give them social-engineering abilities second only to Bards, and include decent morale rules so that you need to worry about social engineering even in combat(*).

            Diplomacy needs to be house-ruled in any campaign that’s going to make extensive use of it; that’s unfortunate but there are some good takes out there on how to do that. I can live with it.

            * And factor CHA into that, while you are at it. Mooks should run away from a 5th-level CHA-15 fighter who they simply will not fear at 10th level but CHA-5.

          • johan_larson says:

            Putting more emphasis on retainers and followers (and having CHA strongly affect their loyalty and numbers) would go a long way toward making CHA matter. That part of D&D seemed to get dropped over the years.

        • Nornagest says:

          Charm is one of those effects that players treat as more powerful than it actually is. Rules as written, it makes the target treat you as a friend; that’s it. They won’t attack their allies for you, they won’t kill themselves for you, they probably won’t give you money (though they may loan you money). They might not even stop attacking you if they already are, depending on the circumstances. Is that world-changing? Well, how much could you expect to change the world if you could wave your hands and make Vladimir Putin your drinking buddy for 1-20 hours (or until he makes his saving throw)?

          There are higher-level spells that rise to the level of how charm is treated by players (the dominate and geas lines are most obvious, but a well-timed and well-worded suggestion can work almost as well), but even they tend to be more limited.

    • John Schilling says:

      New D&D thread: how would you balance heroes with spellcasters?

      I’m not sure you can, at least in a way that will be satisfactory at 1st level and 20th level. D&D through 3.5e was pretty good at mid (roughly 3rd through 6th) levels and not intolerable across the 1st-10th level range.

      If you are using the D&D “limited spell slots per day” approach to achieve balance, you pretty much need to have and be willing to use intelligent adversaries who will TPK any party that tries to take an eight-hour break mid-dungeon after every significant encounter.

      An alternate approach is to make magic cumbersome, with powerful effects requiring significant ritual preparation and magic mostly used for non-combat utility purposes. That’s problematic for D&D for two reasons. In combat, D&D wizards are almost completely useless without their spells, particularly at low levels. Out of combat, D&D wizards are almost completely useless with their spells because the required effects usually cannot be predicted a day in advance. But if I were designing a system from scratch, an adventurer-wizard should be about as capable in combat as an adventurer-thief or an adventurer-cleric. Wizard magic should be broadly flexible – cast any spell you know at any time – but with e.g. mana and casting-time limitations making it not the default in combat. Note that Gandalf’s go-to for dealing with Orcs is a sword, and if he is presumably not quite the swordsman Aragorn or Boromir is, neither is he a d4 commoner with a stick. And gameplay should encourage a more even mix of spontaneous-combat, prepared-combat, and non-combat encounters.

      I’m not terribly fond of the idea of turning martial heroes into Son Goku just so they can keep up with 20th-level magic-users.

      And mostly, I think we’re discussing all the good reasons pre-D&D fantasy writers had for making magic an NPC-only thing. If you have magic in your campaign, you want it to at least occasionally be world-changingly powerful, not just a way to duplicate the effect of a heroically-wielded sword or bow with different special effects. But if you give wizards in general the ability to do world-changingly powerful things, most PCs are going to whine if they can’t do them right now, particularly in combat. And if you give bad-guy NPC wizards the ability to do world-changingly powerful things on demand in combat, they are effectively invincible in combat against non-wizard heroes. If magic is mostly an NPC thing, this is much easier to deal with.

      • idontknow131647093 says:

        I mean. ARPGs like Diablo and MMOs solved the problem pretty easily…

        • Civilis says:

          Most computer RPGs solved the problem by making everyone effectively a wizard: in many cases, every character has a pool of magic energy (generically, Magic Points or MPs) and they spend MPs to use special abilities suited to their class fluff. There’s usually some difference between classes: fighter-types generally have more HP and less MP and hit harder with their normal attacks, for example, but all those special MP attacks may as well be spells.

          There is a class balance feature from MMO-type RPGs that might solve the problem, if you can get over the hurdles that make it much more suitable to computer RPGs than tabletop, and that is the MMO balance ‘holy trinity’ concept, where classes are divided in combat role between defense, offense, and support (or Tank, DPS, and Healer). Fighters are naturals for the defense / tank role, with the job of keeping the enemy’s attention and taking the brunt of the hits so the squishier DPS classes can do damage. Usually this involves a mechanic for determining who the target must be (usually called ‘aggro’) that breaks immersion so that despite the squishy classes doing the damage (or the healer keeping everyone alive, for that matter) the enemy is required to attack the big armored guy.

          The plus side to this is that it makes classes in the Tank role mandatory for party survival. The downside is that there are a lot of issues with implementing this in a tabletop RPG.

          The first one is that this is not something that can be easily turned around on the players. Players aren’t going to like being told that they can’t attack the relatively squishy wizard doing damage to them and must attack the heavily armored but low damage enemy fighter. You can bypass this if the aggro rules don’t apply to players, but it then unbalances fights against NPCs built using player rules. It’s also very hard to justify this as anything other than a game mechanic.

          Second, it’s very boring to be the defender in a tabletop game; in a computer game, defense is a matter of active positioning and timing skills to keep the enemy attention, which works to keep the player engaged in the battle. I haven’t seen a good tabletop RPG combat system where defense was an active action.

          Third, this makes the rules even more complicated and slows down combat because you have one more thing to keep track (that computer RPGS make keeping track of this a trivial background task is why it works for them).

          You can sort of see a primitive ‘aggro’ mechanic in the attacks of opportunity rules in some of the later D&D versions, but it requires all those complicated miniature rules and movement rules and zone of control rules that bog down combat in those versions.

          • AlphaGamma says:

            4e D&D brought in an additional aggro mechanic in the form of ”marking”, which is something that defender (ie tank-type) PC classes, and monsters that fulfill a similar role, can do to opponents- for instance, Fighters can do it to everyone they attack.

            A creature (whether PC or monster- 4e went back on the 3e decision to have monsters and PCs be built on the same rules) can only be marked by one opponent at a time. Once marked, it takes a penalty on its attack rolls against anything other than the opponent who marks it. Most things that can mark opponents also have some form of reaction power to punish opponents that break their mark by attacking somebody else.

            (4e is the most miniatures-dependent of D&D combat systems, but I can certainly imagine a rule like marking working in a system that doesn’t use minis)

          • Civilis says:

            AlphaGamma, I didn’t play enough 4e to remember that there was such a mechanic, but it makes sense given 4e pulled a lot from MMO games, including the party role systems (and at the expense of making combat much more complicated).

            As a GM, I can fully understand a fighter player saying “I try to make myself big and intimidating and get the goblins attention so they go after me and not the wizard” and having them mark a target (especially if you add a STR or CHA check as a mechanic). I can even slightly work the mechanic from the NPC side, and tell the fighter “the enemy knight is up in your face; your experience tells you if you try to attack someone else, he’ll use the opportunity to strike you while you’re vulnerable.”

            The problem is that there’s too many mechanically driven scenarios besides the basic ones that break immersion and/or balance. Can I mark / aggro at a distance? If not, I’m going to end up facing off against any enemy tank’s mark, while the combat is decided by a duel between spellcasters. If I can, how do I justify it in universe, especially the ‘inflict damage if they attack someone else’ part? How does it deal with enemies that outnumber the tank? If I can only mark one of the dozens of goblins, I’m useless as a tank. How does it deal with big enemies? Sure, I can mark that dragon, so I get an attack bonus and perhaps a little extra damage, but if the GM is playing the battle straight, the Wizard I’m protecting is still dead. What happens with two tanks on the same side… which one gets the mark? (It honestly makes sense that if someone has two fighters concentrating on them, which is what a non-magic mark has to amount to, both of the attackers are going to have an easy time getting hits in).

          • Randy M says:

            Can I mark / aggro at a distance?

            Usually only Paladins. I believe most fighter marking abilities were limited to melee attacks. But you could usually charge in and mark.

            How does it deal with enemies that outnumber the tank?

            The Warden (Druidic like defender) can mark multiple adjacent creatures. The fighter may have a few limited use abilities to mark multiple, otherwise they need to pick the biggest threat to mark–but they still get opportunity attacks to try to control the others. Although the rules on out of turn actions were a bit complicated; I think you got up to 1 triggered action per round but opportunity attacks were once per any creature turn.

            If I can only mark one of the dozens of goblins, I’m useless as a tank.

            If you are against a dozen goblins, and they aren’t all goblin heroes or something, then the wizard you are worried about protecting laughs and drops a fireball or flame pillar or something on them, perhaps after burning a shield spell on some lucky goblin javelin.

            How does it deal with big enemies?

            The mark also has an effect of giving a -2 penalty to all attacks that don’t include the marker, and fighters have the ability to stop movement with their attacks of opportunity–yes, this is getting complicated to keep track of. Complication and depth are not exactly the same but the former often accompanies the latter. If you are fighting something that can shove you aside, then fly over you and chomp at the wizard, either you are high enough level that the wizard can survive a hit and be healed by the Warlord while you get back into position, or you guys really took a wrong turn back at the treasure horde.

            What happens with two tanks on the same side… which one gets the mark?

            The newer one supersedes, which may be a bit gamey, but think of it as there’s only so much face for you to get into.
            Flanking an enemy will still give you both a bonus to hit, and you both get opportunity attacks if it moves. What the mark does is penalize the target for making attacks that don’t include you, and usually give you an extra attack in that case.

          • Civilis says:

            Most of these examples don’t seem to be describing someone skilled at combat with a sword but instead describing some power that a class has. I don’t have a problem with giving all classes powers, but if you’re going that route you don’t need to balance the fighter class by assigning it to a specific role in combat.

            All these restrictions on who can be marked and how the mark works just make the fighter more complicated and more constrained when compared to the spellcaster, which is the exact opposite of what we want. If we’re going to make the fighter a co-equal necessity with the spellcaster to the success of the party, the fighter needs to contribute as much to the party. The ability to lock down a single melee enemy of comparable power isn’t an equal contribution. Since the fighter isn’t duplicating the spellcaster’s damage output, we need them to offer something unique; in the case of aggro, we offer them the ability to control the flow of battle.

          • Randy M says:

            Most of these examples don’t seem to be describing someone skilled at combat with a sword but instead describing some power that a class has.

            I can phrase it more flavorfully if that’s what you are looking for.

            Can I mark / aggro at a distance?

            No, you skilled in hand-to-tentacle combat as a fighter, if you want to keep someone pinned down at range, become a ranger.

            If I can only mark one of the dozens of goblins, I’m useless as a tank…How does it deal with big enemies?

            Your role as a fighter is to lock down the baddest single foe on the other side and prevent them from messing up your allies. Get up in their face and focus on them, and you can easily accomplish it, forcing them to chose between opening themselves up to your devastating and often crippling counter attacks, or swinging ineffectually at your armor. Those other, weaker goblins? If your allies can’t handle them, then you need to use the terrain to your advantage and find bottlenecks.

            What happens with two tanks on the same side… which one gets the mark?

            Choose your tactics such that one is holding them down and the other delivering damage, or split up, etc. The Paladin is too honorable to challenge a foe to single combat while it is engaged with the fighter, and the two fighters can’t both effectively draw it’s attention at the same time.

            If we’re going to make the fighter a co-equal necessity with the spellcaster to the success of the party, the fighter needs to contribute as much to the party. The ability to lock down a single melee enemy of comparable power isn’t an equal contribution. Since the fighter isn’t duplicating the spellcaster’s damage output, we need them to offer something unique; in the case of aggro, we offer them the ability to control the flow of battle.

            This is not a complaint made by someone with experience with 4E. The fighter is a bad ass who more than holds his own. He doesn’t need to surpass the caster’s damage (though, depending on the caster and the choices of the enemy, he might) since he mitigated a lot of damage by forcing the powerful enemy to waste its attacks against him.
            How often are all your foes equally threatening? Maybe you are fighting a pack of wolves or a scout patrol or so forth. But often, and most climatically, you will face down a powerful boss and his minions, or a single powerful foe. The fighter is also interesting to play because he must be aware of the terrain, how to position himself and the enemy (via pushes and shoves and feints), which is the bigger threat, and what tactics that enemy is likely to take. Then he presents the enemy with a lose-lose choice.
            Whether this is as interesting or as powerful as what the caster can do depends largely on what spells you are offering to the caster, of course. Defining the fighter as the guy who doesn’t get to use magic means its relative complexity and effectiveness is still tied to spell casting system.

          • idontknow131647093 says:

            There is the traditional Tank/DPS/Support solution, there is also the ability to allow for kiting of enemies, with an emphasis on rolling D20 to see whether you are caught. In addition, higher damage characters (mages, archers) should basically be 1-shottable by bosses including via random AOE attacks.

        • dndnrsn says:

          ARPGs and then increasingly videogame RPGs in general dealt with it by handing out cooldown powers, whether the nature of the powers is justified in-game (Mass Effect’s biotic abilities differing by class) or not (more common) 4th ed gave Vancian powers to other classes, but 4th ed was relatively unpopular (a lot of people went over to Pathfinder, and the degree to which people keep using old editions or buy new versions of them indicates their opinion of the current edition), and Wizards excluded most of what made 4th distinctive from 5th. People seem to like this in videogames but dislike it in tabletop games; doing nonmagic stuff Vancianly (new word) seems kind of uncommon.

    • Nabil ad Dajjal says:

      The problem with making every fighter Hercules is that you’re now no longer playing a Sword & Sorcery game where clever and brave heroes delve dungeons and fight dragons but an Anime game where heroes throw mountains at giant monsters. That sounds like a fun concept for a game but it’s still an entirely different game and probably calls for an entirely different rules set.

      The problem with D&D is that magic was allowed to grow wildly out of control over the editions. Game balance is much better now than it was in 3.X but magic is still a headache from a world-building and adventure-writing perspective. It’s very hard to create a consistent setting or believable scenarios when most non-combat problems in the world can be fixed by a motivated mage in under six seconds.

      At this point, I would just relegate all “utility” spells and cantrips to rituals only and leave Vancian casting to combat spells. A mage can throw little bolts of fire all day if he wants to, plus a few flashier attacks a few times before he burns himself out, but if he wants to pick a lock or breathe underwater he is going to have to at minimum spend ten minutes sitting there chanting first and possibly buy some expensive material components. It was one of 4e’s better ideas which was unfortunately tainted by association with the other more-objectionable changes.

      • John Schilling says:

        A mage can throw little bolts of fire all day if he wants to,

        Why is this preferable to “a mage can swing a sword if he wants to and is at least as good at it as a thief”? The latter doesn’t require a cumbersome split between combat and non-combat systems, is better for plausible world-building, and is more consistent with the source material (e.g. Gandalf and the Grey Mouser).

        • Nabil ad Dajjal says:

          Because players really don’t like it. They chose the Wizard class to cast spells, not to stab goblins with a dagger, and if the ratio of spell-casting to goblin-stabbing gets too low they start complaining about not having enough spell slots.

          If that 1d4 to 1d8 damage comes from a cantrip instead of a simple weapon it doesn’t make a big difference to the game’s math. But it does make a big difference to the player’s enjoyment. As long as it’s not also making an even bigger difference in the game’s implied setting I let that sort of thing slide.

          • Le Maistre Chat says:

            As long as it’s not also making an even bigger difference in the game’s implied setting I let that sort of thing slide.

            I wouldn’t. Next time I run 5E, I’m going to remove cantrips, or just about. A magician doing a sword thrust’s worth of fire damage every 6 seconds all day every day doesn’t fit my source material (folklore) and has nasty setting implications: why did some governments make being a witch a crime in the pre-Christian era? Presumably so they could chain 1st level witches to steam engines to power an Industrial Revolution.

          • Jiro says:

            Presumably so they could chain 1st level witches to steam engines to power an Industrial Revolution.

            You could just as well argue that they could chain fighters to treadmills or wheels to do the same thing. That’s not a problem with mages as compared to fighters, that’s a problem with not having rules for long term endurance and applies to characters of all types.

          • bullseye says:

            If all it took to have an industrial revolution was a reliable source of fire, it wouldn’t have taken so long. Hunter-gatherers living in a forest can keep a fire going indefinitely if they need to.

          • Statismagician says:

            But if you can use witches instead of coal once you’ve got all the other ingredients, odd stuff can start to happen. See e.g. Eberron.

          • bullseye says:

            Indeed. Eberron requires easy low-level magic, and conventional D&D settings aren’t harmed by it.

            I suppose if you wanted a setting with steam engines but no witches powering the engines you’d have to make sure that witches can’t do that.

      • dndnrsn says:

        Yeah, I’m finding the difficulties more in worldbuilding than in the game itself. I’m running a retroclone, and while the MU is key – he is the one party member that they won’t do serious stuff without – he doesn’t replace the other party members. However, I keep finding myself adjusting spells on the basis that they have major worldbuilding implications. For example, the original game’s charm spell is wild: it doesn’t specify how long it lasts, regarding effect all it states is that the spell “will cause the charmed entity to come completely under the influence of the Magic-User” until it is dispelled. A world where this is the case is interesting – wizards are able to enslave people magically – but it’s not necessarily a world that the GM signed up for, or the players, and it could be a lot of work to put together if you take everything to its logical conclusion.

        Even when later editions cut down the power of some spells (charm and sleep get less overpowered) there’s still lots of spells that, if you take them to their logical conclusion, means you get something more like Eberron or Planescape instead of Faerun or whatever the current fluff calls the place where the current equivalent to the Oriental Adventures book is set. Which are cool, but acquired tastes.

    • Nornagest says:

      There are two issues that need to be addressed for casters in D&D. The first is a theme issue: there’s little flashy battle magic in pulp or Tolkien and very little in folklore, so D&D casters don’t end up playing much like their archetypes. Barbecuing someone with a fireball doesn’t say “wise” or “mysterious”: in the source material, wrecking face is the sword guy’s job.

      The second is a gameplay issue — actually several different ones depending on edition. The old-school D&D model for casters is “do anything, if you predicted what you’ll need to do right”, which is fairly well-balanced against non-caster classes in ordinary play — a mid-level wizard going nova on prepared ground can be a challenging boss, but PC casters are much more limited if the adventure is properly paced and not too predictable, and they face a power/versatility tradeoff. The downside is that playing a caster isn’t that fun half the time — you can shut down a combat encounter with sleep or trivialize a skill challenge with knock, but if you prepped the wrong spells, or the encounter isn’t worth burning a spell slot on, you’ll end up sitting in the back row playing with your phone and plinking lethargically with a sling when your turn comes up.

      Thanks to easier magic item creation, that became “do anything, with prep time” in D&D 3+. This solved the empty rounds problem, but at the cost of caster supremacy, and it made the theme problem worse; it’s a non-starter. 4E and to a lesser extent 5E tried to tone it down by giving casters combat powers on the same scale as everyone else and relegating utility powers to rituals, but that has its own problems.

      Both of these problems can be solved, first by changing the library of caster powers and secondly by changing how they play out over time. Keep or even expand on spells like sleep and polymorph, but have them require multiple rounds of concentration first, so trivializing an encounter is possible but requires surprise or good tactical planning. Remove most direct damage, and most spells like knock that infringe on skillmonkey territory. For round-by-round play, either leave the option of Gandalf-level swordplay open or put more emphasis on tricks and lore: Vance’s mages lean very heavily on magical items, and Howard’s often rely on drugs and poisons. If your wizard wants to trample a city under the hooves of hellborn steeds the size of skyscrapers, as in “The Dark Eidolon”, don’t say no, but recall that in that story it took a lifetime of effort and a pact with, basically, Satan.

      • Le Maistre Chat says:

        There are two issues that need to be addressed for casters in D&D. The first is a theme issue: there’s little flashy battle magic in pulp or Tolkien and very little in folklore, so D&D casters don’t end up playing much like their archetypes.

        Gandalf uses battle magic once per day and a sword, so I’m not eager to remove the combat spells he used, nor to give casters free magic every six seconds when I could give them swords instead.

        The old-school D&D model for casters is “do anything, if you predicted what you’ll need to do right”, which is fairly well-balanced against non-caster classes in ordinary play — a mid-level wizard going nova on prepared ground can be a challenging boss, but PC casters are much more limited if the adventure is properly paced and not too predictable, and they face a power/versatility tradeoff. The downside is that playing a caster isn’t that fun half the time — you can shut down a combat encounter with sleep or trivialize a skill challenge with knock, but if you prepped the wrong spells, or the encounter isn’t worth burning a spell slot on, you’ll end up sitting in the back row playing with your phone and plinking lethargically with a sling when your turn comes up.

        This part is tricky. As soon as you give casters something fun and magical to do every single round, balance goes out the window.
        Really, a wizard should be a support character to the heroes. We’re talking about the fundamental building blocks of pre-1974 storytelling here. Thumb through Joseph Campbell’s The Hero with a Thousand Faces and you find “Supernatural Aid” but you don’t find “Wizard Protagonist.” The trick is how do you keep that fun?

        Thanks to easier magic item creation, that became “do anything, with prep time” in D&D 3+. This solved the empty rounds problem, but at the cost of caster supremacy, and it made the theme problem worse; it’s a non-starter.

        Yes, it’s awful for running anything but the implied setting of 3.X

        Keep or even expand on spells like sleep and polymorph, but have them require multiple rounds of concentration first, so trivializing an encounter is possible but requires surprise or good tactical planning. Remove most direct damage, and most spells like knock that infringe on skillmonkey territory. For round-by-round play, either leave the option of Gandalf-level swordplay open or put more emphasis on tricks and lore: Vance’s mages lean very heavily on magical items, and Howard’s often rely on drugs and poisons.

        Bah, I don’t want to remove knock either, since it was one of Gandalf’s spells. Yes to the swordplay, though. Note that Howard’s drugs and poisons are basically spell components: when Conan teams up with the thief Taurus in “The Tower of the Elephant”, he casts Sleep on lions and talks about how acquiring enough lotus powder to cast it was as much an adventure as the one they’re on now.

        • Nornagest says:

          Gandalf uses battle magic once per day and a sword, so I’m not eager to remove the combat spells he used…

          I can only think of one straightforward combat spell he uses: the flash of light that takes out some goblins halfway through The Hobbit, which you could gloss as a lightning bolt but also a bunch of other stuff. Most of the magic he uses is more on the utility side: he analyzes the One Ring, calls the eagles (twice), attempts but is unable to open the door to Moria, creates light once he’s inside, breaks the bridge while he’s fighting the Balrog, breaks the spell on Theoden, and makes a lot of fireworks for a party. A few more are borderline: he sets the surrounding woods on fire when fighting the wolves in The Hobbit, and later when fighting the Ringwraiths semi-offscreen in Fellowship, but that’s still not quite like what D&D wizards do. And a lot of the more dramatic stuff is arguably him using Narya, the Ring of Fire and the second- or third-most powerful artifact available in the setting at that point, not using his own power.

          Note that Howard’s drugs and poisons are basically spell components: when Conan teams up with the thief Taurus in “The Tower of the Elephant”, he casts Sleep on lions and talks about how acquiring enough lotus powder to cast it was as much an adventure as the one they’re on now.

          You could gloss them that way, sure, but you don’t have to. And from a theme perspective I think it makes more sense if you don’t: a wizard should be steeped in lore and full of obscure but not quite magical tricks. The wizard archetype is a keeper of secrets first, only secondarily someone who can break the laws of physics with their brain. And you get better gameplay out of it as a bonus.

          • Le Maistre Chat says:

            Yeah, I don’t disagree that most of Gandalf’s magic was utility. In The Hobbit, he tricks trolls with Ventriloquism (arguably not magic, but D&D made it a 1st level spell), Pyrotechnics to make a distraction when Bilbo and the Dwarves are Goblin prisoners, Light on multiple occasions. He also lights pine cones on fire with his staff as a weak weapon.
            LotR: attacks with Lighting Bolt, Phantasmal Forces (?) in the river washing the Nazgul away. Utility includes eagle summoning (though an optimizing player would totally make that a combat spell) making a fire in a snow storm on the way to Moria, Light again, Knock, Hold Portal, ESP with Galadriel & Elrond, Dispel Magic (?) on Theoden.
            You definitely need a short, utility-focused spell list to play him.

            You could gloss them that way, sure, but you don’t have to. And from a theme perspective I think it makes more sense if you don’t: a wizard should be steeped in lore and full of obscure but not quite magical tricks. The wizard archetype is a keeper of secrets first, only secondarily someone who can break the laws of physics with their brain. And you get better gameplay out of it as a bonus.

            Yeah, this is true. A wizard is a wise man (“wise woman” is proper Early Modern English for a female wizard) keeping secrets, steeped in lore not everyone has.
            So rather than trying to balance a PCs around “how many times a day can I make a cloud of sleep gas or a hand grenade with sheer mental power”, I guess you could balance them around scarcity of material components and how long the rituals to turn them into useful form take.

          • Nornagest says:

            It’s also worth mentioning that divination is incredibly central to the folkloric wizard archetype, probably more so than anything else the folkloric wizard can do: think Merlin’s vision of the red and white dragons, or the three witches in Macbeth. But it tends to get sidelined in games. In CRPGs there’s a good reason for this, since it’s near-impossible to give a vision or a prophecy satisfying gameplay effects outside a cutscene, but there’s no good reason not to use it on tabletop. The D&D spell lists have some of it, but it’s either unreliable or underwhelming, and by old-school rules could break your character pretty easily.

          • Le Maistre Chat says:

            @Nornagest: Yeah, it’s central enough that there are even characters in, e.g. Greek mythology, who are seers without otherwise being wizards.

          • Nabil ad Dajjal says:

            The problem with divination is that players at the table tend to take prophecy as a challenge.

            As an illustrative anecdote, once I ran a brief gimmick campaign. Each player played two versions of the same character. One was a level one adventuer ripped from his bed to an alien world by a race of confused seers trying to narrow down the number of candidates to fulfill an apocalyptic prophecy twenty years from now. The other was a high level version of that same character twenty years later. Some sessions were “flashbacks” which revealed key information or allowed the players to establish what kind of relationships they had with characters who would appear “later on.” Others were in the present day trying to avert the prophecy.

            In the beginning I made the players promise not to deliberately cause time paradoxes. The only way the game would work is if we cooperated on that point.

            The very first time they met a character they had already run into as an adult in the future? Within minutes they were plotting to kill her in the past.

            It’s a very difficult temptation for a lot of people to resist.

          • The Nybbler says:

            The problem with divination is that players at the table tend to take prophecy as a challenge.

            This is a typical problem for characters in traditional mythology and folklore as well, to the point where NOT fighting the prophecy is usually a subversion of the trope. Traditionally, fighting the prophecy turns out badly, of course.

          • Nornagest says:

            Yeah, it’s central enough that there are even characters in, e.g. Greek mythology, who are seers without otherwise being wizards.

            The “-mancer” suffix that tends to pop up around magic — necromancer, pyromancer, geomancer — is Greek, and refers directly to divination. A necromancer was originally someone who’d talk to the spirits of the dead, not someone who’d raise zombies. A pyromancer was someone who’d see the future by gazing into flames.

            (It’s also related to “mantis”, as in “praying mantis” — the insects were named for a supposed resemblance to soothsayers, but the link’s now almost completely obscure in English. And to “mania” — madness and prophecy are related in a lot of traditions.)

    • Tatterdemalion says:

      The root of this problem is the linear warrior/quadratic wizard phenomenon: roughly speaking, spellcasters gain power quadratically, because they get both more spells and better spells, whereas warriors gain it linearly, or not much better than that, so there’s no choice of constants that will balance them at all levels.

      Fortunately, there’s a simple solution: play at the crossover point. As Jonas says upthread, D&D 3.5 is a fun and tolerably-balanced game between about levels 2-3 and 6-8.

      Going above that runs into a whole bunch of other problems, anyhow – powers like flight, mind-control, esp, divination, etc make it much harder to run stories other than “battle the Big Bad”.

      • Nornagest says:

        “Linear warriors, quadratic wizards” misunderstands the problem. Wizards have gained more and better spells in pretty much the same way since the early days of D&D, but balance problems were only notable from 3.0 forward; AFAICT they became notable because of magic-item creation rules that let them do an end-run around the power/versatility tradeoff I mention above (creating scrolls and potions is difficult and basically up to DM fiat in 2E, but it’s trivial in 3E), and because of a proliferation of splats that let them get around Vancian limitations in other ways.

        • Tatterdemalion says:

          I don’t agree that this is a new issue – my recollection of 1st ed is that at 1st level wizards were basically dead weight, with a single magic missile per day that didn’t come close to balancing them against fighters,whereas by 20th they were massively superior.

          • dndnrsn says:

            1st level wizards can cast sleep. It’s poorly defined in 1st ed, but even by the strictest definition, you’re putting a lot of low-hd foes down. Allows a 1st or 2nd level party to take on a lot of situations that would be overwhelming without sleep.

            Let’s say a 2nd level wizard (memorized two this morning) plus the rest of the classic party lineup run into twenty orcs. By hd, their force is half as strong, however, the party can probably keep the wizard safe enough to get off one or two spells. Each casting is going to put down 4d4 enemies – average ten, tending towards the median since it’s four dice. Morale checks probably get involved.

            Magic missile is good once you’re fighting stuff that is too high hd to sleep. By then, the MU is throwing multiples, and it hits automatically. Against orcs or whatever, it blows, because it’s about the damage of an arrow and being an autohit doesn’t make up for being one-use.

            A first level wizard has the ability to make 10 guys just fall down unconscious once a day. Not hugely useful against many encounters with comparable numbers of similar-strength enemies, but once a day it allows the party to behave more recklessly than normal. Plus some of the spells are useful outside of combat, eg charm is weaker in 1st ed than the original version but it’s still pretty good.

    • cryptoshill says:

      The Tome of Battle is a great book for D&D. People call it “broken” because it is trivial to write a broken character. In fact – it is slightly more trivial to break any spellcaster, but they are in the same order-of-magnitude such that they can be in the same game when you’re using Tome of Battle.

    • Drew says:

      Start by breaking “caster” into a bunch of sub-classes, each with a central theme and very limited core spell list. Give out some priest-style domain abilities to compensate.

      Fictional wizards specialize. A fictional Necromancer knows the secrets of the dead, and has all kinds of undead minions. He might also be able to do other sorts of magic. But 50%+ of his on-screen casting should be dead thing related.

      Someone else might control the elements. I’d expect to see him throwing around all kinds of fireballs and lighting and that kind of thing. Enchanters should mostly enchant people.

      So, if I wanted to be lazy about the character re-design, let people memorize stuff that wasn’t on their spell list, but I’d increase casting times by an order of magnitude. So, the Necromancer technically can cast a Fireball, but since he hasn’t done much evocation since college, so it’s going to take 1 minute rather than 1 round.

      Then, if you’re going to have out-of-combat utility spells, particularly ones that negate a non-spell mechanic (“You invested 10 ranks in climb? Watch me make that irrelevant.”), I’d turn them into rituals and make them available to pretty much everyone.

      The fluff is that these aren’t spells, they’re people invoking secrets and pacts and gifts. Getting access to a ritual should be a major event.

    • Dack says:

      There is a strong argument to be made that a party of PCs aren’t supposed to be balanced against each other. They’re on the same team, fulfilling different roles. Nobody looks at a football team and asks why the kicker isn’t balanced against the quarterback, so why ask about the rogue vs the wizard?

      If you still think it’s a problem, you could just make them all play a gish though.

      That’s what I did running a Star Wars RPG. “How are you going to “balance” jedi vs non-jedi?” “You’re all playing jedi.”

      • Le Maistre Chat says:

        There is a strong argument to be made that a party of PCs aren’t supposed to be balanced against each other. They’re on the same team, fulfilling different roles.

        This is how it worked until 3.0 was released. Starting then, you had the problem that a Druid fulfilled the same role as TWO martial heroes while also being able to cast spells as needed.
        You could tell everyone to play a full caster, but caster supremacy gets extremely unsatisfying if you’re trying to emulate any culture’s mythology. Outside of saints’s legends, the supernatural person was always a support character to the protagonist.

        • Dack says:

          If we’re specifically talking about 3rd, yes, Druid is one of the most powerful classes, even without really trying hard. While you can put in a lot of effort to making one of the PHB non-caster classes shine, you’re better off as a Warblade if you want to smash face without spells as a single class character. Of course nothing is stopping you from multiclassing, or smashing face as a Cleric or whatever for that matter except your own self-inflicted fluff. Ignoring the fluff is why most people are bad at being Druids, IMO. My belief is that they are powered-up because they are supposed to be played how they are described in the book. (A loner nature fanatic, dedicated to neutrality, but also beholden to serve and obey the orders of their secret druidic society.) They don’t really make sense in most any adventuring party, let alone murderhoboing. They really should have had some sort of code of conduct mechanic a la Paladin.

          This is all besides the point. Even when two players use the exact same class, there are going to be disparities in effectiveness. Usually it comes down to system mastery. But this isn’t unfair because it’s not a PvP game. Well, sometimes it is, but they’re doing it wrong. Even if it is a PvP game, is it unfair when a sportsball player who practices a lot is better at the game of sportsball than someone who is new to it?

  29. Nick says:

    I have a dumb question, SSC.

    I was on my way home from the holidays last Sunday at the Greyhound station in Cleveland. While I was sitting waiting for my bus, a stranger sat down a few seats from me and said something like, “Hey big guy, you got change for a 20?” I said no. He asked for change for a 10. I said I only had $8. He asked if I could trade him anyway. I said no; he insisted. At that point, as I was telling him he ought to ask the customer service or cafe for change, one of the bus operators came by and instructed him to leave, and he did.

    It struck me as very sketchy, but what was the goal? My first thought, since he was asking for change for a 20, is counterfeiting, but I do not live in a Neal Stephenson novel. My second thought was that he had some other way of cheating me, but I can’t think of how, and it doesn’t really sound worth it for $8. My third thought was that he really wanted the change, but the only reason I’ve ever had to prefer small denominations to large is soda machines and the like, which back at college only accepted $5 or less. But to my knowledge there weren’t any of those in the building, and he could have bought a soda at the cafe anyway. So what was going on?

    • Nornagest says:

      I’ve run into small-denomination counterfeit bills before (in Oakland, California, where they were discovered — or “discovered” — by the same convenience store that gave them out as change). But it’s not very likely.

      If the guy was planning any kind of crime, I think he’d most likely just have snatched the money and run (risky, yes, but I’ve gotten my windows smashed for a five-year-old iPod that’d be worth maybe five bucks on the black market). But it’s also possible that he just needed change for a twenty.

    • Deiseach says:

      All I can think of is if you took your wallet out to give him the change, that’s when the snatch-and-grab would happen. If he just wanted change, he could make a small purchase at the café or wherever. I wouldn’t have thought counterfeiting but then again the kind of small-time thievery at a bus station probably means swapping a fake twenty for real fifteen plus bucks is worth it at that scale. The only other thing I can think of is the scamming for “I need to buy a bus ticket” and maybe the idea is that if you’re a soft enough touch to be willing to give change, you’re a soft enough touch to fall for the sob story and let him off with not taking the (fake) twenty and giving him the change once he launches into “Actually man, I really really need to buy a ticket home to see my dying mother, could you spare the change this once?” story. Asking for change breaks the ice and if you’re willing to listen that far, you’re less likely to brush him off than if he launched straight into the begging.

    • theredsheep says:

      In my experience, some people are just desperately eager to rip somebody off, to a degree that’s ludicrously far out of proportion to the effort, risk, or payoff. Most of the scams I’ve encountered probably didn’t pay better, on an hourly basis, than minimum wage. It’s sort of a self-image thing, I guess, where $5 earned after two hours of trying to hustle strangers conveys far more satisfaction than three times as much from an equivalent amount of time at McDonald’s. It’s sort of the converse, or inverse, or something–I never was much good at those Aristotelian opposites–of extreme paranoia about welfare or voting fraud. Some people can’t suppress their fears that someone is going to screw them in a really pointless and inefficient manner; others can’t resist the urge to be the ones doing the depressingly unimpressive deed itself.

      Or maybe these people are just too stupid to do cost-benefit analysis. One of those.

      • Jacobethan says:

        where $5 earned after two hours of trying to hustle strangers conveys far more satisfaction than three times as much from an equivalent amount of time at McDonald’s.

        Of course it does, how could it not? The one way, you’re your own boss and get to experience the delicious feeling of superiority to someone who outwardly regards you with contempt and pity, playing upon his gullibility, vanity, or guilt to extract his money from him. The other way, you get to act servile for eight hours in front of the same moron in an atmosphere that couldn’t be worse designed for a young guy with limited impulse control. Is that delta worth $5/hr to me? Yeah, plausibly, conceivably.

        • j1000000 says:

          God dammit now I’m gonna watch Fight Club again tonight

        • theredsheep says:

          My limited foodservice experience suggests that it’s really not that hard to get by in fast food with very little effort. And in lower-end retail in general. You’re competing in a labor market with a bunch of druggies and resentful high-schoolers, who frequently no call no show when they aren’t simply quitting without notice. Anyone who has enough pride to do good work in spite of the bad incentives will quit for better work, or else get promoted to manager where they’ll have too many responsibilities to take on the Sisyphean task of trying to foster a work ethic in people doing a job that only exists because it’s not currently cost-effective to automate it. Over time this adds up to a state of war between labor and management, where each is striving to get as much as possible out of the other while giving as little as possible in return, and assumes bad faith is at the bottom of every action. It’s not pleasant, but it’s far from difficult with the bar set so low, and it requires negligible initiative.

      • It’s sort of a self-image thing, I guess, where $5 earned after two hours of trying to hustle strangers conveys far more satisfaction than three times as much from an equivalent amount of time at McDonald’s.

        A friend of mine was both an economist and a serious gambler, a card counter at blackjack. I once commented to him that I felt better about money earned for something specific, such as giving a talk, than about my salary from a university, because I had a clearer intuition of the link between producing and getting paid.

        He told me he felt best about money won at cards, because he was, as it were, going out into the jungle in competition with other people and coming out the winner.

    • Paul Brinkley says:

      Wait, how do you know you’re not living in Anathem?

    • dick says:

      “He that has once done you a kindness will be more ready to do you another, than he whom you yourself have obliged.” – Ben Franklin. It’s possible the point was just to get your wallet out or to see where you keep it, but getting someone to do something trivial for you is also just a common opening line for various hustles. A good con artist doesn’t need to run a set line of patter, he can improvise and lead you down a merry path that ends with you $20 lighter and not really sure how it happened, but he can’t get anywhere without an exchange of confidence (hence the name). So a lot of hustlers start by asking for you to make change, give directions, etc, and if that goes well and you seem manipulable, it would move on to something else.

      If you want to see a rather extreme version of this, search for “derren brown russian scam”.

      • Doctor Mist says:

        If you want to see a rather extreme version of this, search for “derren brown russian scam”.

        Holy cow.

        What did that second guy think, the second time?

  30. theredsheep says:

    Since asking about Monty Hall worked so well, here’s something else I’ve always wondered, and haven’t managed to google: how do computers coordinate encryption? That is, how does the machine on the other end know exactly what operation to perform to reverse the encryption your machine put on it? I’m sending the friendly merchant my credit card info in a magical, peek-proof, tamper-proof box that cannot be opened without either the magic key or else the power of a wizard so mighty that he is unlikely to ever be born, or to work for people who want my credit card info. But how do I arrange for the friendly merchant to have the magic key so he can open the box, without sending it by the same route and compromising the whole system?

    • Nornagest says:

      how does the machine on the other end know exactly what operation to perform to reverse the encryption your machine put on it?

      Because it sent your machine enough information to make that possible.

      Oversimplifying like crazy here: certain mathematical operations let you create a pair of keys such that you can mangle data with one key but you need the second key to unmangle it. Once you have a pair like that, you can send the first key in the clear to anyone you want to communicate with, and they can use it to encrypt data, but once it’s encrypted only someone holding the second key can decrypt it.

      There are lots of different ways to create these pairs, but the general approach is to find an operation that’s easy to do in one direction but hard to reverse unless you have some extra information. The RSA cryptosystem (until recently the standard, and still widely used) relies on prime-number factorization: first you find a product of two large primes p and q, which is the modulus you’ll be working with. Then find a number e that’s coprime to the least common multiple of p-1, q-1, and compute d, e‘s multiplicative inverse relative to that modulus, which can be done efficiently if you know p and q (but not if you know only p * q alone). Send p * q and e to your peer, retain p * q and d, and you can encrypt any number by exponentiating it by e modulo p * q. Exponentiation is easy and finding modular roots is hard, so it can then only be decrypted by exponentiating it by d modulo p * q unless you want to do some very difficult math to it.

      Since data in a computer is just big numbers, that means you can encrypt any data, although in practice this is slow enough that you’ll generally be using the above scheme to encrypt only the key to a second, faster, but symmetrical (i.e. both sides use the same key) encryption algorithm, then using that one for whatever you actually wanted to send.

      (Source: am security engineer.)

      • theredsheep says:

        Are you referring to the product-of-two-huge-primes thing? Does it work out somehow if you only send them the product, or one of the factors, while you keep the other, or something similar?

        (my “math sense” is somewhat idiosyncratic; I understand some things readily but am hopelessly baffled by others)

        • Nornagest says:

          Added details to parent post while you were writing this, probably. The product-of-two-huge-primes thing is one common one-way function but not the only one; modern crypto libraries most commonly default to functions of elliptic curves.

        • Eric Rall says:

          It’s a bit hairier than that. There are several approaches, but they run on similar principles. Wikipedia has a decent technical explanation of the RSA algorithm.

          The short version is that the public key consists of the product of the two primes (N = PQ) and an exponent. The sender turns the message into a numbers (or a series of numbers, for longer messages), and each number is encrypted by raising it to the exponent mod N (i.e. divide by N, discard the integer part of the quotient, and return the remainder; e.g. 7 mod 3 = 1). The formula for getting the plaintext back from the cyphertext (which is a bit over my head) requires knowing the two primes, so either you need to know the primes already (the private key) or you need to factor N.

          • Nornagest says:

            You actually don’t need to know the primes to decrypt RSA; they’re usually thrown out after you create the keys. The keys consist of a modulus (the product of the two primes), an exponent coprime to that modulus, and that exponent’s multiplicative inverse relative to that modulus, which is calculated using p and q but doesn’t include them. The peer gets the modulus and the exponent; you keep the modulus and the inverse.

          • Tatterdemalion says:

            You actually don’t need to know the primes to decrypt RSA; they’re usually thrown out after you create the keys. The keys consist of a modulus (the product of the two primes), an exponent coprime to that modulus, and that exponent’s multiplicative inverse relative to that modulus, which is calculated using p and q but doesn’t include them. The peer gets the modulus and the exponent; you keep the modulus and the inverse.

            Almost – if the two primes are p,q, so that N = pq, and the encryption exponent is e, then the decryption exponent is d = e^-1 mod (p-1)(q-1), not e^-1 mod pq.

            This means that de = 1 mod (p-1)(q-1), and so by Euler’s Little Theorem m^de = m mod N

    • The Nybbler says:

      There are two basic kinds of encryption involved.

      1) Symmetric key encryption. This is the kind that has been known since antiquity. You have a secret key, the other party has the same secret key, and there are complementary decryption and encryption operators which, can convert between plaintext and ciphertext given that secret key.

      2) Public key encryption. This is a 20th century invention which depends on certain operations being easier to do than to reverse. Two examples are multiplying two very large prime numbers together (the reverse is factoring), and raising a number to a power in a finite field (the reverse is called a ‘discrete logarithm’). There are ways to use these to create a cryptographic algorithm with two keys, a “public” key and a “private” key. That which is encrypted with the public key may be decrypted wth the private key (but not the public key), and that which is encrypted with the private key may be decrypted with the public key. It is feasible to derive the public key from the private one, but not vice-versa. (note: for a given level of security, public key encryption is much more expensive than symmetric key)

      Add to this a few more concepts:

      * Hashing: This takes a message of arbitrary length and distills it down to a message of fixed length (the hash). Obviously this is a lossy operation and cannot be reversed. For cryptography, we use hashing functions which have a few important properties
      — It should be infeasible to generate a message with an arbitrary given hash.
      — It should be infeasible to generate two messages with the same hash.

      These properties let us say, with very high probability, that two messages with the same hash are in fact the same message.

      Hashing plus public key cryptography leads us to

      * Signing: Signing is a way to prove that a message has not been tampered with, and that its origin is what it is claimed to be. The way a message is signed is to take a hash of it, and then encrypt that hash with the private key. The message is then verified by the receiver by hashing the message, decrypting the signature with the public key, and checking that the results match. Since only the private key holder could have encrypted the hash, this demonstrates the source of the message.

      The last thing we need is certificates. A certificate is a message containing identifying information about the organization who owns it, information about how it can be used, and most importantly, a public key. It is also signed — either by the private key corresponding to its own public key, or by the private key corresponding to some other certificate’s public key.


      Step 0 (perhaps a bit fanciful): Back in the dawn of time, a bunch of corporate CIOs met in a back room in Silicon Valley somewhere. Some of these people were representatives of browser manufacturers, others were people who wanted to run the certificate racket — Certificate Authorities, or CAs. The CA people all brought with them certificates that were signed with their own private key (self-signed certificates). The browser manufacturers took these certificates and installed them in their own products.

      Step 1: Your merchant went to one of these CAs with a Certificate Request and some cash. The CA took this request and generated a certificate signed with one of the CAs own private keys.

      Step 2: Your browser contacted the merchant and requested the certificate. It checked that the certificate’s information was valid, and that the certificate was signed by an authority it recognized. (there could actually be “intermediate certificates”, but that’s a detail). This gave the browser a public key to use.

      Step 3: The browser can now do a key exchange. Key exchange requires the browser generate a key, encrypt it with the server’s public key, and send it to the server. The server can decrypt with the private key, and now both parties know a key and can use symmetric encryption.

      So now the browser knows who the server is (the server cannot authenticate the browser in this scheme, however), and can trust that no one can read the information.

      There’s at least one problem with this. Suppose “someone” has been recording this exchange. And they’re very patient. And a few years down the line, by hook or by crook, they obtain the server’s private key. They can now go back and decrypt everything encrypted using the server’s keys, including this recorded exchange with your credit card number and the exact type of porn you were ordering. Oops.

      There’s an alternative way of doing key exchange called Diffie-Hellman key exchange, which is based on the difficulty of the discrete logarithm problem. Each party picks a secret number. It takes that secret and raises it to the power of a generator of a finite field, and sends that to the other party. The secret key is then the generator raised to the power of _both_ secrets, which both parties can compute (because exponentiation is commutative) but no eavesdroppers can. The weakness of Diffie-Hellman is that an interceptor, a “man in the middle” who can control the network, can arrange to pretend to be the browser to the server and pretend to be the server to the browser. The way around this is that the number the server sends (the generator raised to the power of the server secret) is _signed_ by the server’s secret key. So this key exchange can neither be intercepted, nor can the key be derived by any party who obtains the server’s secret key in the future. This latter property is called “Perfect Forward Secrecy”, and if your threat model includes multinationals or state-level entities, you want it; it limits the damage enormously if your key is ever compromised.

      • theredsheep says:

        Holy cow, did you go the extra mile. A big thanks to you, and to everybody else who filled me in here.

    • toastengineer says:

      But how do I arrange for the friendly merchant to have the magic key so he can open the box, without sending it by the same route and compromising the whole system?

      I think the specific thing you want is the Diffie-Hellman key exchange. See second answer for non-mathematical explaination.

      But Diffie-Hellman only gives you secrecy (no-one can spy on us…), not authentication (…but how do I know YOU’RE not a spy pretending to be my friend?) which is provided by the asymetric key system described above.

    • TDB says:

      At first I thought you were asking how they know which protocol to use.

    • LesHapablap says:

      I recommend Simon Singh’s The Code Book (link text which goes through the whole history of cryptography.

      The real world analogy works like this: you have a secret message that you want to send to someone. So you put the message in a lockbox with your own padlock on it, to which only you have the key. They receive the lockbox, put their own additional padlock on it, to which they have the only key, and send the box back to you. Then you remove your padlock, and send the lockbox back to them. Now they remove their their own padlock, and done.

  31. Estera clare says:

    Tried to open SSC at school today during a study hall and it was classified as “Alt/New-Age Content” and blocked: I think someone was confused by the cactus person post.

    Previous blockings include Zoroastrianism for “non-traditional religions” and Christianity for “traditional religions,” so this was surprising, but for some reason this time SSC was the only site blocked. ¯\_(ツ)_/¯

    • toastengineer says:

      REASON: Insufficiently offensive

    • Le Maistre Chat says:

      Previous blockings include Zoroastrianism for “non-traditional religions” and Christianity for “traditional religions,”

      Wow, your school is bigoted.

      • Nick says:

        I wonder if they block traditional non-religions and non-traditional non-religions too.

      • theredsheep says:

        I’m not sure what’s non-traditional about Zoroastrianism, anyway. It’s like three thousand years old. Perhaps they’re referring to the part where a lot of their tradition got deliberately destroyed by a previous generation of censors (“May offend long-dead Iranian Muslim zealots”).

    • theredsheep says:

      Imagine how excited the censor would get if you tried to read that post about all the stuff people searched to get here (“why is my sister so hot?” X 25 … ).

    • Deiseach says:

      If they blocked a site for “no religion at all”, you’ve have the entire collection! 🙂

    • Dog says:

      I’m not familiar with the particular filtering system used by NYC, but my advice as a previous school sysadmin is to click on the “request a review” link and suggest a different category, and it will most likely get reclassified. Most of these systems are automatically categorizing sites and typically there aren’t a lot of people dedicated to validating review requests, they just assume it’s a teacher asking and make the change.

  32. TDB says:

    Some recent thread (survey comments?) brought up Scott’s old fish farming thought experiment (originally from his Non-Libertarian FAQ 2.0: Here’s my “purely voluntary” solution:

    Use common law or customary law. The cooperative fish farmers sue the holdouts for damaging their operation with their unfiltered pollution. (But wait, this is a purely voluntary system, how can they sue? ) They find a mutually acceptable arbitrator. (But wait, what if the parties to the dispute can’t agree to a mutually acceptable arbitrator?) The cooperative fish farmers tell their story, as above. The holdouts say, “In a purely voluntary society, of course we can pollute all we want! Don’t you believe in property rights?” The arbitrator decides in the cooperators’ favor and imposes a fine on the holdouts that incentivizes them to cooperate. (But wait, what if the arbitrator decides for the polluter?) The holdouts are better off joining the agreement and everyone lives happily ever after. (But wait, what if the holdouts refuse to pay the fine?) (But wait, couldn’t aggressors use a similar procedure to recreate a parasitic organization and impose crazy drug laws and everything else we hoped to escape by creating a purely voluntary society?) (But wait, haven’t we just created a monopoly?)

    I think I can answer all these objections, but that’s an awful lot of objections. Maybe it’s just too many objections to be credible? That’s why I actually advocate running experiments rather than a one size fits all reform of everything everywhere. (See Patri Friedman on startup cities.) So why do I find the PVS so fascinating? I just can’t help it.

    • Jiro says:

      I think I can answer all these objections, but that’s an awful lot of objections. Maybe it’s just too many objections to be credible?

      If you don’t know how a system works it will always be possible to make a lot of this kind of objection. So “too many objections to be credible” can’t be a real thing. It depends on how thoroughly each objection is refuted.

    • rahien.din says:

      Go discover Holmström’s theorem.

    • ing says:

      I think the holdouts can just say: “No, I’m not interested in you suing me, no arbitrator is acceptable because I’m not going to participate in your made-up legal process.”

      • Winja says:

        That situation looks a lot like the struggles with water rights in the old west.

        And when it came to that point, people ended up beaten or shot.

  33. Anonymous says:

    Is something wrong with comment subscriptions? I noticed the lack of notifications in my inbox.

  34. johan_larson says:

    No madcap mission this thread. Feel free to propose one of your own.

    • Unsaintly says:

      Your Mission, should you choose to accept it, is to fundamentally alter the current structure, use, culture or function of the internet via time travel. You may travel to any point in history after 1960, but may not bring back any modern technology or information beyond what is in your head. You will be provided with credentials and documentation allowing you to insert yourself at any stage of the internet’s development or early history, if needed.

      • cassander says:

        Go to whatever meeting where they decided to start URLs with and burn the building to the ground so we don’t spend the rest of time saddled with 3 letter but 9 syllable acronym before every website.

      • theredsheep says:

        It seems best to tell people to do IPv6 from the get-go. I can’t think of anything else wrong with the internet that isn’t caused by the people using it. Of course, I don’t know all that much about the technical side of the internet.

      • Mark Atwood says:

        What I would bring to 1960 in my head:

        GPL-2.0 and Apache-2.0 licensing

        An improved C and an improved Lua.

        Public key crypto, probably ECC. In fact, I would probably redesign it so that the ECC public key *is* the endpoint address.

        DNS is more flexible. I’d write a better programmable BIND, and make it easy for apps and protocols to assume it’s always present for discovery, auth, and small data sharing.

        REST becomes a thing from the beginning.

        These are all things I could do with what is in my head right now.

        • Nick says:

          How would you improve C and Lua?

          • Mark Atwood says:

            If I was cold re-implementing C in 1960, it would probably look like Rust without the type checking. (Because I don’t have enough CS in my head to implement it, and no computer in 1960 has enough memory to run a compiler that contains it.

            And my cold implementation of Lua would look syntactically like it, except it would be scripted, dynamic, and would have a straightforward semantic translation to Lisp.

          • SamChevre says:

            Use string encoding that stores the string length, or a pointer to the end of the string, as the first “character”.

          • Mark Atwood says:

            string length as first character

            aka “Pascal strings”. That’s basically how everyone was already doing it, until C got popular.

        • albatross11 says:

          From Wikipedia, IPv4 came out in 1981. At that point, public/academic/industrial crypto was barely a field at all, so it would have been impossible to just incorporate existing crypto into the standard–you’d have had to invent a lot of it–sometimes algorithms, often protocols and formats and all the surrounding engineering details that make the crypto algorithms actually work.

          a. Public key cryptography had just been invented–New Directions in Cryptography was only published in 1976. RSA was invented in 1978. Some GCHQ guys had apparently invented public key crypto a few years earlier in 1973, but they sure weren’t telling the guys designing the internet about it.

          b. DES (basically the first example of public, standardized crypto) had been standardized in 1977. (Basically everyone spent the next several years learning about symmetric crypto from it.)

          c. It was around 1990 before we had any hash function designs that look remotely modern. (Snefru and MD4)

          d. Elliptic curve crypto was only proposed in 1985.

          • Mark Atwood says:

            If you dropped me into 1960 with a time machine, I can explain RSA public key, EC public key, and DH key exchange with a pencil and 12 sheets of paper. The hardest slog part of it would be explaining what an Elliptical Curve is to someone of the era. I could put the whole thing in a 1/4-page ad in the New York Times, and then the whole world would know it, 25 years early. Or better yet, I would go explain it to Isaac Asimov and to Martin Gardner. Then it would appear in the next issue of Scientific American and the next issue of Astounding Science Fiction.

          • albatross11 says:

            Sure, but then you’re really starting the public crypto field 20 years early. And there’s a lot of engineering necessary to get from there to good built-in security for the internet. Think of the zillions of hours that have gone into IPSec, for pretty meager actual improvements in the world.

          • johan_larson says:

            I guess the question is whether you really need public-key crypto to secure the internet. Or could you get somewhere with private-key crypto and some notion of trusted servers?

            I work in computing, but I don’t have the expertise to make even an educated guess.

          • Mark Atwood says:

            Even if it changes nothing else, getting PKC into the public domain 25 years early will mean that Vint Cerf would have been able to bake it into IP in 1972, and then into all the higher level protocols.

            But getting it out 25 years early would change a lot more than that. Even not counting the proto-internet, it would have completely changed how the inter-bank protocols were designed, and how the international wire transfer networks were done. Having it be an option in unclassified teletype-like machines would have pushed so many details of the course of events down different paths.

            The reason why IPsec has been so difficult and so megre is because it is a bolt-on, and always will be.

          • dick says:

            I guess the question is whether you really need public-key crypto to secure the internet.

            Alice and Bob must have a shared secret to communicate securely, and the point of PKI is that it allows them to create a shared secret over an insecure channel. In the absence of PKI, a third party could serve that function – both Alice and Bob ask Charlie for a secret – but now you need to explain how Alice is able to communicate securely with Charlie.

            I’m not sure what a practical solution to this would look like in a world where PKI just didn’t exist. Disposable Yubikey-like one time pad devices that you buy at the Post Office? But until something like that is invented, the solution would probably be to just not encrypt anything and live with the fact that if you want to buy something from a website you need to call an 800 number.

          • johan_larson says:

            Disposable Yubikey-like one time pad devices that you buy at the Post Office?

            Or maybe little devices that you get from your internet provider that contain a specially generated encryption (outgoing) and decryption (incoming) keys for communicating with the internet provider. If the internet providers were big enough to be banks (or bank agents?) that would probably be enough to secure consumer payments in most scenarios.

          • John Schilling says:

            If it’s an insecure channel, how does Alice know she’s communicating with Bob rather than an impostor? For anything that requires real security, e.g. e-commerce, even PKI usually falls back on a centralized trusted authority to certify everybody’s keys.

            And in the e-commerce case, you’re presumably using a major bank credit card or an e-payment system like PayPal, so there’s no loss in privacy or security in having them handle that function. So, for e-commerce with only private-key encryption and otherwise 1970s-ish technology:

            Establish your PayPal77 account with whatever out-of-band identity verification would normally suffice for commercial purposes – and in 1977 that maybe is just a phone call with no ID blocking and your credit card at hand. Also you should know your mother’s maiden name.

            PayPal77 snailmails you a 5.25″ floppy with 10,000 indexed 168-bit 3DES keys, or if you’re in a hurry just emails the file to you under 3DES and reads you a 168-bit passphrase over the phone. Either way, you load the file into your PayPal77 browser plug-in (which won’t be called a “browser plug-in” in the 1970s, but whatever). One of those indexed keys will henceforth be used for all your communications with PayPal77.

            Every retailer who wants to do business through PayPal77 does the same.

            When you place an order with Retailer X, they ping the PayPal77 server with a packet (encrypted with their PayPal77/X 3DES key) saying “I need to establish a secure line with dick”. PayPal77 sends them an encrypted message saying “This is dick’s 3DES key #8124, and trust us, we know dick and this is him, that’s what we do here”. PayPal77 simultaneously sends you an encrypted packet saying “From now on, use your 3DES key #8124 for communications with Retailer X and no other purpose – also, let us know if you ever were or ever are hornswoggled into using #8124 for any other purpose”.

            Your browser seamlessly incorporates this new information – or maybe requires you to approve the update manually.

            From then on, whenever you do business with Retailer X, you and they use 3DES key #8124, which only you, they, and PayPal77 should know unless one of you has an internal security breach.

            This should be secure so long as PayPal77 and Retailer X are trustworthy, and if they aren’t then not even public-key cryptography will let you safely use them for commercial transactions. And it is compatible with many competing service providers, particularly if the browser plug-in is generic and open-source.

          • dick says:

            Yes, that is exactly what I was describing, except with the government rather than Paypal77 handing out the keys. (I’m sure that asking people their mother’s maiden name over the phone will work in the beginning, but at some point you’re going to have N people all claiming to be Jane Doe, and the only way to find out who’s really Jane Doe is the old “come in between 9 and 4 with a photo ID and two recent utility bills” method. The party that does this doesn’t have to be the government, but it’d be convenient, since that’s who has already built and staffed an office in every town in America)

          • b_jonas says:

            @Mark Atwood: Funnily explaining the assymetric crypto primitives is the easy part, on the education side. Could you also teach a symmetric crypto primitive that is fast enough to encrypt and digest everything on a 1970s computer, and convince the people back then that it is likely secure enough to be worth using?

      • Well... says:

        Not sure of how to do it, but make it so the default assumption is NOT that everything online is free.

      • johan_larson says:

        What’s wrong with the internet? At a low technical level, my impression is that we suffer from two low-level problems: security was bolted on fairly late in the game, rather than being designed in from the start, and the system does not distinguish between time-critical non-time-critical data streams.

        The two issues seem fairly easy to fix. It was the cold war, and the system was funded by the government. Raise fears of spies and saboteurs, and fund projects to deal with the possibility early. And for time-critical data streams, look for some application that is feasible at the low speeds of the early internet, such as telemetry of some sort, and fund efforts to make it happen.

        Culturally, I think many of the problems of how the internet is used have their roots in the notion that everything should be free, meaning without cost or all-you-can-eat. Spam, for instance, would never have been a problem if email had been a bit more centralized and charged by the message from the start, even at some low low rate like a penny a message. Similarly the vast and intrusive jungle that is adware would never have become much of an issue if there had been a firm cultural norm that services are paid for.

        This one is harder, because it rubs up against one of our cognitive blind spots. We love free. We love free stuff much more than we love really cheap stuff. But maybe baking in a billing system right from the start would help.

        • toastengineer says:

          Actually, front what I’ve heard from the people who were there, they WANTED to add security in from the start. The government stopped them, mainly via the whole “encryption is a munition and cannot be exported” thing.

          • Mark Atwood says:

            Vince Cerf does tell that story. He knew what public key crypto was, when IP and TCP were being designed, but he couldn’t use them, because it was classified.

        • John Schilling says:

          Culturally, I think many of the problems of how the internet is used have their roots in the notion that everything should be free, meaning without cost or all-you-can-eat.

          That’s a technical issue as much as a cultural one. If micropayments had been included in the architecture from the start, the prevailing notion would probably have been “ordinary things are cheap”, not “all things are or ought to be free”.

          Also, having payments of any sort baked into the architecture, would be a powerful motivator for getting the security right.

          • 10240 says:

            And the prevailing form of shady business on the internet would be “creative ways to milk people for micropayments”.

          • sorrento says:

            Hmm… I was there, and I remember it differently. The original online systems, like CompuServe, did have micropayments. Your time was metered, of course, and you were billed per minute you stayed online. But there were also areas of CompuServe and America Online that cost extra to access.

            People hated micropayments, and they hated being billed by the minute. As soon as the internet became available, people wanted to use it instead of those systems.

            Micropayments is NOT an idea that was never tried. People would not stop talking about micropayments in the 1990s. Big companies like IBM and DEC had whole divisions dedicated to it. The idea was tried and it just never worked.

            People just hate the idea that the meter is running. They will often take a free alternative over a much better cheap one. Partly it’s because people are irrational, of course, and they don’t always correctly value their time. However, I think there are good reasons to prefer a free thing over a cheap one, even if the free one is much worse. Choosing to spend money on something is a decision, and decisions take mental energy that people would rather not spend. Also, if the meter is running, there is a possibility of screwing up and ending up with a much larger bill than intended.

        • albatross11 says:

          Micropayments are hard to get right, though. Especially in a world where an RSA signature takes like 30 seconds. You might have killed the internet in its crib by binding a micropayment scheme that failed to it.

          More generally, a lot of this implies more centralization. It’s not at all clear to me that we’d have anything like the modern internet if it had been built with a lot of centralized control from the beginning.

          • johan_larson says:

            Some sort of online services were coming. It was steam-engine time. Quite a rich culture developed in France around Minitel, for instance, until it was done in by the internet. The question is what these counterfactual online services would be like. The premise here is that we want something at least a bit different from the internet of 2019, and maybe quite a bit different.

          • Jaskologist says:

            Does anybody have a good analysis of why Minitel failed where the internet succeeded?

            I personally think a whole lot of the success of HTTP, HTML, and even JS owes to the fact that they are text-oriented, and dead-simple to implement to the level of “non-robust, but kinda works enough to do stuff.”

          • John Schilling says:

            You don’t need centralization to do micropayments if you incorporate it into the architecture from the start. Crudely speaking, you can assign each packet a “cost” field, positive for content that will only be released on payment, negative for packets that carry a commitment to pay for their delivery (and perhaps prepay for a positive-value response). Increment infinitesimally each time they pass through a node, to cover transit cost, and cancel reciprocal transactions in bulk. Major backbone nodes would see enough positive- and negative-value transactions in each direction to mostly cancel and e.g. Sprint writes a small check to AT&T at the end of the week. An ISP whose customers are mostly content providers will be sending them weekly aggregate payments, one with lots of consumers will be sending them bills. Each node, each ISP, each user, deals only with their immediate neighbors and will have secure records to ensure those deals are fair.

            It would be possible to get this wrong, yes. But not having done it at all puts us in a different wrong place, one where we have to pretend advertising is going to pay for everything else. Or graft a centralized system like Paypal onto the rest of the internet. And as johan_larson points out, ARPA wasn’t the only game in town and someone else would probably have gotten it right.

      • 10240 says:

        Integrate a decent programming language (or bytecode interpreter, or sandbox for machine code), GUI toolkit and application development API into web browsers from the get-go, rather than using an ad-hoc patchwork of technologies, extending what was originally a document markup language, to develop web applications, leading to websites that use a multiple of the RAM and running at a fraction of the speed of an equivalent desktop application, chaotic look and feel, and limited functionality that is only slowly approaching completeness in recent years.

        • johan_larson says:

          Well, we did have Java applets there for a while. I’m not quite sure why they were abandoned.

          But yes, virtually anything would be better than JavaScript. To be sure, Eich was under extreme time pressure when he implemented it. But Scheme was right over here, and UCSD Pascal with a bytecode interpreter was right over there, and and …

          • albatross11 says:

            Having my browser execute code provided by a site on the internet seems really, really hard to get right in security terms. You can restrict the language, try to build sandboxes, etc., but it’s not clear to me that you’ll ever be rid of the security issues all this causes.

          • sorrento says:

            I actually don’t think Javascript is that bad, when used for its intended purpose. It’s a dynamic language that you can rapidly prototype things in.

            Writing graphical user interfaces in Java is like trying to play volleyball with a suit of full plate mail on. Especially with the early versions of Java which didn’t have things like lambdas, anonymous classes, and so on.

            Perhaps we could have removed some of the embarrassing mistakes from Javascript like the lack of integers and the difference between null and undefined, but I don’t think it would change things THAT much.

      • sorrento says:

        One really interesting question is whether centralized services like America Online or Compuserve could have survived or even thrived, instead of the internet.

        I think if the telecom companies had been more technically savvy, they could have merged with those guys and kept the internet balkanized for a long time, possibly forever. The telecom could give you a discount on things inside CompuServe, but charge you extra for accessing things from other services. Since there isn’t much competition in the local telecom markets, they could probably set up a cozy duopoly or tri-opoly.

        In fact, there is some evidence that this sort of thing will start to happen, with the end of net neutrality. The telecoms have a huge amount of control over local consumers, but have historically been technically inept, which let internet companies grab a slice of the pie. That could change.

        Another question is whether we could have avoided the whole “everything is just a document” phase in the World Wide Web’s lifetime. Could we have started with something like Hypercard instead of something like HTTP?

    • RavenclawPrefect says:

      Your mission, should you choose to accept it, is to make a human go as fast as you can under their own power on level ground.

      Your budget: A eugenics program, an atomic-scale matter assembly factory, and as many trillions of dollars as you need. No technology known to be possible is beyond your reach.


      (1) Your human must be a garden-variety homo sapiens – selective breeding is allowed, but you can’t tamper with genes directly and they’d better not have achieved any sort of evolutionary divergence. (Let’s set a time limit of 20 generations or so.) You can only start with one human, and they can’t bring materials to impregnate themselves or otherwise create additional manpower. Excellent health may be ensured, but no extraordinary anti-senescence measures (lifespan capped at 120, and less if you couldn’t reasonably expect them to make it that long). Conjoined twins and other highly noncentral examples of humans don’t count. Doping / drugging / illegal training methods / etc are all fair game, though.

      (2) Your vehicle(s) and track(s) must not contain any artificial mechanical force, whether that be electronically generated, battery-powered, or collected in some other fashion. You also cannot accept imparted force from external sources other than the usual ones acting on a human in still air – no earthquakes, no wind power, no surfing, no Project Orion, etc. Supplementary electronics, like a drone flying alongside your human to blow oxygen-rich air in their mouth, are OK.

      (3) You cannot store energy in any form – gears and levers and odd mechanical contraptions are all right, but at every point in time where force is being exerted on some object, there had better be a direct chain of cause and effect leading back to the human doing work at that moment. (So you can’t do things in the vein of “have them crank a turbine for 60 years then go ride in the plane they’ve charged up”.)

      (4) Your human and all materials must start at rest in a stable position relative to the Earth and be directly attached to the Earth’s surface (no geosynchronous orbit or letting go of a skydiver). They don’t need to stay this way, subject to the next constraint.

      (5) Your human’s center of mass should not move more than 5 feet from its initial position vertically, as measured by their gravitational potential energy. (Possible exceptions if you’re clearly obeying the spirit of “level ground” but that tight a constraint isn’t practical.)

      (6) You may not alter the Earth, or the orbital course thereof, in any way substantial enough to affect the conditions given above. (Minor terraforming projects, like flattening a continent or two, are fine, just don’t pull a Ship of Theseus on the planet.) You can build exotic structures, but your human shouldn’t be positioned above the current elevation of Mount Everest.

      The metric you are optimizing for is maximum observed instantaneous speed as measured relative to a stationary observer positioned at the starting point (which may be located wherever on Earth’s surface you choose).

      Everything else is fair game; you are not constrained by ethics, money, practicality, or the limits of what mere non-munchkins can dream up.

      • Aapje says:

        Is it allowed to produce a wake in front of the person that does not impart force from an external source on the person itself, but reduces air resistance?

        On a related note, is it allowed to create a long vacuum tube and have the person reach maximum speed inside the tube? A human-powered hyperloop seems like an option, if so.

        • RavenclawPrefect says:

          Yeah, wakes are fine, as are vacuum tubes. Part 2: what sort of actual speeds can you get out of such an operation? Seems like the limiting factors here are friction and the amount of sustained effort a human can put in to keep accelerating the hyperloop (high speeds = lots of kinetic energy to pour into the system, so you’re looking for a marathon runner rather than a sprinter).

          • Aapje says:

            I don’t know the achievable friction and residual air resistance in a hyperloop, but if a person can achieve 300 km/h (180 mph) when being towed, then I would expect something way higher.

            Especially if you put the oxygen supply on a separate hyperloop vehicle that follows the record-breaking vehicle (weight still matters, as it causes ball bearing friction).

      • AlphaGamma says:

        Bicycle-powered draisine in a very long evacuated tube.

        We know that a human on a bicycle in conditions where air resistance doesn’t apply can exceed 180 mph- the record was set a few months ago. Steel wheel on steel rail might be more efficient than motorcycle tyre on salt flat.

        Of course, Mueller-Korenek had to be towed up to 90 mph, as her bicycle was geared too high to pedal at lower speeds. So we need some form of propulsion system that works at low speeds as well…

        • Statismagician says:

          My heart now deeply desires a bicycle with rocket-assisted takeoff.

          What could possibly go wrong?

        • dodrian says:

          I’m thinking a reticulated bicycle, as you say on steel tracks in a vacuum tube.

          You could use a geared system connected to a hand-powered cycle to push the bicycle up to speed first. Use a clever interlock so that the pusher can be ditched for the final record.

          I think the biggest drawback with this approach is you need a fair amount of weight if you want a pressurized capsule for the human to lie in. Possibly a pressure suit would work, but they need unencumbered motion of their legs.

        • Lambert says:

          Now that I think about this, I realise that a vehicle stores energy in much the same way as a flywheel, except linearly.
          Maybe boosting speeds by putting weights around the outside of the bike wheel, then suddenly pulling them all in to convert to higher speeds would be close enough to be allowed.
          Use liquid ventilation, or even run their circulatory system open-cycle on perfluoromethyldecalin.
          Replace any non-driving wheels with superconducting maglev. Look into replacing a mechanical drivewheel with some kind of mechanical inductive system. Like , which spins an array of permanent magnets. I’m sure that setup can be modified to induce a linear force.
          With no friction or air resistance, size and weight become less of an issue, as does power output.
          The hard bit will be all the gearing that is required to reach these speeds.

          • AlphaGamma says:

            I was wondering whether maglev would fall foul of Rule 2.

            If it doesn’t, would we be allowed to just power the propulsion motors of a maglev using a human-powered generator? IIRC maglevs normally tend to have the propulsion magnets outside the train (for obvious reasons, as it means the train doesn’t need catenary or an onboard power source), but I imagine this can be reversed so a human-powered electromagnet on the vehicle acts on a permanent-magnet track…

          • Lambert says:

            I don’t think electromagnets (or any electricity beyond eddy currents) are necessary.
            A moving array of permanent magnets should be equivalent to the moving magnetic field used in an asynchronous linear motor.

      • bullseye says:

        It seems to me that the human should be selected (or bred) for smallness, in order to maximize strength/weight ratio. But wait, are real life champion cyclists small? Quick internet search says they’re average height and lean. Maybe strength/weight ratio does matter, but there’s also an advantage in having longer legs? Maybe human size is already optimized for speed and it somehow works out the same for running and biking?

        • Aapje says:

          No, maximizing the strength/weight ratio is important when climbing. When the main resistance to overcome is air friction, the key is having good power to drag.

          Ultimately an important question is for how long power should be generated. Road time trials are really endurance events, where a high speed has to be maintained for about 50 kilometers. Road sprinters typically have to ride 100’s of kilometers they get to the finish where the sprinting happens, so any sprinter who is optimized for the sprint won’t actually get to the finish in time to sprint for the win. In contrast, you have indoor cycling events where the time trials/sprint events are as little as 1 kilometer.

          Here you can see a picture of the legs of a road sprinter on the left and a 1 kilometer indoor specialist on the right. The latter’s thigh’s were measured at 73 centimetres (29 inch).

          Note that this amount of muscle can’t be fed by oxygen, so they are only beneficial if the effort is so short that much of it can be anaerobic (so less than two minutes or so).

          If you can somehow store great energy in the vehicle, for instance by using a hyperloop with very low friction bearings, the best approach is presumably to generate power for a long time. So that would mean limited muscle mass that can be fed aerobically continuously, albeit perhaps with a sprint at the end. So very similar to cyclist who sprints in road events.

      • Skivverus says:

        A note: since we don’t have to worry about repeat usability, we should probably include methods for converting arm muscle power to forward speed in addition to legs. For that matter, while we’re considering non-leg muscles, the human body’s got a *lot* of muscles that pull in all sorts of different ways, and net-positive work can probably be extracted from a lot of them, though per “not constrained by ethics”, maximum extraction is likely to involve vivisection of the “runner”. Hopefully those trillions of dollars extend to machines to reverse the process once the maximum speed’s been achieved; will bill it under “methods to ensure the subject does not freeze up or expire from trauma” alongside the anaesthetic.
        Side benefit: reconstructive surgery gets a big boost from the blueprints for machines at the end.

    • Paul Brinkley says:

      Your mission, should you choose to accept it: figure out a way to mine landfills for useful materials. Bonus points if the process could be made cost effective. (Theoretically, I suspect this is possible, since you would be saving yourself the effort of finding such caches of material in the first place, and since you know those materials were valuable at some point.)

      (Inspired by a comment by Mark Atwood.)

      • Mark Atwood says:

        It’s already sort of done. Quite a few “mines” have been taking the yuge pile of gravel left over from an existing older played out mine, and processing it again with newer tech for same or different materials, as economics and processing tech changes. I know of one tails pile in the US southwest that has been reprocessed 5 times for 3 different metals, with the original “mining” having been done over 100 years ago.

        • Paul Brinkley says:

          That is indeed pretty interesting.

          How well would those processing methods fare against a block of compacted iron, steel, couch cushions, bed frames, paper, wax, glass, plant fiber, and clay?

          • Mark Atwood says:

            How well would any known ore extraction process work against that? Not well at all. Trash contains enough surprise! bits of lead, mercury, cadmium, and/or chromium to surprise poison any one intermediate processes any one random day out of each month. At which point, your plant shuts down, AND you’ve just turned part of your plant into more trash.

            And then, back uptopic, there is lots and lots and lots of plastic stuck to nearly everything, that will detach from what you want only with fire (bad) or really hard to clean-for-reuse toxic solvents (bad).

            The only reprocessing processes that come to mind are: a star trek filtering transporter, or an army of billions of literal Maxwell’s Demons, or else just magic up a free-energy plasma arc to reduce it to plasma and then sort the resulting atoms by mass.

            Just bury it, and wait for it to subduct.

          • hoof_in_mouth says:

            You can get a lot of the organics out via superheated water, but since you need a basically infinite energy source for that you might as well go directly to the plasma.

          • johan_larson says:

            What might you find in a trash dump that would make it profitable to dig up the dump and pass it on conveyor bets past pickers, with maybe an intermediate step to chop things into fist-sized lumps? Let’s be very optimistic and suppose you could pay your pickers $1 per hour, not including the digging up and processing.

      • CatCube says:

        This is an interesting one. All of the current methods of doing this for mining ore rely on beneficiation, which made mining formations of lower-percentage ores feasible.

        To explain what I mean, a bit of history for iron ore mining in the Great Lakes region: Underground mines from 1844 to about the late ’40s (with a few continuing for another couple of decades) mostly relied on direct shipping ore. It’s worth explaining what is precisely meant by “ore.” Ore isn’t just “rock with the metal you want in it.” It’s a specific mineral. For example, hematite and magnetite are both common ores that are about 70% iron by weight. Direct shipping ore is simply when the run-of-mine rock has a high enough percentage of iron ore relative to waste minerals that you can just dump it into a blast furnace, generally about 60% iron or better.

        The town I grew up in was founded around underground mines. They mined veins of very pure iron ore, which paid out in 1928. However, in the 50s, they came back and put an open pit mine directly over the old shafts, even though they had already mined out the high-percentage ore. What changed? Beneficiation. The rock surrounding those high-percentage ore still had a lot of iron ore in them (~50% IIRC, or around 30% iron), just scattered into little pieces throughout the rock. They now would take this what was considered waste rock before, grind it up so small that any particular piece is going to be either mostly ore or mostly waste rock (tailings), then use the different properties of the ore and tailings to separate them. At that mine, hematite was the common ore, so they couldn’t use magnetic separation; they used froth floatation to cause the lighter tailings to come to the surface and the ore to sink, Edit*: to cause the ore to float and leave the tailings to sink, concentrating the ore until it was about 60% iron. This used far fewer people and made use of what was originally a waste product, but is very capital intensive; the difference between the surface plant of an open-pit mine using concentrators to the old underground shafts shipping run-of-mine rock is pretty stark.

        Areas where the native metal is found used gravity separation–just using an additive to water to adjust the specific gravity of the fluid to be between the SG of the metal and the tailings–even earlier (I know of the late 1800s), but that didn’t require nearly as much grinding effort as froth floatation or magnetic separation because the metal was in larger chunks and had a much higher SG differential.

        To come back to your question, how would you beneficiate the trash? It’s a much sticker problem than grinding rock, which would be much more uniform in composition and therefore much easier to subdivide into small, even pieces. You’re going to be mostly dealing with native metals, so if you can solve that, you can probably use gravity separation since most metals will be the densest stuff in the trash.

        * I had a brain fart and wrote this backwards; the chemicals added stick to the iron ore and form bubbles to bring the ore to the top.

      • hoof_in_mouth says:

        You would treat it like a normal mine… grind or crush the “ore” into a smaller and smaller bits until a combination of mechanical separation and chemical treatment can be used to separate what you want from the tailings. You do have to pick end-products to optimize for and the “ore” is really low grade. Think of e-waste and you wanted to recover the metallic tin (there’s not really much lead in electronics, if any). Grind sufficiently small, separate tin-containing and non-containing if you can, then treat with acids to get the tin into solution, then get the acid off the solids and treat that, then get the tin out of the solution and hopefully recycle the acid…, and that’s just one thing, if you wanted to recover lead or mercury or copper it’s a different acid, and what happens if there is a contaminant that poisons your reaction? It’s never easy, never clean, never efficient, which is why the Chinese do it instead of us now. We already recycle the stuff that is easy and cost effective (plain metals), very little of that goes to landfills.

      • bullseye says:

        Low-wage countries have people living in landfills who go through the trash for a living. Electronic waste is valuable enough that China buys it from overseas.

      • Tenacious D says:

        I’m going to go with one that’s already done at some landfills: tapping them for gas.

        Another possibility might be to selectively precipitate some useful stuff out of the leachate since it needs to be treated anyway.

  35. idontknow131647093 says:

    The new Netflix Black Mirror episode was called “Bandersnatch” and was a choose your own adventure thing.

    I have to say I’ve never encountered a less compelling RPG in my life. Did anyone like it? Why?

    • I agree with you. It felt like it’s entire point was just to rely on its gimmick without any deeper meaning going on.

    • cassander says:

      I was baffled by it. I can’t quite figure out why anyone involved thought what they had was compelling. I can only imagine that they started out with some fun ideas (there are hints of them in the early stages) but had to massively scale back when they figured out how much filming their original plan would have cost and ended up putting out this.

    • Gobbobobble says:

      Black Mirror continues burrowing through its own large intestine, news at 11.

    • LadyJane says:

      I suppose I’m the odd one out, I actually enjoyed the experience for what it was. Admittedly, it didn’t have the depth of characterization or social commentary that I’ve come to expect from Black Mirror, and the introduction of supernatural elements was a pretty stark departure from the show’s normal techno-dystopianism (unless you go with the fan theory that Stefan was in some kind of simulated reality, as hinted at by the White Bear symbol). But it was still a fun and compelling experience to me.

  36. Radu Floricica says:

    Looking on recommendations for (practical) political reading. I may or may not be involved with the strategy part for local elections in my home town, and if I am, I want to do things properly. I budgeted a couple of months for studying.

    Open to anything good, both theoretical, like for eg. Haidt’s Moral Foundations, or very practical – anything on early communism, maybe multiculturalism in US. We may need to do a lot of offline work, because we’ve mostly topped the online electorate and there is quite a bit of rural, old population here. Also stuff on training/organizing the team.

    • Erusian says:

      Politics is one of those industries where knowledge is very much passed around through networks and not openly available. I’d recommend finding a mentor. Otherwise, Shattered, Fear and Loathing on the Campaign Trail, and Losers: The Road to Everyplace but the White House all seem to contain fairly accurate facsimiles of what I’ve seen in politics. You’re generally going to find more material from losers than winners though, since winners go on to office.

      I’ve also found political campaigns tend to have no business or organizational sense as such. To some degree, I get that. Political campaigns aren’t self-perpetuating profit driven machines. But managing finances and personnel are needed and often rare. And politics is more insulated from… well, the effects of their decisions, so a lot of people rise high without much in the way of hard skills. So I’d read some books on business management and probably statistics etc.

      Old school books on propaganda (like the CIA field manual) can also have some utility. But for the most part, the situations are only somewhat analogous.

      If you’re dealing with old, rural voters, you just need to use the channels they do. Television, radio, billboards, plane trailers…

      • Mark Atwood says:

        a lot of people rise high without much in the way of hard skills

        That was my primary takeaway from reading both “Shattered” and “What Happened” about the failed campaign of the 2016 election.

        So many many really terrible managers. At every level from the top to the middle. (The bottom levels were mostly ignored by those two books. One of my theories is the bottom levels are full of unsung highly talented hardworking skilled people who are killing themselves keeping everything sort of kind of working at all.)

        They were terrible at management, terrible at people management, terrible at meeting-running, terrible at accounting, terrible at resource management, terrible at mass logistics, and really not good at marketing.

        They were terrible to their staff, they were terrible to their peers, they were terribly organized. They weren’t even good at “bad politics”, their rounds of backstabbing were laughably incompetent from the outside.

        And these books were written by *sympathetic* people *on their side*.

        I have some exposure to corporate politics that have iterated hard in century old poorly managed companies, and some slight exposure to the viciously poor people management in academia. All of which were staffed with more effective and more humane people managers than that campaign was.

        • Deiseach says:

          So many many really terrible managers. At every level from the top to the middle. (The bottom levels were mostly ignored by those two books. One of my theories is the bottom levels are full of unsung highly talented hardworking skilled people who are killing themselves keeping everything sort of kind of working at all.)

          Haven’t read the books, but from the extracts and similar articles published online that I have read, it seems like (a) the bottom levels were the traditional grass-roots who got ignored (e.g. union chapters who put together fundraising and organised meetings, expected the candidate to at least show up for one ‘hey guys great work’, and were totally blown off by her campaign management because pffft, bluecollar white guys are so not the demographic we’re targetting) and (b) everyone was so completely convinced Hillary was going to win that they didn’t bother running the campaign, they were more concerned with the backstabbing to get close to Hillary*, ensure they were her new BFF, and cut off access to her for their rivals – since they were all competing to be considered for the plum jobs that would be handed out by the Empress when she would be putting her administration together. So this meant Jack would not talk to Jill since that would give Jill valuable data which she could use in her campaign to unseat Jack.

          I still love reading this piece on Hillary’s data-driven campaign because as results show, they got it so wrong, yet at the time this was “Hillary inherited Obama’s data campaign and since that was how he won she is gonna crush Trump who is a dinosaur and hasn’t the deep pockets to compete with the money her campaign is throwing at targeted ads” triumphing:

          It was an algorithm from Kriegel’s shop — unreported until now — that determined, after the opening states, where almost every dollar of Clinton’s more than $60 million in television ads was spent during the primary.

          The tool bypassed the expertise and instincts of her traditional media buyers by calculating the “cost per flippable delegate,” in the words of one senior Clinton official, and then spat out what states, television markets, networks and shows to buy. Obama veterans were wowed by its advancement; internally, some Clintonites saw it as their secret weapon in building an insurmountable delegate lead over Bernie Sanders.

          Now, with Donald Trump investing virtually nothing in data analytics during the primary and little since, Kriegel’s work isn’t just powering Clinton’s campaign, it is providing her a crucial tactical advantage in the campaign’s final stretch. It’s one of the reasons her team is confident that, even if the race tightens as November approaches, they hold a distinctive edge. As millions of phone calls are made, doors knocked and ads aired in the next nine weeks, it is far likelier the Democratic voter contacts will reach the best and most receptive audiences than the Republican ones.

          Yep, you guys sure wowed them! 🙂

          *Another, related, flaw seems to be that Hillary did/does have these little inner circles of tightly-knit advisers whom she trusts completely and confides in, and unless you are one of these and have her ear, fuhgeddaboudit. Hence all the climbing over each other to reach the level of court favourite.

          • Plumber says:


            “….it seems like (a) the bottom levels were the traditional grass-roots who got ignored (e.g. union chapters who put together fundraising and organised meetings, expected the candidate to at least show up for one ‘hey guys great work’, and were totally blown off by her campaign management because pffft, bluecollar white guys are so not the demographic we’re targetting) and (b) everyone was so completely convinced Hillary was going to win that they didn’t bother running the campaign, they were more concerned with the backstabbing to get close to Hillary…”

            Anecdote isn’t evidence, but I remember many more bumper stickers and pins being handed out for Kerry in 2004 and Obama in ’08 than were handed out in both 2012 and ’16.

            Clinton would have had a higher climb regardless, support was stronger for either Sanders or Trump in the primaries, my wife liked Clinton in the primaries, but that was rare in my neck of the woods.

          • Nornagest says:

            I noticed a lot more Hillary ’16 bumper stickers after the election than before it. (Lots of Sanders stickers before, though.)

        • Winja says:

          And these are the same motherfuckers who want to run the country.

          Jesus Christ.

          • CatCube says:

            I hate to break this to you, but they didn’t come out of nowhere, and were merely expanding existing campaigning. This is also found among the motherfuckers who do run the country.

        • BBA says:

          We’ve discussed the phrase “late capitalism.” I think we’re living in late democracy.

          • Tarpitz says:

            Reading Dickens’ account of an 1830s election in the Pickwick Papers does not incline me to believe in some bygone golden age of democratic virtue and lack of cynicism.

    • Paul Brinkley says:

      It’s somewhat adjacent, but still strongly implies a great deal of what you see in US politics: Neil Postman’s How To Watch TV News.

    • TDB says:

      Maybe not what you really want, but Rick Falkvinge’s description of the birth of the Pirate Party (title is Swarmwise, he has it available online with a Creative Commons license) is interesting.

      It may be out of date by now, has anyone seen a comparable thing since Trump?

    • Anon. says:

      Dominic Cummings on the Brexit referendum is the best thing I’ve read on practical politics. It’s huge but 100% worth it.

      • Aapje says:

        Yes, extremely rationalist-friendly and interesting.

        Though as far as I recall, his main points can be summarized as:
        – Get a core group of very competent people together
        – Measure whether your slogan(s)/messages are memetically strong
        – Focus on getting your memetically strongest slogan(s)/messages out there in the most cost-effective way possible

        • Statismagician says:

          This seems like a pretty fair summary, but also I can’t conceive of how this wouldn’t already be standard practice. I notice that I am confused; does anybody know what’s up here? Is it really just that politics is almost completely insulated from/politicians somehow incentivized to avoid rational decisionmaking? I mean, to some degree, sure, but to the ‘use the best* advertising cheaply** is a new idea’ degree?

          *And test this.
          **Avoiding obvious trap options, e.g. the 3:00 AM broadcast slot right after [I don’t know, something that’s on TV at 3am].

          • Kyle A Johansen says:


            I would say what is up is:

            a) It is a question of priorities. Should you ‘use the best* advertising cheaply** is a new idea’ or focus on getting the best people where they are most seen, or something else. I don’t see any obvious reason for why it ought to have been the former that is where one ought to prioritise the former over the latter.

            b) How do you test for the best? Referendums aren’t that common, and Cummings in that post warns against putting too much weight on the result of this one (and many people suggest that project Fear put too much weight on the Scottish one). of course, you might try to optimise for something other than effectiveness at winning the election, but what? Cummings obviously says a mix of polling and lots of focus groups, but there are theoretical arguments against that* as well as different approaches within that broad method.

            c) Cummings doesn’t claim that it is a new idea, but more that it is the best idea and not recognised as such. The post is about its efficacy, not its novelty.

            * e.g. ‘Shy Tory’ effect.

            Edit: I would say that my general point is that if Cummings had written a post with different priorities – perhaps if he had stressed the ‘Boris’ effect and his Wetherspoon polling – that you could have gotten the same ‘well, duh’ reaction. (Similar to the ‘well, birds of a feather’ ~ ‘unsurprising, opposites do attract’ effect.)

            I wouldn’t trust my reaction to evaluating the content. I would base it on his clearly having a track record of success, and people who write giant blog posts generally being the honest sort; leavers doubly so.

          • Statismagician says:

            In re: your edit – that’s just the thing; I would be a lot less confused if this seemed like the class of issues where all positions seem equally reasonable if argued well. I’ve never believed the opposites attract canard, I know enough about sampling, social-desirability, and self-report biases not to put much stock in that sort of polling for epistemological reasons, and I honestly don’t know a great deal about Boris (sorry, American) but I have a bunch of posts here and a lot of real-life history arguing that no, it’s economics/vast social currents rather than Great Men for a whole slew of issues. This just seems… I don’t know, weird, in a way I can’t quite pin down.

          • Kyle A Johansen says:

            Ah, but Cummings is arguing against ‘economics/vast social currents’, and the idea that Brexit was inevitable. Instead, he is saying that ultimate victory came from solid strategic/tactical choices (on the part of Leave), luck, a strategic/errors on the part of Remain.

          • Aapje says:


            I can’t conceive of how this wouldn’t already be standard practice.

            I think that if you allow things to develop naturally, you end up in a poor place. For example, the people who tend to be most driven and competent at political games, yet not very competent in other ways, seem most likely to end up working for your campaign for free or little money.

            These people also tend to be very partisan and thus very much in a partisan bubble. So they like slogans/memes that appeal to the (very much) already convinced, well educated, etc & to forms of advertising that appeal to them (because they are ‘proven,’ respectable, target people like them, etc).

            Cummings’ proposal is to turn things up to 11, not just by picking the best people from those who volunteer, but by pro-actively approaching the top tier of problem solvers (he suggests physicists).

            I think that Cummings’ proposal in general is not actually radically different from what people are already trying to do, but that he is mainly radically less willing to compromise or take shortcuts.

          • Statismagician says:

            @ Kyle A Johansen

            Fair; I should have been clearer. My interpretation is that Cummings’ position is that the pro-Brexit side had any chance of winning at all due to economics/broad social currents, and only actually won because (the specific set of things: Vote Leave did a lot of things right, Cameron et. al. did a lot of things really really wrong, and sheer dumb luck, where if any of those were not true they wouldn’t have won). Necessary vs. sufficient, in other words.


            All of that makes sense, it just seems like it should be insufficient to account for the degree of incompetence Cummings describes. I admit I haven’t worked for a political campaign or the government and so could just be wrong about the scale of the problem.

          • Walter says:

            I think the reason behind the immense failure of the political hirelings is simply that what looks like a failure to us, but falls within their own tolerances.

            Relevant, Mr. Levine’s observation that a successful hedge fund manager is not necessarily one whose picking is accurate, but rather one who is able to remain a hedge fund manager.

            To the Minister the disarray of Whitehall is a catastrophe, an insupportable disgrace, but to Sir Humphrey it is a feature.

    • JohnBuridan says:
      Jeremy Bird has some stuff online that may be useful.

  37. dndnrsn says:

    RPG thread: what is the earliest game to have “plot points” or “fate points” or any number of similar names, and are they more common now than they used to be? Points that can be spent (temporarily or permanently) to give some sort of ability to cheat fate – at minimum, a reroll or similar,

    I had always thought it was a relatively recent element, showing up in force around the same time storytelling games or storygames or whatever you call them did (so, early to mid 2000s). But I found a used copy of the 1983 James Bond game, and it has points that can be used to turn failure into success, enhance success, or negate damage, to the degree of being able to survive impossible (or at least highly improbable) things. I think Warhammer Fantasy Roleplay from 1986 had them; I know the second edition did but it’s much later. I had thought that it was a fairly “new school” thing but am now thinking it might just be due to uneven exposure to different games from different periods. Alternatively, if those two are outliers, I’m still sorta right, but that raises the question of why they didn’t become more popular – given that mechanics of the sort in James Bond (which also has the cool wrinkle that villains get points that can only be used defensively) would fit perfectly a lot of campaigns where people are expecting a game where the heroes have some plot armour, are using a system that doesn’t give heroes plot armour, and the resulting disjunction either causes problems (people who want plot armour but do not have it are unhappy, scenarios where plot armour should be available get messed up) or is fixed by GM fudging, which creates more problems.

    (Of course, given that D&D is the majority of RPG sales, what D&D is doing at a given moment in time is more important than anything else – Inspiration is kinda a plot point mechanic. I don’t know about 4th, though. I don’t think 3rd had anything like that out of the box.)

    • Nornagest says:

      3rd didn’t have anything like that in the core books, but Pathfinder started experimenting with them somewhere in its splat lifecycle (though, to be fair, it experimented with just about every conceivable mechanic somewhere in the splats). 4e sort of did, but denominated them in actions and didn’t explicitly couple them to plot or fate or anything like that.

    • bean says:

      I know there were Force Points in Star Wars D6 (late 80s, reasonably prominent), which were brought into Star Wars d20 (Wizards-owned, also prominent.) The later d20 Star Wars had Action Points, too, which were a toned-down version of the same (1 point/level for FP, a lot more for AP).

    • sandoratthezoo says:

      You could spend Karma (experience points) for temporary boosts in Marvel Super Heroes roleplaying, which dates to, um… 1984.

      • broblawsky says:

        I think this actually might be the first one.

        • Protagoras says:

          Nope, as dndnrsn mentioned, the James Bond role playing game had that in 1983. One of the games I remember from my first gaming experiences in the early 80s. It’s starting to sound like that might have been a pioneer here, however forgettable it might have been in other respects.

    • C_B says:

      D&D introduced “Action Points” in the original 3.0 Eberron campaign setting (2004), but they remained a variant rule outside of the Eberron setting for the rest of 3.0/3.5’s run. Originally, you just got a set number of them per character level, but they later tacked the “also, the DM might award you an action point if you do something especially cool” advice onto the same mechanic.

      • Unsaintly says:

        D&D’s first (as far as I can tell) introduction of action/fate points was in AD&D 2nd edition, with the book Skills & Powers. This provided characters with Character Points, which could be spent on leveling to gain traits, proficiency, class skills, etc. However, CP could also be used in the course of play to reroll any failed check. While this suffered from the (at the time common) problem of using your advancement points as rerolls, it was still an early pioneer of the idea. Nothing predates the 1983 James Bond instance though, at leat to my knowledge

    • broblawsky says:

      Shadowrun had a Karma Pool in 1st edition, AFAIK, and that came out in 1989.

    • AlphaGamma says:

      WFRP 1e definitely has Fate Points. They can be spent only to avoid death. The example given in the rulebook is someone spending a Fate Point when his PC is about to be killed in combat- the GM decides that the character is instead knocked unconscious and captured. They are almost impossible to get back- GMs are supposed to give them out only once per adventure or less.

      2e split them into Fate and Fortune Points. Characters start with the same number of each, but Fortune Points replenish every day and are much less powerful, allowing a re-roll on a single failed roll (or similar), while Fate Points work much like in 1e.

  38. Walter says:

    Random question, is there anyone on here who can give a deep dive on how Warhammer 40k got started? Like, nowadays there is all this baroque IP covering everything, 1-3 new novels from Black Library every month, etc, but it feels like the initial creation of the whole setting was incredibly seat-of-the-pants.

    I recognize that there’s no particular reason to expect SSC to just happen to have someone deeply interested in random ancient wargaming story in the commentariat, but in all honesty I feel like I’ve got about even odds on this. Any takers?

    • cryptoshill says:

      Everything 40k pre 2010 or so was *entirely* seat of the pants. Sex-demons? Check. BDSM-fetish elves that worship the Sex-demons? Double-check! Every single big bad alien race from every sci-fi ever? Check! Weirdly medieval future-marines? Yup! Rambo? Yep he’s in there! 70s biker gangs with mysteriously advanced technology? They disappeared from the canon, but there’s a periodic grassroots effort to bring them back.

      I don’t know who actually is responsible for the horrific mess that 40k is, but I appreciate the insanity for its own glory.

    • Deiseach says:

      Oh gosh, all I remember is very vague from reading a run of White Dwarf back in the day (the 1987 magazine shown in the Wikipedia article is one I remember), but it seemed to grow out of their original tabletop fantasy wargaming – hey we’ve done fantasy, how about if we do a sci-fi version as well? And thus you get a far-future space opera where Orcs are, um, Orks and there are versions of Elves etc. and you have the God-Emperor and Space Marines (because tabletop wargaming) fighting existential threats to the very survival of humanity and sanity, turned up to 11 just because!

      There seemed to also be a healthy influence of Michael Moorcock in the Law versus Chaos, Chaos Gods, and so on but I couldn’t swear to it 🙂 Had a bit of a search around online and this gives a good recap of the run of the magazines and the contents, you should get some hints to the creation and rise of WH40K as chronicled in the official Games Workshop publication (the mid-80s are when the change in emphasis happened and they dropped scenarios for other games and concentrated heavily on their own stuff). Issues 91-100, according to this run-through, are where Warhammer (both fantasy and 40k) take over:

      The imagination of Rick Priestley begins to inform a lot of the content of the magazine with his many articles discussing the WH40K universe in exhaustive detail, but there is a steady trend towards supplements for the miniatures wargames at the exclusion of all else, and the rise in features where articles on new mechanics for Warhammer and adverts for figures begin to bleed into one another.

    • Nabil ad Dajjal says:

      In the 80’s there was a British comic called ‘Nemesis the Warlock’. It was published by 2000AD, the same people who gave us Judge Dredd, and had a similar tone of dark humor in an over-the-top satirical dystopia. The Terran Empire, the antagonists of the comic, are hyper-militaristic religious zealots fighting against the demonic alien mage Nemesis with the decaying remnants of their formerly technological society. Sound familiar?

      Games Workshop started producing Warhammer 40K seven years after Nemesis was published. A lot of that early material was similarly satirical and visually very similar; it wasn’t until the lunatics started running the asylum years later that it began being played straight. The earlier material was much more self-aware.

      It’s actually a shame, because a lot of what I consider the cooler elements in the fluff were part of that stuff that ended up side-lined to publish more Ultra Smurfs books. Things like the Imperial Knights and the Rogue Traders drove home the idea of a rotting yet still-mighty empire a lot more than bodybuilders with oversized pauldrons punching demons.

    • dndnrsn says:

      I’ve got a couple White Dwarfs from the late 80s. First edition Rogue Traderlooks like the fluff was… if not less grimdark, then at least the theme of the human team was less “enormous theocracy where everything is covered in skulls and scrolls and candles” than it became… Maybe in 3rd ed? All the 2nd-ed era stuff I saw was less grimdark and less about nothing but WAR. The Imperium is a lot more interesting when it isn’t just “Catholic space fascists” or whatever the joke was. There’s a lot more in the grim darkness of the 41st millennium than war: the Imperium is not on a total war footing, nor has it instituted universal conscription. This is true in later fluff, but later fluff is about nothing but war (isn’t there some fan theory or something where the picture of the Imperium in 40K is that presented by Imperial propaganda?)

      As noted in the post above, there’s a point where W40K starts getting played straight. This was probably due to the shift to a younger, more American audience. Teenagers are bad at telling when something isn’t being played straight and teenage boys are prone to see anti-war stuff featuring war and think “war is cool”. Meanwhile, Americans do satire differently from the British (although the culture gap may have narrowed in recent years, presumably due to DVDs and then video files making transmission easier). I’d guess that the majority of the adult fanbase now recognizes that the whole thing is kind of silly, but in a “so dumb it’s awesome” way (like Metalocalypse, not Spinal Tap).

      The rules system from the early days looks more RPG-ish; named characters get more detail. Probably it got played on a smaller scale. I think you could do less martial setups, like “authorities vs genestealer cult” but I don’t know for sure.

      Interestingly, the 40K RPGs (or, the first couple) present an Imperium that’s not just at war constantly, and I keep hearing about Kill Team, which I gather is a skirmish game, less straight-up martial. I gave up on the 40K RPGs based on the awful rules and flimsy binding of the first couple and haven’t touched a miniature in years.

      • Le Maistre Chat says:

        As noted in the post above, there’s a point where W40K starts getting played straight.

        Well, having their art subverted away from them couldn’t happen to more deserving people than subversives who treat the Catholic Church and Nazism as the same.

        • dndnrsn says:

          The presentation of the Imperium in the early stuff is both less of a theocracy and less “weird baroque” in its presentation. Stuff’s less dialled up to 11. The “Catholic Space Nazi” thing is something I’d associate more with the presentation starting maybe late 90s or early 00s.

        • Gobbobobble says:

          Ehh, I know it’s in vogue these days to brand everything as Nazism but that’s a bit of a stretch.

      • AlphaGamma says:

        I think the famous line “in the grim darkness of the far future there is only war” is from 2nd edition, where it is on the introductory box set. Though the cover art of that box, while it has got skulls and scrolls, is much less dark than the 3rd edition cover art.

        This may also be related to changes in GW’s art style and direction in general, I think at some point between the release of WFB 5e (1996) and 40k 3e (1998).

        • dndnrsn says:

          The art is both brighter and less… there’s less dangly stuff and holding a flamethrower next to your tunic and parchment seals.

      • Nick says:

        “Catholic space fascists” or whatever the joke was.

        Yeah, I’ll be honest, I don’t even see the appeal of 40K. I actually like grimdark stuff, generally! but it seems to me like there is just nothing worth saving in the grim darkness of the far future. Gratuitous space gothic is its only redeeming feature.

        • dndnrsn says:

          It’s too bad because a lot of the fluff behind the main RPG line shows an Imperium that’s an interesting place. It’s basically got a major problem of medieval societies: bad communications and travel. The bits of it that are near trouble take on a disproportionate share of that trouble – there’s War Planet Warplanetius, where the economy makes that of the USSR 1941-45 look pacifistic, which is right next to a bunch of orks or something, and then an impractical distance away is the paradise-planet Pax IV or some hive world or something.

        • theredsheep says:

          Do they have wives and children in this horrible far future, or just cloning pods to produce more perfect soldiers for war against the damned? I know little about 40K except that it has ludicrously over-the-top space soldiers who love their emperor, and also that Blizzard ripped it off extensively for their own world of space humans vs. space elves vs. space orcs.

          • Gobbobobble says:

            The over-the-top space soldiers are selected from the general population as youths and given a fuckton of drugs to kick puberty into crazy overdrive and also a bunch of surgeries, primarily the implanting of a bunch of extra organs.

            And, at least in the canon of my teenage years, actually those dudes are the least religious about the Emperor. View him more as a GOAT Martial Leader to respect and emulate than an “I’m averting my eyes, oh Lord” type of God. It’s the unaugmented humans that go for the dialed-to-11 space catholic rites.

  39. j1000000 says:

    Last Open Thread, Aapje recommended the documentary Wild Wild Country, which I was a big fan of. Made me wonder if anyone around here is into documentaries and which ones they like/recommend.

    • Well... says:

      I just re-watched “Koyaanisqatsi” last night for the umteenth time since I was a teenager. Never disappoints.

      Herzog’s “White Diamond” was pretty great, though it’s been a while since I saw it.

      One I watched more recently and enjoyed was “Room 237” (or, whatever number, I forget), the behind-the-scenes story of Kubrick’s “The Shining”.

    • Chlopodo says:

      I was unusually impressed by “Breaking the Maya Code”, a 2-hour documentary about the decipherment of Mayan hieroglyphics, based on the book by the same name. I do not recommend, however, the similarly-named “Cracking the Maya Code”, which is a 1-hour long, dumbed-down version of the same documentary.

    • nkurz says:

      Adding to my response to Aapje in the previous thread, I want to second his recommendation of Wild Wild Country. I’m only three episodes in at this point, but it’s fantastic. It touches so many of the threads that preoccupy SSC. I’d love to hear what others here make of it.

      For my own recommendation, the best documentary I’ve seen in recent years is “The Salt of the Earth”: It focuses on the life and work of the Brazilian photographer Sebastião Salgado. It’s co-directed by Wim Wenders and Salgado’s son. It’s visually stunning, and reaches some of the lowest emotional lows and highest emotional highs of anything I’ve seen. It’s available as a DVD from Netflix, or as a $4 rental from iTunes or Amazon.

    • Anon. says:

      F for Fake

    • Aapje says:

      The King of Kong is a great docu about a fierce competition over the Donkey Kong record.

      The Last of the Unjust, a documentary with Rabbi Benjamin Murmelstein who worked with Eichmann and was a high-ranking member of the Theresienstadt concentration camp. Fascinating!

      The Act of Killing & The Look of Silence are two very interesting documentaries about the (CIA-assisted) mass murders in Indonesia of commies and ‘commies.’

      Waltz with Bashir is a very good documentary about an Israeli soldier who tries to reckon with his actions.

      Jodorowsky’s Dune is a documentary about the failed attempt by the prolific LSD user and very magic realist director to make Dune with some of the greatest stars.

      Werner Herzog docs (I really like him):
      – Grizzly man, man lives with bears, man gets eaten, Herzog chastises man and mankind for their hubris
      – Little Dieter needs to fly, about a pilot shot down in Laos
      – Happy People: A Year in the Taiga, trappers living a hard and primitive life in the Russian taiga
      – Lessons of Darkness, less a true documentary than a mesmerizing and bleak (in the best way) meditation on mankind

      The Thin Blue Line, about a man convicted and sentenced to death for a murder he did not commit. The first of the many crime investigations that we now have (Serial, Making a Murderer, etc) and perhaps still the best.

      The Fog of War: Eleven Lessons from the Life of Robert S. McNamara, very interesting

      The Queen of Versailles, family gets rich and loses their humanity, then become poor (or less rich). See it if you get envious of rich people and want to feel better.

      Brother’s Keeper, a very interesting documentary about how a group of rural social outcasts become loved by their community when the state goes after one of them. Very relevant given the recent discussions on tolerance in rural communities, but also very telling on how tribalism and social dynamics work.

      Ken Burns:
      – The Civil War
      – The Vietnam War

      • j1000000 says:

        RE: The King of Kong, Billy Mitchell’s 1 million point score (a record during the recording of KoK) was recently disqualified by Twin Galaxies b/c they said they had new proof he used emulators. He has been on a Twitch/Twitter rampage recently trying to get back his records and trash talk the haters.

        He’s the best. He’s like a living wrestling heel.

        • Chlopodo says:

          “Recently disqualified by Twin Galaxies” is technically correct… but the fuller* story seems to be that the Twin Galaxies organization had been all-but-complicit in the cheating to begin with, and was dragged kicking and screaming into admitting that Mitchell was a fraud only when they absolutely couldn’t deny it any longer.

          [* “Fuller” is a relative term, of course. Who knows what the whole story really is?]

      • nkurz says:

        Thanks for the recommendations. I’ve seen most of these, and will try to seek out the rest.

        About Herzog, though. I’ve soured a lot on Herzog after realizing how loosely he is bound by the truth. He’s a great filmmaker, but even accounting poetic license, I don’t think he should be trusted as a documentarian.

        The most direct place to see this is with “Happy People”, which I thought was really good until I saw the original. The original, you might ask? When I first saw it, I hadn’t realized that Herzog wasn’t actually involved in the filming. In actuality, it was a complete 4-hour series broadcast on Russian television before Herzog was brought on as a “big name” to aid international marketing.

        While there are bound to be details lost in editing a series down to a feature, I was bothered by how many of the interesting “details” were in fact intentional misconstruals and deceptions, designed to improve the narrative flow. Watching them both made clear to me that while Herzog is a great storyteller, he simply doesn’t feel any shame at fabricating details to make a better story. Trust him at your own peril.

        For a hilarious alternative view of Herzog consider Zak Penn’s “Incident at Loch Ness”: The three (!) layers of directors’ commentary should be considered essential for this one. It’s a fake documentary posing as a Herzog film, starring Herzog, and might actually get closer to the truth than some of the “real” Herzog documentaries.

        The original Russian series of Happy People (which is great) is available subtitled on Youtube:

        • Aapje says:

          Yeah, it’s definitely true that Herzog tends to fit his movies around his preferred narratives, rather than have the material lead him. I had concluded that before and should have noted that.

          As such, he is indeed a poor documentarian, but I do like him as a storyteller.

          Thanks for the links!

        • Anon. says:

          He’s always been up-front with it, and at least from his perspective there’s a good reason for it.

          You should bear in mind that almost all my documentaries are feature films in disguise. Because I stylize, I invent, there’s a lot of fantasy in it—not for creating a fraud, but exactly the contrary, to create a deeper form of truth, which is not fact-related. Facts hardly ever give you any truth, and that’s a mistake of cinéma vérité, because they always postulate it as if facts would constitute truth. In that case, my answer is that the phone directory of Manhattan is a book of books. Because it has 4 million entries, and they are all factually correct, but it doesn’t illuminate us. You see, I do things for creating moments that illuminate you as an audience, and the same thing happens with feature films as well.

          Famously the door thing in Little Dieter Needs to Fly is completely made up, but you can see what he’s getting at.

    • Kyle A Johansen says:

      The Shock of the New was my favourite television of all time when I first watched it.

      It is a series of BBC programs about modern art host by Robert Hughes. It isn’t too wacky or avante-garde, but it is quick and quite stuffed. Robert Hughes is great; very knowledgeable, with character and humour, and good voice and presence.

      I think it is on YouTube, or at least was the last time I checked. It is/was the American edit with is almost exactly the same, but the almost makes watching it uncomfortable. If one hasn’t seen the BBC version, then the differences – which are mostly, as far as I can remember – the cutting of musical montages of random people and street scenes – won’t be noticed.

    • herbert herberson says:

      Dark Days, a documentary about people living in abandoned NYC subway tunnels in the mid 90s

    • WashedOut says:

      Ken Burns’ The Vietnam War should be compulsory viewing, for both style and content.

      Searching For Sugarman – doco about an underground/cult folk singer who, without knowing, became a sensation.

      Valley Uprising – on the history of rock climbing in Yosemite

      • johan_larson says:

        Valley Uprising – on the history of rock climbing in Yosemite

        Definitely worth seeing.

    • Tarpitz says:

      They Shall Not Grow Old is unlikely to tell you much concrete that you don’t already know, but it’s an astonishing piece of film-making. Peter Jackson and his team have taken a load of archival footage from the Western Front in WW1 and colourised it and made the speed uniform, adding sound effects where appropriate and using lip-readers and actors to revoice anyone who can be seen speaking in the footage. That may sound like it has the potential to be naff, but the execution is so stunning that in fact it simply makes the Great War real and immediate in a way nothing I’ve seen before does. The narration is entirely in the form of recordings of veterans talking about their war experience in subsequent interviews, which among other things has the virtue of making it clear just how different individual experiences of and attitudes towards the war were. I can’t recommend it enough.

    • Elephant says:

      if it’s still in theaters, Free Solo is excellent, especially on a big screen. (It’s the documentary about the guy who climbed El Capitan, without ropes or harnesses.)

    • Douglas Knight says:

      Exit Through the Gift Shop (dir. Banksy)
      And I second Democritus’s recommendation of Orson Welles’s best film.

  40. Jaskologist says:

    Apropos of the recent post, something I was going to ask anyway:

    Or maybe climate change took over everything, became so important that everything else faded into the background. This is certainly how it feels to me. Whenever I hear about the rainforests nowadays, it’s as a footnote to some global warming story where they add that we should save the rainforest as a carbon sink. Whenever I hear about landfills or recycling today, it’s in the context of trash giving off greenhouse gases.

    Suppose for the sake of argument that government can do just about anything well, but only for 5 things. Beyond that, attention gets divided and it actually becomes actively harmful to the things it is trying to do.

    What 5 things would you choose for your government to spend its attention on?

    (Limit yourself to things governments attempt now. No cheating by having the government focus on perpetual motion machines.)

    • bean says:

      I think the question as specified is too broad. Can I say Defense as one thing? What about Universal Healthcare? What about, say, the FAA? How many things does it do? Or the FBI? Can I say “Fight Crime”, and if so, does that include, say, border enforcement? At some point, this becomes a game of trying to figure out the broadest categories you can for the government to be looking at.

      • Controls Freak says:

        “Army, Navy, Marines, Air Force, and uhhhh, Space Force. Oops, I ran out already.”

    • Plumber says:

      1) The Health of it’s citizens.

      2) The prosperity (a high minimum and a high median) of it’s citizens.

      3) The security of it’s citizens (law enforcement and national defense).

      4) The Post Office.

      5) Public Libraries.

    • Telemythides says:

      – Fight crime.
      – Preserve the environment.
      – Ensure national security.
      – Help the poor.
      – Promote economic growth.

      Would’ve like to include healthcare and education, but there’s no more room and those can probably be handled better privately than any of my five.

    • Anonymous says:

      1. External warfare.
      2. Internal security.
      3. Justice in applying the law.
      4. Stability and sustainability.
      5. Economic prosperity.

    • Erusian says:

      In order of importance:
      1.) The promotion and positioning of its leaders and bureaucrats
      2.) Taxation
      3.) War
      4.) Diplomacy
      5.) Law (courts/police)

      Firstly, a state must invest in making sure its leaders and servants are the best quality of people and incentivized to take the common good and interests of the state above their own. If good leaders are not available, the instructions will be bad or perverted by self-interest. If good bureaucrats are not available, the instructions will be poorly carried out or perverted by self-interest. And good is not merely a moral character: they must be presumed to be selfish and their incentives must be aligned to act in this manner.

      Next, a state without good taxes has no way of doing anything. If the taxes are laid in a destructive or arbitrary manner, people will hate the state and the state will find its initiatives poorly funded.

      Next, a state must be able to defend itself militarily. If it cannot, it exists on the sufferance of its neighbors. Any good that might be done by the state will disappear the moment another state decides to invade.

      Next, the state must maintain diplomatic efforts, both open and covert, to maintain its position and advance its standing. If it does not, it will end up in unnecessary wars or fighting alone when it might have had allies. It will miss economic opportunity and be forced to spend more on its military than is optimal. It will have enemies where it could have had friends.

      Lastly, the state must set up a just system of laws and see to their just enforcement. They must root out corruption to maintain integrity, enforce property rights to grow wealth, resolve disputes to maintain the public good.

    • theredsheep says:

      Am I allowed to be a wiseass and quote five of the six purposes mentioned in the Preamble to the Constitution? It’s interesting to ask myself which of the six I’m willing to ditch. Probably “form a more perfect union,” since I’m not totally sure what that would mean in a modern context.

    • TDB says:

      Digression: if we thought organizations could only do 5 things well at once (which might actually be optimistic), why not have more than one government in the same territory, but with different mandates? Statist Kickstarter?

  41. Plumber says:

    Okay a deeply culture war topic:

    I say that Fritz Leiber’s Fafhrd and the Gray Mouser series is more “in the bones” of Dungeons & Dragons than J.R.R. Tolkien’s Lord of the Rings

    Any dispute?

    • Le Maistre Chat says:

      How so?
      Neither of them is an elf, dwarf or halfling.
      They never fight goblins/orcs or trolls.
      Lord of the Rings has a wizard with 3rd level spells and a Cleric who turns (Ring)wraiths and commands some sort of undead (yeah, Aragorn is no D&D Ranger: I said it).

      • Plumber says:

        @Le Maistre Chat

        To quote Gygax:
        Dungeons & Dragons, 
        Book 1: 
        Men & Magic

        “These rules are strictly fantasy. Those wargamers who lack imagination, those who don’t care for Burroughs’ Martian adventures where John Carter is groping through black pits, who feel no thrill upon reading Howard’s Conan saga, who do not enjoy the de Camp & Pratt fantasies or Fritz Leiber’s Fafhrd and the Gray Mouser pitting their swords against evil sorceries will not be likely to find Dungeons & Dragons to their taste. But those whose imaginations know no bounds will find that these rules are the answer to their prayers. With this last bit of advice we invite you to read on and enjoy a “world” where the fantastic is fact and magic really works!”

        E. Gary Gygax 
        Tactical Studies Rules Editor 
        1 November 1973 
        Lake Geneva, Wisconsin

        and from the 1979 Dungeon Masters Guide

        “The most immediate influences upon AD&D were probably de Camp & Pratt, REH, Fritz Leiber, Jack Vance, HPL, and A. Merritt”

        16 May 1979

        The Dragon even published a Fafhrd and The Grey Mouser story in 1977!

        Besides, Aragorn and Frodo are mopping ponces, but Fafhrd and the Grey Mouser frequently meet at the Silver Eel and peruse gold to spend there! 

        Not once do the Fellowship grab gold pieces! 

        Gold for XP is the essence of D&D (dropping that mechanic ripped the soul out of the game!).

        Dungeons & Dragons was (and should be) Sword and Sorcery not “Epic Fantasy” world saving lameness!

        Look at the early modules: The Village of Hommlet listed how much gold each shop keeper had!



        Dying Earth?


        Even The Hobbit?


        But The Lord of the Rings?


        Proper D&D was and should be about looting, world-saving is antithetical to spirt of the game!

      • Deiseach says:

        High Fantasy versus Low Fantasy. Fafhrd and the Mouser aren’t particularly even heroes for hire, they go on adventures because that’s how they get money, or are tricked/arm-twisted into it by someone, or there’s a particular object or end they wish to achieve. It’s much closer to the “group of adventurers hang out/meet up in local tavern, hear about haunted caves/treasure dungeon/evil witch king in the hills and set out on a quest for XP and gold” version of a campaign. It’s also slightly grittier (if you can use that term) in tone, a bit more cynical and “beer and busty barmaids”, and the magic system in it is a lot more amenable to the “learn to be a wizard and have an arsenal of spells to cast” than the way Tolkien handles magic.

        • Nornagest says:

          the way Tolkien handles magic

          Step 1: be an angel. There is no Step 2.

          • Evan Þ says:

            Alternate Step 1: Be an elf.

            Step 2: Protest that you don’t know what the word “magic” means.

          • Le Maistre Chat says:

            Oh man, yes. It darkly amuses me when one of them (Galadriel?) protests “I don’t know what the hobbit word ‘magic’ means. You seem to use the same word for what we do and what the Enemy does.”

        • Plumber says:


          You totally get it!

    • John Schilling says:

      Yeah, I’ll dispute that. You can’t model Fahfrd or the Grey Mouser as D&D characters without multiclassing, which was a clumsy after-the-fact graft onto core D&D mechanics. But the entire Fellowship is a near-perfect match for Basic Edition character types, except that Gandalf sometimes wields a sword. Note that in the original edition, “hobbit”, er “halfling”, was a character class, one that appears to perfectly match Bilbo and Frodo while having little precedent in mythology and none whatsoever in Sword and Sorcery fiction. Including both Elves and Dwarves, neither of them tiny and with a racial animus between them, matches Tolkien very well, mythology poorly and only in the parts that Tolkien cribbed from, and S&S not at all. Also, as LMC notes, the list of monsters.

      When it came time to put flesh on those bones, and later when it came time to take up arms against the Tolkien estate’s lawyers, yes, they started mining Lieber et al for material. But the bones are deeply Tolkien.

      • Brett says:

        The spin-off stuff was even less subtle. I remember thinking, “How did they get away with calling the Dwarvish mine taken over by a monster they dug up ‘Moria’?” when I read one of those Salvatore books ages ago.

    • Nornagest says:

      D&D treats Tolkien more as something to mine for mooks and character concepts than something to model for gameplay purposes, but I think that’s true mostly by default. Lord of the Rings is fundamentally a story about ethics, and that doesn’t play well at the table. The Lankhmar stories* are about dashing heroics, and that does.

      (*) Though honestly all I’ve read of Leiber is Swords of Lankhmar and a couple of short stories. I’m much more familiar with some of the other source material: I’ve read most of the Conan stories and all the Dying Earth stories, for example.

      • Le Maistre Chat says:

        Lord of the Rings is fundamentally a story about ethics, and that doesn’t play well at the table.

        Haha, yes.

        I’m much more familiar with some of the other source material: I’ve read most of the Conan stories and all the Dying Earth stories, for example.

        Same, I think. I’ve got a Dying Earth omnibus and one of all the Howard Conan stories. I’ve also read the de Camp/Carter stories out of some of the Frazetta-cover paperbacks, which is how Gygax and friends would have known Conan. Turns out some iconic D&D monsters like the remorhaz are swiped from those two, while many Howard monsters (extraterrestrial elephant wizards, winged apes, etc) went unused.

        • Nornagest says:

          I read more of the Robert Jordan-era pastiches, which were more easily available when I was a kid, than the Carter/de Camp ones. It’s striking, reading them alongside Howard, just how good a writer Howard was, and how effortless he makes it look. Jordan is no slouch, but he gets repetitive. Howard’s formula is basically the same (though he does depart from it more often), but he makes it feel fresh every time, even when he’s recycling not just a formula but the whole plot, as in Black Colossus vs. half of The Hour of the Dragon, or The Devil in Iron vs. any of his other lost-city-with-mystic-guardian stories. And he’s clearly very well-versed in history, even if it’s the 1920s understanding of it with a bunch of fringe esoterica thrown in because it plots well.

          (It’s also striking how, uh, kinky the Jordan ones are. The books he’s more famous for have their share of fetishy material, but his Conan work is about two steps from Gor.)

          • Le Maistre Chat says:

            how effortless [Howard] makes it look. Jordan is no slouch, but he gets repetitive. Howard’s formula is basically the same (though he does depart from it more often), but he makes it feel fresh every time, even when he’s recycling not just a formula but the whole plot, as in Black Colossus vs. half of The Hour of the Dragon, or The Devil in Iron vs. any of his other lost-city-with-mystic-guardian stories.

            Heh, yes. He was writing formulaicly for a market (Weird Tales which his pen pal Lovecraft resented: Conan stories often made the cover because Howard slipped in a scene to appeal to house artist Margaret Brundage/Bondage), and it’s the mark of superiority that he was able to make it feel fresh each time.

            And he’s clearly very well-versed in history, even if it’s the 1920s understanding of it with a bunch of fringe esoterica thrown in because it plots well.

            Conan zips around from proto-history (eg Queen of the Black Coast, where he has a proto-Semitic paramour) to 16th century Spanish pirates ‘n’ Indians (The Black Stranger, I think?), in what I assume was a shorthand for “I want to write historical adventures like Harold Lamb, but using the same protagonist every time lets me write faster.”
            How close that fringe esoterica was to the mainstream of the time is an interesting question.

            (It’s also striking how, uh, kinky the Jordan ones are. The books he’s more famous for have their share of fetishy material, but his Conan work is about two steps from Gor.)

            Wait, what?

          • Nornagest says:

            Wait, what?

            You know the obligatory Margaret Brundage scene? Think that, but at a rate of about one a chapter, and some mind-control stuff thrown in for flavor.

          • Nornagest says:

            Conan zips around from proto-history to 16th century Spanish pirates ‘n’ Indians

            Oh, yeah, the Hyborean Age clearly isn’t trying to rigorously model any one period or tech level; even leaving all the piracy stuff aside, you get scenes with Late Middle Ages knights clashing against Roman-era shield-wall infantry and both against Early Iron Age skirmishers. Throw in the odd antediluvian wizard that some schmuck woke up, and it gets very confusing. But he understands the individual elements well enough to give them the right tactics and equipment and fit correctly into their parent societies.

    • Walter says:

      Nowadays I’d say you are correct. Modern adventures are much more akin to Fafhrd’s shenanigans than they are to Bilbo’s.

      But I think the inspiration for D&D was much more Tolkien than anything else, and thus LOTR would be more ‘in the bones’.

      • Protagoras says:

        Yeah, Tolkien didn’t write about murder hobos.

        • sandoratthezoo says:

          Early D&D was very murder-hobo-oriented. Gygax, for example, famously didn’t care much about characters or plots and was very oriented towards plopping PCs into hazardous environments and asking them to problem solve getting treasure away from monsters who would murder them in a fair fight.

    • sandoratthezoo says:

      I haven’t read Fafhrd and the Gray Mouser. But the claim that these stories are more in the bones of D&D was fairly common currency on and .misc back in the 90’s, so there are people who agree with you.

      I think that the general consensus at the time was that there was a very Tolkien-esque gloss of paint on D&D, with the races and classes and so forth, but the story beats, the kinds of things people did, the attitudes that the game presumed were more influenced by F&GM and Conan than LotR.

      • Le Maistre Chat says:

        There’s a guy named Mike Mornard, one of Gygax’s original players, around on RPG forums. He comes across as saying that the setting was very Tolkien with a sprinkling from every fantasy/horror book and movie Gygax liked, but the PCs were amoral at best. He was allowed to play characters like a baby Balrog, starting with 1 hit die out of the 9-ish the standard Balrog had, with an ability to ignite his body 10% of the time that would advance to at-will as he grew up.
        So baby Balrog was adventuring in the megadungeon, and he gets the idea that he should have orc henchmen. He tries to impress a random room full of orcs by igniting himself… the roll fails. So he excuses himself, douses himself with flammable oil in the next room, walks back in, and sets himself on fire with a torch. Gygax is so impressed, he hands out a hefty bonus to the roll to persuade the orcs again.

        • Plumber says:

          Le Maistre Chat

          “There’s a guy named Mike Mornard, one of Gygax’s original players…”

          Indeed there is, check out his post in this thread Originally Posted by Mike Mornard, who played in both Arneson’s & Gygax’s campaigns!

          …is there anybody who DIDN’T know Gary hated Hobbits?

          “Hey, Gary, have you read Lord of the Rings?”


          Direct quote

          D&D co-creator, Arneson has the inspiration as:

          “Some months after Mr. Wesley left, a local TV station had on several old monster movies, which I watched while eating popcorn and reading old Conan novels. It was then that Blackmoor
          Dungeon was first conceived.”

          Wargaming #4 1978, p47

          I’ve the distinct impression that the main original influences on Arneson & Gygax were those explicitly mentioned in the Dungeons & Dragons preamble (which I quoted upthread), and it was players who first brought in Tolkien.Yes Tolkien had Moria, but I’d cite Howard’s “mega-Dungeon in Red Nails and Leiber’s Quarmall from the Fafhrd & Grey Mouser story The Lords of Quarmall (started by Fischer and finished by Leiber). By the time of the story, much of it was uninhabited, but it clearly had many levels–it was a whole kingdom underground (complete with slaves running on treadmills to keep ventilation fans turning), as bigger influences on early D&D.

      • Jiro says:

        The actual answer is “both Lankhmar and Tolkien had a lot of influence on Dungeons and Dragons. Which one counts as “the” influence depends on what parts of D&D you’re emphasizing.

    • dndnrsn says:

      I’d argue that both are about equal influences. It’s one-third wargaming, one third pulp, one third Tolkein. Some of the notes are pulp, some are Tolkein, filtered through a wargame taken to the smallest possible scale.

      What’s “in the bones” though? What they were trying to do, what they ended up producing, ? If the originators had meant to produce something that exactly mimics Leiber, or something that exactly mimics Tolkein, they didn’t. What they later said they were aiming at varies based on how popular Tolkein was at a given moment in time and how litigious his estate looked (as John notes).

    • John Schilling says:

      One aspect of D&D that I think breaks any analogy with its literary sources, and in a big way, is having magic-users as PCs. The Grey Mouser had a bit of hedge-magic, IIRC none of it useful in combat, and even that was a genre exception. Swords and Sorcery was all about wizards as adversaries for mundane heroes to struggle against, or as mysterious dispensers of plot coupons, not as heroic adventurers. Tolkien is a better fit, because Gandalf, but from a role-playing perspective Gandalf really works best as an NPC ally. But from the start, D&D tempted players with the chance to be the Guy What Can Cast Fireball.

      Which requires a more mechanical approach to magic than any literary antecedent save Vance, and encourages direct combat spellcasting that is generally rare in pre-D&D fantasy fiction, and needed a fundamentally new character archetype for which there were few “role models” in fiction or mythology. And, of course, was prone to creating linear-fighter-quadratic-wizard game imbalance.

      It’s a somewhat closer match to Tolkien than to Sword and Sorcery, again because Gandalf, but it is a major divergence from any source material and it has been “in the bones” of D&D from the start.

      Which raises the question, are there any RPGs that follow the Sword and Sorcery tradition of having magic be real and powerful but largely limited to NPCs? Call of Cthulhu sort of does this, with SAN costs limiting heroic PCs to maybe a few dozen spells cast per career and so discouraging specialist spellcasters, but that’s yet another literary tradition only tenuously connected to S&S.

      • Hoopyfreud says:

        that’s yet another literary tradition only tenuously connected to S&S.

        Not sure how true that is, given the close relationship between Lovecraft himself and Howard himself. I’d say weird fiction is generally much closer to S&S than either is to high fantasy. Runequest, by the way, follows this structure pretty well.

        • Plumber says:


          RuneQuest is a great game, and most every complaint/wishlist of what I’ve seen those who say “I wish D&D was more….” lately was answered by Melee/Wizard/In The Labyrinth/The Fantasy Trip 1977 to 1980, RuneQuest in 1978, and Champions in 1981, and folks who describe D&D as synonymous with “D20/3e/3.5” really grind my gears!

      • Plumber says:

        @John Schilling,

        You have a very good point there, the closest literary antecedents that I can think of with D&D like Magic-Users are Vance’s Dying Earth and Niven’s The Magic Goes Away stories (which I highly recommend).

        I actually like the “magic systems” of Call of Cthullu, Pendragon and Stormbringer better than most versions of D&D, and for D&D I like keeping PC’s hands off the above 6th level spells introduced in the Greyhawk supplement, and just plain limiting “levelling up” to the rate articulated by Gygax in the “D&D is only as good as the DM article (if that link doesn’t work I also quoted it here)

        • John Schilling says:

          I had forgotten Pendragon, probably because I died halfway through the only session I ever played (hint: if unhorsed by a still-mounted knight, do not try to continue the fight). But it’s a perfect example of what I am talking about. Magic exists, it is quite powerful and a Big Deal in a suitably mythic way, and far too cumbersome to be used by PCs. If someone were to apply a system like that to a Swords & Sorcery or a High Fantasy RPG, that would be a great improvement.

          Alas, it is far too late to pull convenient magic out of Dungeons and Dragons. But I’ll put that on my list of things to bring up with Gygax and Arenson if I ever time-travel to 1970s Wisconsin. Right up there with “hit points” and “character classes”.

          • Plumber says:

            @John Schilling

            “…..Magic exists, it is quite powerful and a Big Deal in a suitably mythic way, and far too cumbersome to be used by PCs. If someone were to apply a system like that to a Swords & Sorcery or a High Fantasy RPG, that would be a great improvement.

            Alas, it is far too late to pull convenient magic out of Dungeons and Dragons. But I’ll put that on my list of things to bring up with Gygax and Arenson if I ever time-travel to 1970s Wisconsin. Right up there with “hit points” and “character classes””

            I like the cut of your jib!

            From Greyhawk through “3.5” D&D has made Magic-Users progressively more powerful, WotC “5e” (the 1991 Rules Cyclopedia was the first “fifth edition of Dungeons & Dragons“) does reduce “caster supremacy” somewhat, but not enough!

            Okay, the Gray Mouser did cast a spell in “Unholy Grail”, but imagine instead a party of Thieves and Warriors trying to work unearthly magic from some old scroll, with results both greater and more dangerous than what they expect. 

            Now that’s genre!

          • Nornagest says:

            Lamentations of the Flame Princess is probably my least favorite of the major retroclones, but it does have a bunch of Weird Fiction-style rules for the unintended consequences of tampering with reality that’d mesh well with that sort of concept. (Hint: don’t cast any of the summon spells unless you’re desperate enough to risk something worse than a TPK.)

          • Plumber says:


            Lamentations of the Flame Princess is probably my least favorite of the major retroclones…..”

            Lamentations of the Flame Princess, and Labyrinth Lord are the only “retro-clones” I own, and I kinda feel I wasted my money with Labyrinth Lord as it is such a very close clone that it just didn’t add much to the rules I already owned, about the only use I could see would be to hand out to players who didn’t have the old rules.

            The early modern/17th century with Dwarves, Elves, and Halflings default setting of Lamentations doesn’t appeal to me that much, but at least it was less than a complete clone so it added something.

            What do you recommend?

          • Nornagest says:

            I like Adventurer Conqueror King, but it might be less compelling to people who haven’t spent as much time thinking about the price of wool in 9th-century Icelandic villages as I have. Still, even if you ignore all that stuff, I think it removes enough of BECMI’s warts to be worth the money, and there’s something to be said for a system that actually takes the old-school follower/stronghold rules seriously.

            You’re right about Labyrinth Lord: it, and a whole bunch of similar titles, basically exist to duplicate old-school editions for people that weren’t around for them the first time. Since used copies can be pretty hard to get ahold of if you aren’t and don’t know any gamers over the age of 40, I think that’s a legit niche to fill, but I can also see how you wouldn’t find it useful.

          • Plumber says:


            “I like Adventurer Conqueror King…”

            Thank you very much for the tip!

            “…it might be less compelling to people who haven’t spent as much time thinking about the price of wool in 9th-century Icelandic villages ….”

            I thought Chaosium’s Mythic Iceland was quite good, you may want to check it out.

            “…gamers over the age of 40…”

            Well I’m a 50 years old gamer, and as much as I dislike the “build” culture of the youngsters and 3.5/Pathfinder which they favor, my generation was worse.

            We tasted the promised land in the 1970’s and ’80’s but by the early ’90’s, no one (that I could find) would play D&D anymore, which would be fine if the replacements were AWESOME games like Pendragon or Traveller, but NO!, it was just the bleak horrible landscape of the modernday-ish claptrap of Champions, Cyberpunk, and most (and worst) of all Vampire! 😠

            I walked away from the hobby in ’92 rather then play open-my-own-door-and-go-outside settings like that.

            If I wanted to experience a modern day setting, I didn’t need to game it!

            I don’t like default Epic/High Fantasy “5e” D&D, but at least it can be tweaked into something resembling Sword & Sorcery settings (more by subtraction than addition), but there was just no saving superheroes, “Dark Future”, and “World of Darkness”!

          • bean says:

            I like Adventurer Conqueror King, but it might be less compelling to people who haven’t spent as much time thinking about the price of wool in 9th-century Icelandic villages as I have.

            Wait, what? What has David Friedman been up to now?

          • MrApophenia says:

            It might also be worth checking out Dungeon Crawl Classics, which is not a retro-clone but rather tries to capture the feel of “How people imagine classic D&D,” complete with art that looks like it should be doodled in a study hall notebook or airbrushed on a van.

            A lot of effort is spent making magic seem exceptionally rare and dangerous and unique to a particular wizard, instead of a common feature of the world.

          • Le Maistre Chat says:


            Wait, what? What has David Friedman been up to now?

            Maybe it should have been titled Anarchist, Capitalist, King.

          • Plumber says:


            “It might also be worth checking out Dungeon Crawl Classics…”


            I flipped through the pages of DCC a few years book and was amused by the art, which I realized was supposed to be both a homage and a parody, but it really did remind me of game art from the ’70’s and ’80’s, and no not from TSR, instead from the The Arduin Grimoires by Dave Hargrave, which were third-party supplements that me first DM used all with TSR’s “LBB’s” the AD&D Monster Manual, and another third-party book, Chaosium’s All the World’s Monsters, until that dark day he decided we should play Villains & Vigilantes instead.

          • Nornagest says:

            Wait, what? What has David Friedman been up to now?

            No, no, I wasn’t reading his book. I was thinking about the price of wool in Iceland for entirely different reasons (the same ones that led to me picking up this handle).

            It’s a weird coincidence that there are two people on this board who’ve been obsessed with Medieval Icelandic economics at some point in their lives, but it is just a coincidence.

          • Jiro says:

            If I wanted to experience a modern day setting, I didn’t need to game it!

            Did you also say that people who read superhero comic books were just experiencing a modern setting? I mean, there are some people who criticize superheroes, but I’ve never heard of them being criticized on that basis.

        • dndnrsn says:


          If you like CoC and you haven’t checked out the new-ish Delta Green give that a look. Best Cthulhu game I’ve seen. Its magic system is even more useless for PCs than CoC’s, because to cast a spell you have to either roll your inverted SAN (so, the crazier you go, the better you get at casting spells) or sacrifice POW.

          (A fantasy game run in a system like that would be interesting – high lethality, relatively low-powered PCs, magic as either unavailable or sanity-wrecking)

          • Plumber says:


            That sounds a lot like the same magic system used in the 1981 version of Call of Cthullu (I never got any subsequent ones), I found CoC incredibly easy to GM (or “Keeper”) and the link between madness and magic seemed very thematic to me.

      • Le Maistre Chat says:

        Tolkien is a better fit, because Gandalf, but from a role-playing perspective Gandalf really works best as an NPC ally.

        Nah, I disagree. As a D&D character, Gandalf is quite well-balanced with Aragorn.
        The Fellowship really looks like an Old School open table. Gandalf and Aragorn start at Level 5, the DM recruits two new players and make them start at Level 1, and then his old players Legolas, Gimli and Boromir wandering back, with the last rage-quitting over how the DM is treating magic items.

        • John Schilling says:

          Not buying it. A D&D player with a fifth-level wizard, completely failing to fireball even a single ringwraith or oliphaunt at the Pelennor Fields? Next you’ll be telling me Aragorn didn’t search the room for treasure after subjugating the Army of the Dead.

          You can model most of Gandalf’s abilities as a mid-level D&D wizard, but it’s much harder to model his actions as any sort of PC. He’s physically absent for most of the campaign. When he is present, he mostly serves as a technical and political adviser. He casts a few utility spells at need. Swings a sword at some Orcs, but never so much as magic-missiles or sleeps any of them. And then the one time the rest of the party wakes up a CR-inappropriate Balrog, he becomes a badass combatant for that engagement only, then disappears until they need someone to cast dispel magic on Theoden for the plot to move forward.

          That’s NPC-ally behavior. And as an NPC ally, he wouldn’t have needed a magic system, because he’d always have exactly the abilities the DM needed him to have. Allowing PCs to play wizards, and giving them a list of magical abilities they can exercise without the DM’s permission, is game-changing if not game-breaking.

          • cassander says:

            He does cast lightning bolt once during the hobbit, but in general I agree.

          • Le Maistre Chat says:

            He also casts lightning bolt at the Battle of Pelennor Fields. Decades earlier in The Hobbit, he was already a 5th level Mage, since he killed a few orcs with a lightning bolt when they kidnapped Thorin & company. So one important difference between Middle Earth and D&D is the time frame to level up!
            You’re correct that fighting a Balrog all day all the way down and back up a mountain when it’s high enough CR to require “Run, you fools!” is NPC behavior.

          • Jiro says:

            A D&D player with a fifth-level wizard, completely failing to fireball even a single ringwraith or oliphaunt at the Pelennor Fields?

            He’s a player with a fifth level wizard who never got the fireball spell (which was more possible in D&D back then, since you didn’t get to add new spells when going up a level).

            Also, fireball in general is something that D&D didn’t take from anywhere. It’s more like a modern hand grenade than it is like anything that wizards typically did in fantasy novels.

          • Creutzer says:

            Gandalf does cast some fire magic to fend off a wolf attack before they ascend to Caradhras, in the book. I don’t have it at hand to check whether this takes a form that in any way resembles a fireball.

          • John Schilling says:

            After Caradharas and before Moria, actually. And he just scares off the wolves by starting a brushfire in some scraggly (and not notably damp) trees, using the same magic words he used to start a campfire on Caradharas. Gandalf is explicitly limited to igniting the already-flammable, so this is more utility fire magic creatively applied, than battle magic. The closest D&D model would be 1st-level “Burning Hands”, which of course is battle magic applicable to ugly bags of mostly water, because D&D.

            And now that I look, why is there not a cantrip for “I just need to light a candle or start a campfire”? That would seem like Wizardry 101, incendio and flickum bicus and all that. Possibly the single most obvious bit of utility magic, the one that would turn a Paleolithic tribal shaman into a literal Promethean demigod, and it’s beneath D&D’s notice.

          • Montfort says:


            Arguably that’s in prestidigitation’s remit. Actually, it’s explicitly in prestidigitation’s remit as later clarified (or revised) in Tome and Blood, but I think the original description leaves enough room for it. I do consider it an oversight to omit it from the original spell description if they intended it to work that way from the beginning, though.

            With the system they have, it’s probably best to leave most of the mostly-flavor very minor utility effects in one spell like that to encourage players to actually prepare it. And even if things like lighting fires and changing how food tastes were separate spells, you wouldn’t really want to waste time tracking their use in much detail (or at least I wouldn’t, even though I’m rather sympathetic to the logistics & dragons school of play).

          • John Schilling says:

            3.5e rules as written say that prestidigitation has “severe limitations” and that it can “chill or warm” non-living material. Even with maximal charity, I cannot read that any other way than that “warm” marks a limiting boundary and that incendiary heat is Right Out. If they’ve changed that since, good for them, but what took so long? Thirty years of playtesting should have been plenty of time to notice that having to burn a 1st-level spell slot just to light a candle was Not Right.

          • pansnarrans says:

            Having Magic Cigarette Lighter at range as part of Prestidigitation would seem a little overpowered as what can light a campfire can also light that oil that just got spilled over the enemy archers.

            But if it’s a touch spell then it seems fine, as it’s the same as striking a match. And clearly the first spell you’d learn at wizard school to show off to your mates.

          • johan_larson says:

            AD&D V1 didn’t have cantrips, but a level 1 Magic-User could light a fire using Burning Hands.

            “Inflammable materials touched by the fire will burn, i.e. cloth, paper, parchment, thin wood, etc.”

            It’s overkill, but it should work.

          • John Schilling says:

            But if it’s a touch spell then it seems fine, as it’s the same as striking a match. And clearly the first spell you’d learn at wizard school to show off to your mates.

            Right, the moment you come up with the concept of cantrips, having one that’s roughly an enhanced zippo should be no-brainer obvious, and if it’s bundled with other simple utilities even better. Having your simple utility bundle all but explicitly say “no more than warming, and really this is just for useless party tricks”, is getting Gandalfian or folklorian magic pretty much Wrong, as is saying that wizards have no fire magic of less than Goblin-incinerating cost and effect.

            ETA: Burning hands is given as a 15′ cone effect, which is going to be problematic for just a campfire and right out for a candle. Before the splatbook revision of prestidigitation, I’m not sure there’s a good way for a wizard to start a small fire.

          • johan_larson says:

            The earliest mention of cantrips I can find in a D&D context is in Dragon Magazine issues 59-60 (1982), where EGG himself lists a collection of them. This list includes Firefinger, which is ideally suited to lighting fires.

            The firefinger cantrip enables the caster to cause a jet of flame up to one-half
            foot in length to shoot forth from his or
            her finger. The flame is very hot and will
            ignite combustible materials such as
            parchment, twigs, kindling, and the like
            without difficulty, providing the materials are relatively dry. The flame persists
            for up to 1 segment. The caster speaks a
            word of power over elemental fire (ronson, zip-po, or the much revered word,
            dun-hill), extends the forefinger, and
            makes a downward or sideways motion
            with the thumb.


          • John Schilling says:

            I like that list of cantrips, parts of it at least, and I like the idea of bundling them into a 0-level prestidigitation spell that doesn’t require the wizard to predesignate the night before. But it looks like it took them eighteen years to migrate that from the pages of a magazine to the core rules, and another decade to fix the bit where they accidentally excluded the most useful effect on the list.

            So, Gygax thought of it, but didn’t bother to properly implement it. Probably because he thought of it as a party trick rather than a useful tool. I’m going to guess that Gary was never a Boy Scout.

          • Having Magic Cigarette Lighter at range

            Means a simple ranged kill spell–consider the effect of a cigarette lighter in the brain.

            I have had no involvement with D&D for a very long time but my novel Salamander has magic, magic is very weak, and the above point is why, even though a fire mage is more like a match than a blow torch, one is still very dangerous.

          • Montfort says:

            [edited to reflect info from Johan Larson’s comment]

            what took so long?

            Tome and Blood (2001) was released before 3.5 (2003), as an expansion to 3e (2000). The flame effect never made the jump to its 3.5 counterpart, Complete Arcana (2004), nor, as you noted, was it adopted into the base spell description. Whether that’s an oversight or a deliberate walk back I don’t know. But 2001 isn’t so long after cantrips (as a class of spell) came to D&D core.

            As for cantrip as a first-level spell of it’s own (2e, 1989), the original description is extremely vague, noting only the spell can’t do damage, distract spellcasters, or make useful tools. Fire is an open question. An optional replacement of the spell was offered in Dragon 221 (1995, “Little Wishes”) because cantrip was not as popular as other first level spells. In addition to changing it from a spell to a proficiency, they defined a number of specific spell effects by school of magic (available to all, better if cast in specialty). These include “fiery cantrips” that “create sparks, warmth, flickering lights (no stronger than torch-light), and can ignite dry undergrowth, paper, campfires or thin sticks.” This proficiency was presumably built from the cantrip system Johan Larson identifies above, which (modified) made its way into core 3+.

            To me it seems like igniting kindling was on the table from the start, and later went away. (And then it comes back in 4e and 5e, I believe, but I only checked the online wikis for those editions.)

      • Skivverus says:

        Vaguely-connected musing: how much did the ability to do ridiculous over-the-top things in the setting (magical and otherwise) contribute to D&D’s popularity? Err on the side of too much realism, and the game becomes boring.

        • Plumber says:


          “….how much did the ability to do ridiculous over-the-top things in the setting (magical and otherwise) contribute to D&D’s popularity? Err on the side of too much realism, and the game becomes boring”

          Judging by the games that largely replaced D&D in the 1990’s, probably a lot.

          D&D was dying, and then Wizards of the Coast changed and saved the game by making it so that the PC’s got more comic book faster.

          My preference would instead be something more likeThe Seven Samurai with a sorcerer or witch to defeat (or something like The 13th Warrior/Eaters of the Dead), but I think that mine is a minority taste.

      • FormerRanger says:

        One aspect of D&D that I think breaks any analogy with its literary sources, and in a big way, is having magic-users as PCs.

        The major source for D&D’s magic was Jack Vance’s Dying Earth series, mentioned above but not given much love. It had almost every feature in D&D magic, especially the idea that you had to re-memorize your spells after use. The stories collected in The Dying Earth often had “PC” magic users who thought carefully about what spells to memorize. In the later Cugel the Clever stories collected in Eyes of the Overworld, Cugel was a thief who tried to learn and use spells and often suffered from his inability to do so (he was a poor thief and worse magic user). Different planes of existence were a major feature of many of the stories as well. Vance’s monsters (deodands, pelgranes, sandestins, grues, etc.) don’t appear to have made it into D&D, though maybe some of the later editions have them? (I played the original box set D&D and later the classic book set.)

        Gygax would steal from anyone, bless his heart. (That’s meant as a compliment.)

        • John Schilling says:

          Dying Earth was the source for D&D’s magic system, transparently so, but I don’t think it is the reason there is magic system in D&D in the first place. I can’t much of Vance but the magic system in early D&D, nor have I seen any sign of anyone trying to run a Vance-inspired campaign.

          I believe D&D has a magic system because players wanted to role-play Gandalf from the start, and Tolkien is absolutely no help in providing the mechanics you need to do that. Gandalf is exactly as powerful as Tolkien needs him to be at any given moment; a PC needs clearly-defined abilities that can be used as reliably and predictably as a fighter’s crossbow and without negotiating with the DM every time. And if you’re limited to military history and pre-1974 fantasy fiction, about the only place you can get anything like that is Vance.

          • Nornagest says:

            Some of the early adventures are pretty Vancian in tone — I could see Cugel the Clever getting scammed into exploring the crashed spaceship in Expedition to the Barrier Peaks. A lot of magic items are lifted from Vance, most obviously the robe of eyes. Spells too, e.g. prismatic spray, although that might fall under “magic system”. And Gary Gygax’s prose style is closer to Vance’s than to any of the other authors usually cited as influences. But Vance’s work doesn’t seem to have left much of a mark on the bestiary, for whatever reason.

        • Jiro says:

          D&D did have grues, and “demodands”, which do seem to be named after deodands, . (Of course grues were also from Zork.)