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

Open Thread 91.25

This is the twice-weekly hidden open thread. Post about anything you want, ask random questions, whatever. You can also talk at the SSC subreddit or the SSC Discord server.

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

489 Responses to Open Thread 91.25

  1. Anonymous says:

    I wonder – just what percentage of Klingon society is the martial aristocracy that gets 99% of the airtime?

    • Well... says:

      Not sure how serious I can be about this, but to me the Klingons always seemed like a mashup of Samurais, Sikhs, and possibly a distillation of the more warlike American Indian tribes. This means the martial elements run throughout Klingon society from top to bottom.

      Still, it is funny to think about the other 99% of Klingons who aren’t “active duty” aboard starships but instead are back on Planet Klingon or whatever, raising the kids, cleaning the toilets, devising new recipes for gagh, etc. Maybe they have inter-tribal hostilities they can fight about, but you know they can’t all be seeing action. The Klingon professors/engineers who designed and built the Birds of Prey couldn’t have been pulled away from their computers (or sharp metal slide rules?) every few weeks to go fight in some war or else the damn things would never have been built.

      • Randy M says:

        Maybe the way Klingon computers work is very antagonistic. You have to really slam the buttons, the UI insults you when you type incorrectly, you have to verbally threaten it to get any output. Likewise, their farming is done via aggressively bullying the stubborn and of course dangerous beasts of burden. Obviously most of the culinary breakthroughs come in the way to most gloriously vanquish the prey, and in preparation developing new dishes is seen as a martial struggle between the chef and consumer, and though it is bad form if it ends in death, the chef is likely severely beaten by serving a dish that insults the eater by not assaulting his taste buds in an extreme way.
        And, given that, cleaning the toilets is obviously a feat of some renown even without any artificial danger.

        • Well... says:

          ^ Scott probably won’t nominate this for comment of the week but I just want to let you know it deserves to be.

        • James C says:

          Maybe the way Klingon computers work is very antagonistic. You have to really slam the buttons, the UI insults you when you type incorrectly, you have to verbally threaten it to get any output.

          Rather like Windows then?

        • Conrad Honcho says:

          And if you think human holy wars over vi vs. emacs are rough, at least your opponent isn’t wielding a bat’leth.

        • actinide meta says:

          I think there was a whole list of Klingon programmer aphorisms that went around by e-mail in, like, the 90s.

          Klingon software is not released. It escapes.

      • John Schilling says:

        Plausibly most of that work is done by slaves, not necessarily Klingon. That would conflict somewhat with the image that TNG-through-Reboot Trek, at least, was trying to sell us, but I’m trying to think if we ever saw enough of their culture to really know one way or another. TNG did have one episode with a visiting Klingon scientist who seemed to be of low status but not a slave at least. And there were a few Klingon lawyers who were respected as such, but then they did seem to have an adversarial system of justice and even without literal trial-by-combat that might be seen as sufficiently warrior-ish.

      • Incurian says:

        (or sharp metal slide rules?)

        This is not a slide rule, but it is Klingon AF: Range and Deflection Protractor. You can’t tell from the picture, but it’s huge.

        • Andrew Hunter says:

          Unless the nickname is “mek’leth”, which isn’t quite right anyway, I can’t guess what the “obvious” nickname is.

          • Nornagest says:

            I got nothing. Looks like a can opener, though, sized for whole canned sheep or something.

          • Well... says:

            Given the demographics of who needs to “determine the direction towards which artillery must be aligned to hit targets”, I’d guess the nickname has something to do with a penis.

          • Incurian says:

            They are nerdy enough to notice the resemblance but not enough to know the name. They call it a Klingon Battle Axe. They’re pretty easy to use unless CPT Applebaum taught you how to use one, in which case you’ll have to figure it out for yourself over a couple months. Fuck that guy.

    • Evan Þ says:

      I heard the theory once that Klingons are a post-scarcity society that’ve dedicated their resources to letting everyone become part of the martial aristocracy.

      • cassander says:

        adjust that to “almost everyone” and I think that’s a fantastic answer.

      • Deiseach says:

        Klingons are a post-scarcity society

        Huge divergence from TOS then, where they were portrayed as being militarily aggressive and conquering new territories not because they were fanatical killers and warriors but because their home system was very resource-poor and they adopted a “stealing bread to feed your hungry family is no crime” approach to conquering better off worlds.

        From the transcript of the Day of the Dove episode (and interestingly, the Klingons on their side have been told that the Federation mistreats and kills its ‘prisoners of war’):

        MARA: We have always fought. We must. We are hunters, Captain, tracking and taking what we need. There are poor planets in the Klingon systems, we must push outward if we are to survive.

        • Jiro says:

          TOS also did not have the whole “Klingons are honorable” thing, which was a huge change.

          • Nick says:

            This makes the transition occurring between TOS and TNG even more plausible, if at all. How did ENT treat the Klingons? Checking Memory Alpha, it looks like there’s all this stuff about an augment virus to explain the difference in appearance…. I’ve never seen those episodes, so this is new to me.

          • John Schilling says:

            TOS didn’t make a point of saying that Klingons were honorable, but “Day of the Dove” in particular did a fair job of showing it. Other episodes less so, but possibly due to less opportunity as they were shown from a one-dimensional adversarial point of view. Could a GI on Guadalcanal have properly assessed the extent to which Bushido traditions remained relevant to 20th-century Imperial Japan?

        • Aapje says:

          @Deiseach

          Arguably the Mongols became a post-scarcity society as well when they demanding huge tribute from those they defeated. The Mongol home lands were quite poor, I think. The Klingons may have a similar racket going on.

          • hyperboloid says:

            A relevant difference is that because of the technological limitations of the time, advanced technology was not needed to effectively wage war, and as such a relatively primitive nomadic culture such as the Mongols could be militarily self sufficient.

            If the warships you need to maintain your empire are being manufactured by people in some vassal state, then you’re not going to have an empire for very long.

      • Matt M says:

        Yeah, this makes sense to me. To the extent that we can wrap our heads around the idea that replicator = “post-scarcity” = nobody actually has to farm for food to exist anymore, why couldn’t everyone be a warrior? The only limiting factor would seem to be the lack of enemies to conquer, which they seem to solve through inter-house in-fighting.

  2. Sniffnoy says:

    (Originally a comment comment on LW2; reposting it here, edited to fit the differing context, as I think it deserves more visibility)

    So Eliezer Yudkowsky has talked a bunch about the need to solve “paradoxes” such as Pascal’s Mugging that result from unbounded utility functions. (Yes, I’m aware that the Pascal’s Mugging problem that’s often presented doesn’t particularly rely on unbounded utility functions, but Eliezer’s original version, as intended he intended it, does, as he’s expained.) But really one should not be using unbounded utility functions in the first place; the resolution of such problems is that they should never come up in the first place. And I get the impression that many people don’t know why, so I think I should explain.

    (Note for those unfamiliar: I’m talking here about utility functions in the decision-theoretic sense, not the utilitarian sense (“E-utility”). If people are confused about the difference I can I’m sure dig up an old comment of mine explaining the matter…)

    Anyway, the basic question is: Where do utility functions come from? Like, why should one model a rational agent as having a utility function at all? The answer of course is either the VNM theorem or Savage’s theorem, depending on whether or not you’re pre-assuming the notion of probability (you really shouldn’t, I’d say, but that’s another matter). Right, both these theorems take the form of, here’s a bunch of conditions any rational agent should obey, let’s show that such an agent must in fact be acting according to a utility function (i.e. trying to maximize its expected value).

    Now here’s the thing: The utility functions output by Savage’s theorem are always bounded. Why is that? Well, essentially, because otherwise you could set up a St. Petersburg paradox that would contradict the assumed rationality conditions (in short, you can set up two gambles, both of “infinite expected utility”, but where one dominates the other, and show that both A. the agent must prefer the first to the second, but also B. the agent must be indifferent between them, contradiction). Thus we conclude that the utility function must be bounded.

    OK, but what if we base things around the VNM theorem, then? It requires pre-assuming the notion of probability, but the utility functions output by the VNM theorem aren’t guaranteed to be bounded.

    Here’s the thing: The VNM theorem only guarantees that the utility function it outputs works for finite gambles. Seriously. The VNM theorem gives no guarantee that the agent is acting according to the specified utility function when presented with a gamble with infinitely many possible outcomes, only when presented with a gamble with finitely many outcomes.

    Similarly, with Savage’s theorem, the assumption that forces utility functions to be bounded — P7 — is the same one that guarantees that the utility function works for infinite gambles. You can get rid of P7, and you’ll no longer be guaranteed to get a bounded utility function, but neither will you be guaranteed that the utility function will work for gambles with infinitely many possible outcomes.

    This means that, fundamentally, if you want to work with infinite gambles, you need to only be talking about bounded utility functions. If you talk about infinite gambles in the context of unbounded utility functions, well, you’re basically talking nonsense, because there’s just absolutely no guarantee that the utility function you’re using applies in such a situation. The problems of unbounded utility that Eliezer keeps pointing out, that he insists we need to solve, really are just straight contradictions arising from him making bad assumptions that need to be thrown out. Like, they all stem from him assuming that unbounded utility functions work in the case of infinite gambles, and there simply is no such guarantee; not in the VNM theorem, not in Savage’s theorem.

    If you’re assuming infinite gambles, you need to assume bounded utility functions, or else you need to accept that in cases of infinite gambles the utility function doesn’t actually apply — making the utility function basically useless, because, well, everything has infinitely many possible outcomes. Between a utility function that remains valid in the face of infinite gambles, and unbounded utility, it’s pretty clear you should choose the former.

    And between Savage’s axiom P7 and unbounded utility, it’s pretty clear you should choose the former. Because P7 is an assumption that directly describes a rationality condition on the agent’s preferences, a form of the sure-thing principle, one we can clearly see had better be true of any rational agent; while unbounded utility… means what, exactly, in terms of the agent’s preferences? Something, certainly, but not something we obviously need. And in fact we don’t need it.

    As best I can tell, Eliezer keeps insisting we need unbounded utility functions out of some sort of commitment to total utilitarianism or something along the lines of such (that’s my summary of his position, anyway). I would consider that to be on much shakier ground (there are so many nonobvious large assumptions for something like that to even make sense, seriously I’m not even going into it) than obvious things like the sure-thing principle, or that a utility function is nearly useless if it’s not valid for infinite gambles. And like I said, as best I can tell, Eliezer keeps assuming that the utility function is valid in such situations even though there’s nothing guaranteeing this; and this assumption is just in contradiction with his assumption of an unbounded utility function. He should keep the validity assumption (which we need) and throw out the unboundedness one (which we don’t).

    • shakeddown says:

      so in practical terms, what does that imply for the thought experiment where someone says “Give me a dollar or I torture 3^^^^3 people”? Assuming he can add an arbitrary number of ^, isn’t that unbounded?

      • Sniffnoy says:

        OK, so, there’s a few different things going on here.

        Assuming he can add an arbitrary number of ^, isn’t that unbounded?

        So basically you’re implicitly making the same mistake Eliezer is here. You’re implicitly assuming some form of total utilitarianism or something similar in when you make the implicit inference from “arbitrarily many people being tortured” to “arbitrarily negative (decision-theoretic) utility”. And what I’m saying is, no, let’s go back to fundamentals, utility functions (again, in the decision-theoretic sense) must be bounded for very good reasons, and if total utilitarianism conflicts with that, then that indicates a problem in total utilitarianism, not in boundedness of utility functions.

        so in practical terms, what does that imply for the thought experiment where someone says “Give me a dollar or I torture 3^^^^3 people”?

        For the original form of Eliezer’s “Pascal’s Mugging” thought experiment, which involved comparing the growth/decay rates of two functions — the badness being threatened, and the probability that the mugger is telling the truth — it completely kills it. Because the point was that the former grows much faster than the latter falls off, and if in fact the former is bounded, well, the problem goes away. As do the other problems Eliezer keeps pointing out, not noticing that the real solution to all of them is to give up on the idea of unbounded (decision-theoretic) utility.

        That said, this doesn’t necessarily have any effect on what you might call a more practical form of the problem, one that doesn’t necessarily rely on taking the numbers to infinity (and this tends to be the form most people mean, although it was not the originally intended problem). Just saying that the utility function must be bounded doesn’t really tell you anything about its particular growth rate at any particular point. So it doesn’t really affect that; my post above has nothing to say about such a “practical” form of the problem. (Although, you might want to see Zvi Mowshowitz here about that form of the problem.)

        • John Schilling says:

          Because the point was that the former grows much faster than the latter falls off,

          Citation very much needed.

          • Sniffnoy says:

            Sure, here’s Eliezer Yudkowsky writing about the problem. Here’s my attempt to excerpt the relevant parts:

            On conventional epistemology, the prior probability of hypotheses diminishes exponentially with their complexity; if it would take 20 bits to specify a hypothesis, then its prior probability receives a 2^-20 penalty factor […]

            […]

            If Pascal’s Mugger had only offered to save a mere googol lives (10^100), we could perhaps reply that although the notion of a Matrix Lord may sound simple to say in English, if we actually try to imagine all the machinery involved, it works out to a substantial amount of computational complexity. […] And so (we reply) when mere verbal English has been translated into a formal hypothesis, the Kolmogorov complexity of this hypothesis is more than 332 bits – it would take more than 332 ones and zeroes to specify – where 2^-332 ~ 10^-100. Therefore (we conclude) the net expected value of the Mugger’s offer is still tiny, once its prior improbability is taken into account.

            But once Pascal’s Mugger offers to save a googolplex lives – offers us a scenario whose value is constructed by twice-repeated exponentiation – we seem to run into some difficulty using this answer. Can we really claim that the complexity of this scenario is on the order of a googol bits – that to formally write out the hypothesis would take one hundred billion billion times more bits than there are atoms in the observable universe?

            And a tiny, paltry number like a googolplex is only the beginning of computationally simple numbers that are unimaginably huge. […] 3↑↑↑3 is still quite simple computationally – we could describe a small Turing machine which computes it – so a hypothesis involving 3↑↑↑3 should not therefore get a large complexity penalty, if we’re penalizing hypotheses by algorithmic complexity.

            I had originally intended the scenario of Pascal’s Mugging to point up what seemed like a basic problem with combining conventional epistemology with conventional decision theory: Conventional epistemology says to penalize hypotheses by an exponential factor of computational complexity. This seems pretty strict in everyday life: “What? for a mere 20 bits I am to be called a million times less probable?” But for stranger hypotheses about things like Matrix Lords, the size of the hypothetical universe can blow up enormously faster than the exponential of its complexity. This would mean that all our decisions were dominated by tiny-seeming probabilities (on the order of 2^-100 and less) of scenarios where our lightest action affected 3↑↑4 people… which would in turn be dominated by even more remote probabilities of affecting 3↑↑5 people…

            This problem is worse than just giving five dollars to Pascal’s Mugger – our expected utilities don’t converge at all! Conventional epistemology tells us to sum over the predictions of all hypotheses weighted by their computational complexity and evidential fit. This works fine with epistemic probabilities and sensory predictions because no hypothesis can predict more than probability 1 or less than probability 0 for a sensory experience. As hypotheses get more and more complex, their contributed predictions have tinier and tinier weights, and the sum converges quickly. But decision theory tells us to calculate expected utility by summing the utility of each possible outcome, times the probability of that outcome conditional on our action. If hypothetical utilities can grow faster than hypothetical probability diminishes, the contribution of an average term in the series will keep increasing, and this sum will never converge – not if we try to do it the same way we got our epistemic predictions, by summing over complexity-weighted possibilities. […]

            Unfortunately I failed to make it clear in my original writeup that this was where the problem came from, and that it was general to situations beyond the Mugger. Nick Bostrom’s writeup of Pascal’s Mugging for a philosophy journal used a Mugger offering a quintillion days of happiness, where a quintillion is merely 1,000,000,000,000,000,000 = 10^18. It takes at least two exponentiations to outrun a singly-exponential complexity penalty. I would be willing to assign a probability of less than 1 in 10^18 to a random person being a Matrix Lord. You may not have to invoke 3↑↑↑3 to cause problems, but you’ve got to use something like 10^10^100 – double exponentiation or better. Manipulating ordinary hypotheses about the ordinary physical universe taken at face value, which just contains 10^80 atoms within range of our telescopes, should not lead us into such difficulties.

          • John Schilling says:

            On conventional epistemology, the prior probability of hypotheses diminishes exponentially with their complexity; if it would take 20 bits to specify a hypothesis, then its prior probability receives a 2^-20 penalty factor

            And there’s a citation for that, but it’s a dead link.

            And I think the claim is wrong. If I consider the set of hypothesis of the form, “the multiverse over the whole of its existence will contain exactly N sentient entities capable of being tortured”, then the integrated probability of these hypotheses for all values of N must be 1.0, but if I use Yudkowsky’s formulation I think that diverges to infinity. Need to double-check my math on this and don’t have the time right now, but I think he is vastly overstating the fatness of the tail.

          • Sniffnoy says:

            Ah, well, I was just providing a cite for this being Yudkowsky’s point, not for him being right. 😛

            The link there isn’t dead though, just malformed — it’s got two URLs in it concatenated; the points here and the second points here.

            Eliezer is basically trying to apply an informal version of Solomonoff induction here; it’s obviously not actual Solomonoff induction but it’s taking that same idea of how the universal prior works and applying it here. This would seem to indicate that if anything you have the situation backwards; the universal prior is indeed sometimes defined in such a way that it’s not an actual probability distribution, but in those cases the sum is less than 1, not greater, let alone infinite.

            Edit: I guess the easy way to see this is to sum over the descriptions rather than the numbers or situations being described.

            Also I guess the sum does indeed diverge if you leave out the restriction that the descriptions are drawn from a prefix-free language, which is an important part of how the universal prior works but which Eliezer doesn’t explicitly mention. So, uh, make sure you account for that.

    • Ivy says:

      Thanks, really interesting post.

      But I’m not clear on what it means for unbounded utility functions not be “guaranteed to work”. What are the practical consequences?

      Apologies if I’m making a category error, but it seems definitely possible to program an agent with an unbounded utility function (like maximizing accumulated wealth or paperclips), or train a reinforcement learning agent with an unbounded reward signal. These agents would be vulnerable to Pascal’s Muggings and the other problems of unbounded utility that Eliezer talks about.

      Is your point that we should never create such agents? I’d agree, but someone else might, and if they do we really will have to deal with these paradoxes.

      • Sniffnoy says:

        Good questions.

        But I’m not clear on what it means for unbounded utility functions not be “guaranteed to work”. What are the practical consequences?

        So when I speak of a utility function “working”, that’s actually something you need to compare two situations to test, it’s not actually something you can determine looking at an individual situation. So when I say “such a utility function is not guaranteed to work for infinite gambles”, what I actually mean is “such a utility function is not guaranteed to work for comparing gambles when at least one of the gambles has infinitely many possible outcomes”.

        That is to say, a utility function is supposed to reflect an agent’s preferences — that given two possible actions to compare, the agent will prefer the one with the higher expected utility (or be indifferent between them if these are equal). For a utility function not to “work”, in this context, means that it fails to reflect the agent’s preferences; maybe the agent prefers A over B even though B has higher expected utility, for instance.

        Note that here I’m assuming we’re starting with the idea of an agent with pre-existing preferences, and trying to build a utility function to match. The VNM theorem, and Savage’s theorem, both ensure that as long as the agent’s preferences satisfy certain rationality assumptions, there is always a utility function that accomplishes this. However, as I said above, VNM, or Savage without P7, only guarantees that the utility function works properly when comparing actions with only finitely many possible outcomes, which to my mind is not that useful for building a truly general agent.

        This is a little different than the setting of your next question:

        Apologies if I’m making a category error, but it seems definitely possible to program an agent with an unbounded utility function (like maximizing accumulated wealth or paperclips), or train a reinforcement learning agent with an unbounded reward signal. These agents would be vulnerable to Pascal’s Muggings and the other problems of unbounded utility that Eliezer talks about.

        Sure, absolutely you could build such a thing! Rather than starting with rationality assumptions and concluding that one should use a matching utility function, one could just start with a utility function as your basic assumption. I don’t see why you’d want to do that, but you could.

        My point though is that if you do this you will run into more fundamental problems than the sorts of paradoxes Eliezer points out. For instance, consider the St. Petersburg-style problem considered above. The agent can choose between option A, which yields utility 1 with probability 1/2, utility 2 with probability 1/4, utility 4 with probability 1/8, etc; and option B, which yields utility 2 with probability 1/2, utility 4 with probability 1/4, utility 8 with probability 1/8, etc.

        In the original setting, the rationality assumptions made led to a contradiction — they forced the agent to both prefer B to A (because it dominates), but also to be indifferent between the two (essentially, because they both have infinite expected utility). In this new setting, though, where we’re not starting from rationality assumptions but rather directly from a utility function, there’s no contradiction: The agent is, quite definitely, indifferent between A and B. Because they both have infinite expected utility, and the agent’s preferences match the given utility function by definition.

        But while there’s no contradiction here, it does mean that we must be violating some rationality assumption, if we’re using an unbounded utility function, since we know that if we accepted all of Savage’s assumptions we’d have to be using a bounded utility function. And indeed we are: The agent is indifferent between two options even though one of them dominates the other. I don’t think that’s something we want, and, like I said, it seems like a much more fundamental problem than Pascal’s Mugging or anything like that. So yes, if you want to write such an agent you’ll have to deal with the sorts of problems Eliezer points out, but also much more fundamental ones; and the only solution to those more fundamental ones is to use a bounded utility function. (So it is indeed my opinion that we shouldn’t create such unbounded-utility agents, because they’d be necessarily irrational.)

        • Sniffnoy says:

          Oops, I messed up the statement of the paradoxical (or just problem-revealing, depending on context) scenario.

          What it should be is, with probability 1/2 A yields 1 and B yields 2; with probability 1/4 A yields 2 and B yields 4; with probability 1/8 A yields 4 and B yields 8; etc.

          As I originally stated it, the statement that B dominates A doesn’t follow. With the correct statement, it does.

          The rest of what I said about this comparison then goes through.

    • Anon. says:

      Where do utility functions come from? Like, why should one model a rational agent as having a utility function at all?

      You should go up an additional step. EY/LW seem to be stuck in a very old, and empirically unsupported, notion of the self/agents/agency/introspection/self-knowledge/etc.

      • Sniffnoy says:

        Obviously these aren’t very good descriptions of how humans act; we know that people don’t actually obey the rationality assumptions made by VNM or Savage. Hell there’s plenty of examples of this on LW; you can go read about the Ellsberg paradox there, for instance. If you want to build an artificial intelligence though, an artificial agent, I think it does make sense to want it to obey rationality assumptions, of the sort that (per Savage) would force it to be acting as if it had a utility function. So I think such questions are still very much worth considering. LW is only partly about the art of human rationality; it’s also about the art of designing rationality into any future agents we might design.

    • ADifferentAnonymous says:

      By solving Pascal’s Mugging via rejecting total utilitarianism and using a bound, you’re basically saying we should collapse “an incomprehensibly vast number of people being tortured” into one outcome, and only caring about it a finite amount, regardless of what the actual number is. I can’t accept that.

      What if our *priors* are such that expected utility is always finite? That is, no possible evidence can convince us that we’re actually facing a St. Petersburg problem? So considerations like the leverage penalty actually induce a bound on expected utility, despite the utility function itself being unbounded?

      • Sniffnoy says:

        By solving Pascal’s Mugging via rejecting total utilitarianism and using a bound, you’re basically saying we should collapse “an incomprehensibly vast number of people being tortured” into one outcome,

        This is not correct.

        You don’t get the bounded utility function you’re looking for by starting from total utilitarianism and then applying a cap. I don’t think anyone would disagree that more people being tortured is to be dispreferred. However an increasing function can still be bounded. Because of issues like this I don’t think a good utility function should ever actually attain its supremum or infimum.

        Again, it’s very much not clear to me that total utilitarianism, or indeed any sort of utilitarianism, actually makes sense, and I think starting from it as such a basic assumption is a bad idea. To my mind the right question is “What do we conclude from the fact that utility functions must be bounded?”, not “How do we shoehorn bounded (decision-theoretic) utility functions and total utilitarianism into the same framework?”

        (It may also be useful to remember here that the raw numbers coming out of a utility function are not by themselves meaningful…)

        What if our *priors* are such that expected utility is always finite? That is, no possible evidence can convince us that we’re actually facing a St. Petersburg problem?

        I think you need to make this more precise, and I don’t think it works when you do.

        I’m assuming we’re working here from a framework of “here is an agent with preferences that satisfy rationality assumptions, let’s come up with a utility function that describes this”. Because that, I would say, is where utility functions come from. Yes, you can just start out with the assumption that you’re going by a utility function, but such a thing seems totally unjustified to me.

        Note that, as I pointed out to Ivy above, if you do do the latter, then there is no contradiction; there is a violation of rationality assumptions, but no contradiction, the agent is quite definitely indifferent if you define it in such a way. But once again, I think such rationality assumptions are pretty bad, and I think this approach is unjustified in the first place.

        With the former approach… well… what you say doesn’t really help. The fact is that starting from Savage’s assumptions we are able to derive contradictory conclusions about such a situation. It simply doesn’t matter whether the agent is ever presented with such a choice or not; we assumed from the beginning that it has a preference (which may be indifference) between any given pair of actions, and there is no possible preference that it can have here and satisfy all the rationality assumptions we’ve made. (And of course you could toss one out, perhaps, but which one?)

        In order for your approach to make sense we’d have to make quite different starting assumptions — that the agent simply doesn’t need to have preferences between any two actions, that it only needs to have preferences over actions it might have the opportunity to take. I can’t say for a fact that such an approach can’t be made to work, but I’d be very skeptical. (Partly because I’d be worried about trusting the “this can’t happen” assumption; when that assumption is violated, what does the agent do?) But if you want to try to work such a thing out, hey, go ahead…

        • ADifferentAnonymous says:

          Sure, you can approach your bound asymptotically. You still end up basically resisting the mugger because 3 ^^^^^ 3 people tortured isn’t that much worse (to you) than 3 ^^^^ 3 people being tortured. My intuition flatly rejects that.

          I’m philosophically more ok with not having preferences between some impossible actions–saying, essentially, that I’m a dumb finite being and with my current understanding of math (possibly any understanding of math) I can’t really answer St. Petersburg scenarios sensibly.

          • Sniffnoy says:

            <shrug> It’s maybe a possible approach I suppose? Again, the agent has to take some action; it shouldn’t just crash or something in “can’t-happen” scenarios, so I remain pretty damned skeptical such a thing is workable. But hey, if you do come up with a framework that allows you to formalize this idea, do make sure to publicize it…

          • Creutzer says:

            My intuition flatly rejects that.

            Your intuition can’t consistently do that if it also says that one should resist Pascal’s mugging.

          • Sniffnoy says:

            Creutzer: It’s not clear to me that that’s true? That there truly is no other way of solving that particular problem? Like FDT might imply a general “don’t respond to threats” principle. Or one could imagine having such a principle even without FDT.

            Which is why I wouldn’t focus so much on that particular problem. Point is, unbounded utility introduces a number of problems, that in fact Eliezer has written about but refuses to accept the simple solution to. But each problem individually may have other solutions; Pascal’s Mugging likely does.

            …but not the dominated St. Petersburg problem. Unless you’re going to develop some entirely new theory like ADifferentAnonymous wants to, that one really only does have the one solution.

        • Sniffnoy says:

          But once again, I think such rationality assumptions are pretty bad, and I think this approach is unjustified in the first place.

          Argh. Meant to write I think violations of such rationality assumptions are pretty bad. Blech.

    • Nabil ad Dajjal says:

      What I’ve never understood about scenarios like Pascal’s mugging is why you’d bother to calculate expected return in the first place.

      If someone says “I’m secretly an all-powerful god who can sentence uncountably vast numbers of people to eternal damnation but I really need a quarter for the gumball machine” the prior probability for that claim to be true is 0. It’s nonsense and isn’t worth considering.

      • Evan Þ says:

        Well, the Bible does warn us to show hospitality to strangers as people have sometimes entertained angels unawares.

      • Jiro says:

        The prior isn’t 0, it’s just very low. And here the difference between 0 and very low matters.

      • Sniffnoy says:

        That’s easy to say, but how do you formalize that intuition so that you can program an AI so as to reach the same conclusion? Remember, the question’s not just about getting the right answer for yourself here.

        Also saying that the probability is zero is quite a bad solution in the first place because taken seriously it means you can’t update if the person does in fact display what appears to be godlike power.

  3. Well... says:

    OK, I’m now 6 episodes through “Manhunt: Unabomber.” Impressions so far:

    – The way they handle Kaczynski’s philosophy, when they are really focusing on it, is actually kinda interesting. (I just wish that comprised more of the show than it does.)
    – The dialog writing is mostly pretty terrible.
    – The acting is mostly pretty terrible.
    – The casting is mostly pretty terrible.
    – I like certain aspects of the cinematography–specifically, the way many of the shots are composed–but the way they move the camera (or don’t move it) bugs me and for some reason I am particularly distracted by the DP’s choice of lenses. (??)
    – I was surprised to learn the FBI agent who’s the inspiration for the show’s main character was one of the producers, since so much of his character’s role in the case is depicted in a completely inaccurate way.
    – While the main “facts” are stuck to somewhat faithfully, the show takes egregious liberties with historical accuracy at many of the more detailed levels.

    I watch the show for Kaczynski and his philosophy of technology, but apparently most people are watching the show for “workplace drama,” “crime-mystery drama,” “home life drama,” “coming of age drama,” “dealing with a fax machine drama,” “reading a newspaper drama,” etc. In this age of targeted niche content, I expect they should have made a show suited to my particular tastes by now!

    I’ll probably finish the series up tonight.

    • j1000000 says:

      “In this age of targeted niche content, I expect they should have made a show suited to my particular tastes by now!”

      Seems like your niche is one Netflix wouldn’t want to target — persuasive anti-technology shows might hurt their bottom line a bit. The concept of Netflix and binge watching brings to mind Kaczynski’s anti-moderation stance: “Never forget that the human race with technology is just like an alcoholic with a barrel of wine.” I can never stop at just one episode of The Office…

      • Well... says:

        persuasive anti-technology shows might hurt their bottom line a bit.

        Hah, true.

        • j1000000 says:

          So, do you find Kaczynski’s writing particularly persuasive, independent of it being surprisingly sane for a violent murderer?

          I read it after edgelords of the alt-right became interested in it for a period, and while I basically agreed with his view of industrial society (short of his methods of revolt, of course), I didn’t think he was “right” in any way — at this point in my Internet life, I am effectively a nihilist, and read every piece as if it’s people expressing preferences in candy bar tastes. My tastes align with our host Scott’s in almost all regards, but I mostly dream of belief in God at this point.

          • Well... says:

            I don’t necessarily agree with Kaczynski’s ideas, but I do admire them and take them seriously. I have a lot of Luddite tendencies and a visceral dislike of urban environments to boot.

            It’s kind of funny thinking about those all-trite edgelords fawning over Kaczynski. I’ve said on my blog that they ought to join either Al Qaeda or their local Amish church: the society they crave is right there waiting for them and they could be part of it tomorrow. Retreating to an isolated cabin in the woods might be another option for them too. But talk is cheap, which is maybe where you get that candy bar tastes reading.

          • hlynkacg says:

            I think the fact that people find Kaczynski’s sanity surprising is indicative/illustrative of a dangerous blind-spot in a lot of rationalist’s thinking. Why wouldn’t he be sane?

            As for the rest that’s why the old right tends to view the alt-right with disdain. The vast majority of them are only in it for the lulz or for the sake of being contrarian and will scatter like dandelion seeds the moment the road gets hard.

          • Well... says:

            I think the fact that people find Kaczynski’s sanity surprising is indicative/illustrative of a dangerous blind-spot in a lot of rationalist’s thinking. Why wouldn’t he be sane?

            Imagine there are two Ted Kaczynskis. Both live Walden-style in an isolated cabin in Montana. Both were math geniuses. The two Teds have almost identical articulate, well-reasoned cases against modern industrial society and trace its evils to its technology.

            Ted1 argues his case in letters to editors of newspapers, in letters to his brother, to his very few friends/acquaintances at the local library, etc. but otherwise sighs that he will probably never convince a lot of people or make a lasting difference. Also, Ted1’s case against modern industrial society doesn’t call for violent revolution.

            Ted2’s case does call for violent revolution, and he does the same things as Ted1 but instead of sighing he mails bombs to people to get the public’s attention.

            The first time you ever encounter Ted1 is at the library where he is arguing his case. The first time you encounter Ted2 is after a bomb has exploded, killing or maiming someone in the news. And then later, around the time of his trial, a bunch of facts about his life come out that paint a pretty convincing picture that he at least had some schizoid tendencies, possibly (though not likely) up to paranoid schizophrenia.

            Ted1’s ideas wouldn’t seem “shockingly” sane, they’d just seem unusual to most people but otherwise would make sense, the same way Milton Friedman’s case for libertarianism makes sense even though most people ultimately reveal a preference for non-libertarianism. But Ted2 is of course the Ted we all got. And so when we read his manifesto we were expecting it to be Mein Kampf–an even more deranged Mein Kampf by a guy who couldn’t get any newspapers, let alone armies & generals, to listen to him without threatening violence.

            Edit to add: most people have a blind spot about their technology choices and how those choices impact their lives indirectly if not directly. Rationalists are not exempted from this blind spot, which might be noteworthy because being “rationalists” kind of implies they should be eliminating blind spots.

          • albatross11 says:

            I think it’s hard to distinguish from the outside between:

            a. Crazy in the sense of having defective functioning of your mental equipment leading you to express and act on ideas that make no sense to me.

            b. Crazy in the sense of having adopted and acted on ideas that make no sense to me despite properly-functioning[1] mental equipment.

            On one end of this, you get the nut who shot up that theater in Aurora, Colorado. He was apparently a pretty smart guy (IIRC, a neuroscience PhD student), but he lost touch with reality enough to convince himself he was the Joker and shoot up a theater showing a Batman movie.

            On the other end, you get someone coldly rational and evil as that guy who set off a car bomb in Oslo and then shot up a Norwegian summer camp full of the kids of the elites. His ideas may seem nuts to you or me, but it’s pretty clear that there wasn’t something defective with his brain causing him to imagine stuff that wasn’t there or hear voices or something–he just believed weird things and did horrible stuff based on those beliefs.

            [1] That is, mental equipment that works about as well as the average person on the street, including all the biases, mental errors, and self-deception that mortal flesh is heir to.

          • albatross11 says:

            ObLiteraryReference:

            SM Stirling’s _Change_ series includes a pretty important role for the Unabomber, who escapes prison when the world changes and ends up as the head of an anti-technology cult….

    • Anon. says:

      Have you seen Stemple Pass? It’s a really weird minimalist movie seemingly designed to put you in a coma, but it actually managed to shift my view of Kaczynski quite a lot.

  4. toastengineer says:

    Hey, I just wanted to thank this community for all the support you folks showed me when I asked for help finding work; after a lifetime of no-one giving a shit it’s quite a shock to see people making such a fuss over helping me out.

  5. Collin says:

    Have you experimented with transcranial direct current stimulation?

    • Well... says:

      It was a few years ago but I remember reading a bunch of articles about it. I was specifically interested in learning what the lasting side-effects were. If I remember right, after one day of TCS they included headaches and depression lasting about 2 weeks.

      If your head passed quality control inspection it should come with a label that reads “CAUTION: NO USER-SERVICEABLE PARTS INSIDE. REFER SERVICING TO QUALIFIED SERVICE PERSONNEL.”

      • The Nybbler says:

        If your head passed quality control inspection it should come with a label that reads “CAUTION: NO USER-SERVICEABLE PARTS INSIDE. REFER SERVICING TO QUALIFIED SERVICE PERSONNEL.”

        That just indicates one of the places there’s a hidden screw. (the other place is under the warranty sticker, which alas my head lacks)

  6. Le Maistre Chat says:

    Christian transhumanism: what are the issues involved?
    I think the most obvious point is when GK Chesterton ‘said’ “There have been tyrannical angels since the days of Noah, and our tools have been rebelling against us since the first peasant stepped on a rake.” There is absolutely no way that humans who have turned their backs on God will program a human-friendly superhuman computer. If a superhuman computer can be built during this zeitgeist, either the Butlerian Jihad needs to happen or we have to hope that moral realism works such that a superhuman intelligence will know and act on moral facts more completely than we do.
    This latter seems sketchy for Christians, who note no strong correlation between IQ and holiness and believe in fallen angels.

    So perhaps strong AI is off the table. What about things like genetic engineering? Are we called to be against it, as CS Lewis’s “Abolition of Man” suggests, or what?

    • Evan Þ says:

      Counter (and I’m a Christian myself): Humans who have turned their backs on God can demonstrably create a society no worse off than this present world. It is possible that, similarly, they can program a computer that will be no less friendly than this present world. Still, they’d be bulldozing so many Chestertonian fences that it’s unlikely – even assuming God won’t subtly intervene through apparent coincidences to stop them.

      I think Christianity’s farther-reaching implication for transhumanism is the soul: something apart from the material that makes up your brain and body, known to survive bodily death, in which lies your personal identity. First, we don’t need to despair at death “as those who have no hope”; Christ is risen, and God will resurrect His people as well. Second, we don’t need to ponder whether a hypothetical duplicate of yourself assembled after a thousand years will share your personal identity (or whether two simultaneous duplicates will); your identity doesn’t lie in the material of your body. Third… well, other implications are left as an exercise for the reader, or maybe for myself in a followup.

      • Le Maistre Chat says:

        It is possible that, similarly, they can program a computer that will be no less friendly than this present world.

        Well, that’s more reassuring than being killed for paperclips, but not really friendly. 🙂

        I think Christianity’s farther-reaching implication for transhumanism is the soul: something apart from the material that makes up your brain and body, known to survive bodily death, in which lies your personal identity. First, we don’t need to despair at death “as those who have no hope”; Christ is risen, and God will resurrect His people as well.

        Of course. Transhumanists like Yudkowsky are placing their hope in resurrection by a superhuman computer programmed to love them. We don’t need that.

        • Evan Þ says:

          Transhumanists like Yudkowsky are placing their hope in resurrection by a superhuman computer programmed to love them. We don’t need that.

          But they’re hoping for resurrection and eternal life, and those are perfectly appropriate hopes. That’s what first interested me in Less Wrong: here are some atheists who have things exactly right about death. It is bad. Catastrophically bad. The most important thing in the world is to oppose and end it.

          Sadly, Yudkowsky et al’s atheism causes them to miss the correct solution… but they’re almost the only ones who recognize death as such a great problem.

          • Nick says:

            Hah, I had very nearly the opposite reaction to Eliezer’s anti-deathism, and I say this as a Catholic: no, death is fine! The issue is no afterlife! I don’t mind an eternity of the beatific vision, but 10^100 years in this world of sorrows? Count me out.

      • SamChevre says:

        Nitpick that really isn’t: I’m not convinced that “Humans who have turned their backs on God can demonstrably create a society no worse off than this present world.”

        Whether that’s true depends a lot on how much of what makes society work, now, has to do with habits of thought and institutions that were originally based on belief in God.

        • The original Mr. X says:

          Reminds me of an article I read by (I think) Tom Holland, who said that he gained a greater respect for Christianity after realising that places like pagan Sparta and Rome would be considered horrifically cruel dystopias if they existed nowadays. If pre-Christian societies were fine with killing people for public entertainment or declaring a yearly open season on the peasantry, there’s no guarantee that post-Christian societies won’t slip back into similar habits.

    • Well... says:

      People want transhumanism because they see their minds/bodies as blank canvases, platforms to be built on, or else as these horribly broken insufficient shells we’re born with.

      I think God would disagree with that description of Man. At the end of the sixth day when He looked at His work and said it was good, I’m pretty sure that’s exactly what He meant.

      • Evan Þ says:

        On the other hand, now in a Fallen world, our bodies are broken and imperfect. By some interpretations (including mine), under God’s original design, we were supposed to literally live forever in these bodies.

        • Well... says:

          Then is transhumanism just another attempt to build a tower of Babel?

          • Randy M says:

            What was the sin of Babel that a Christian Transhumanist should avoid? Where precisely does it veer off from the impulse to heal and succor shared by physicians great and mundane? And if I add one more rhetorical question, do I set off the Sidles detector?;)

            We have towers now, higher likely that Babel would have been conceived of being. Can there likewise be fundamental biological manipulation that stems from purely good motives?

            [edit: dodrian basically said all that first and better]

    • RDNinja says:

      In my SF writing, I’ve toyed with the idea of soulless AI being the perfect substrate for demon possession. Why fight for space in a human skull when they’re just building luxury apartments for you?

      As far as body modification goes, we’ve been promised new, perfect bodies in heaven, so I don’t see much of an issue.

      • The Nybbler says:

        As far as body modification goes, we’ve been promised new, perfect bodies in heaven, so I don’t see much of an issue.

        But they’d still be modified, as your modification indicates your preference and thus what you consider perfection! OK, maybe not that drunken tattoo with your old girlfriend’s name, misspelled.

        • RDNinja says:

          But as a corrupt, sinful human being, what I consider to be a perfect body is almost certainly warped.

      • Randy M says:

        In my SF writing, I’ve toyed with the idea of soulless AI being the perfect substrate for demon possession.

        Not to compare thinking about writing with the more exalted task of actually doing so, but I had ideas along a similar line stemming from thinking about a set-up for a near-future D&D that mixed plausible science with religious elements. After signs of intelligent life are identified on distant planets, an AI is designed to guide the exploratory spaceships. Demons attempt to posses the computer on the chance this is some Edenic race to tempt and corrupt.

      • Evan Þ says:

        I’ve pondered the same question, too. If we do ever get AI that passes the Turing test (in the sense intended, not in the sense of “pretend to be a teen who doesn’t speak good English”), I’m going to be scared that it actually happened.

        • Nick says:

          Where is your faith? If we truly create intelligent AI, we just will have ensouled them. And they shall know God. 😀

          • We do that with the organic version, after all.

          • Evan Þ says:

            But then, theology has long recognized a difference between creating and begetting. Humans can beget and bear ensouled life, but can we create it?

            If we do create AI and somehow determine it possesses a soul, I’d be more likely to say it was specially given a soul by God.

          • Betty Cook says:

            @ Evan: To quote R. A. Lafferty (1968): “The spirit came down once on water and clay. Could it not come down on gell-cells and flux-fix?”

          • Evan Þ says:

            @Betty, “on the seventh day, God rested” from His work of creation. According to some theologians, we’re still in that seventh day.

            But then, according to others, we’re in a new eighth day. And God said He can raise up even the stones (which probably contained a lot of silicon, right?) to cry out praises to Him…

          • Nick says:

            Evan, yeah, I shouldn’t quite have said that we ensoul it. The point is that if the AI is rational (in the sense Aquinas meant “rational” in “rational animal”), then it will have a soul, regardless of how it got there.

      • Nancy Lebovitz says:

        “In my SF writing, I’ve toyed with the idea of soulless AI being the perfect substrate for demon possession.”

        Something of the sort appears in R. A. Lafferty’s _Past Master_, though it’s Programmed Mechanical People rather than AIs. The idea isn’t especially developed, though.

    • dodrian says:

      A first look at transhumanism brings up comparisons to the Garden of Eden and Tower of Babel: man trying to be like God. Not a great start.

      A second look sees attempts to end poverty and suffering: very laudable and Biblically-aligned goals.

      I think any Christian look at issues of transhumanism needs to balance those two perspectives.

      What are the goals of genetic engineering? Curing debilitating diseases and restoring dignity to the suffering seems like a valid case for Christian transhumanism. Prolonging life is something that’s been happening throughout the 20th century, though mainly through better knowledge of how to keep the body in good shape and prevent/cure common causes of death – would a direct manipulation of genes to slow cell decay really be any different? What if life could be prolonged indefinitely? To me that seems like it leans heavily towards the Babel/Eden side of the spectrum, though if we had that technology I could certainly imagine many Christians making the case that we have a duty to preserve life if we can.

    • Nancy Lebovitz says:

      It may depend on the details of the transhumanism. I doubt that there’s a good Christian argument against putting a sensor under your skin.

      As for death, I doubt that uploading gives complete immortality– even if you don’t get caught by the sun turning into a red giant or by a software glitch, the universe is running down. I assume that either your soul goes to its fate when your body dies or uploading merely delays death.

  7. Andrew Hunter says:

    Let’s say that we had a law where on your 40th birthday, you got to pick a college student, punch them in the face, take all the money from their wallet, and make them your servant for a week. Clearly a pretty shitty law. We should repeal it.

    But suppose this has been the law for a century. If it were repealed, you’d have a lot of 39-year-olds raising bloody murder: they got punched and enslaved 19 years ago, consarn it, and now it’s only fair that they get the benefit. With a law this bad, I think most people would agree that no, it’s not fair, but too bad: this sucks too much to have, and you’re just going to have to eat it. (If 39-year-olds are a sufficiently powerful political constituency, we might see a phase out over ten years where a 29-year-old like me gets hosed, current 34-year-olds get to spit in someone’s face and demand $20, and this year’s crop gets the whole enchilada.)

    But the 39-year-olds do have at least a bit of a point: it isn’t fair that they’ve paid the cost for a system that they now don’t get to use. One could expect that the sudden expected influx of cash at 40 would warp purchasing power–you’d traditionally buy a house on your birthday using the kid’s labor as a down payment (work with me here…) and now these people would be locked out of that system. Phaseouts make this marginally less painful, but not by all that much–they really just smear the unfairness onto new people.

    Now obviously there are a lot of other real-life laws which cause economic distortion in bad ways that should be repealed: putting aside flat political concessions to get bills through congress, have any of them proposed real ways to make the change more fair? I’m not certain I know a good solution for this class of problems.

    • I’m not sure what point you are trying to make here. Yes, we should repeal bad laws. IMO, there are an awful lot of them, surely a majority of all laws.

      And yes, even bad laws often should be phased out, because we don’t want to overly penalize someone who has set up their life based on the law (to me a great current example is the proposed but dropped making “free” graduate tuition taxable. Being tax exempt is bad law in many ways, but a sudden repeal would screw over lots of people). But phasing out laws has a couple of downsides: 1) It makes the law a lot more complicated, and 2) in a few years a different legislature might reverse the phase-out and bring back the bad law, which is much more likely with a phase-out than a sudden repeal.

      But I’m not sure what else there is to say about it.

    • The Nybbler says:

      The traditional way to handle these things is to wait until a small generation (say, Generation X) is about to benefit, then screw them over. Been on the receiving end of that sort of thing so many times I never believe nonsense about paying one’s dues and reaping benefits in the future any more.

      • Andrew Hunter says:

        You are making the rare reverse is-ought error: I’m actually asking for ought.

      • fortaleza84 says:

        Can you give me a couple examples? I am a member of Generation X and I’m not aware of anything like that. The only thing I can think of that might qualify is social security which has raised retiring ages but this hardly feels like I am getting screwed over given that people are living longer and work is much less physically demanding than in the past, all due to the work of earlier generations.

        I actually feel pretty fortunate to belong to Generation X given that none of us (in the US) have had to deal with compulsory military service; I was able to buy a house before the big real estate booms after 2001 while at the same time taking advantage of drops in interest rates to refinance; it was a lot easier to get into an elite college than it is now; we were the last generation where it was normal as a child to just goof off with your friends instead of doing some structured and monitored activity; it was possible to go to an elite professional school without assuming a crippling level of debt and one’s student loans could be discharged in bankruptcy; probably other stuff too.

        I suppose every generation has pluses and minuses; it doesn’t seem to me that Generation X has been uniquely victimized by policy changes.

        • The Nybbler says:

          My examples are mostly on a personal level — a promise I’d get privilege or benefit X after N years, but during year N-1, said privilege or benefit was abolished. As for policy, along with Social Security, older X’ers in the US got hit by the higher drinking age.

          • TheContinentalOp says:

            That happened to me. The year I turned 17, the state of NJ raised the drinking age from 18 to 19. Then a year later when I was 18, the age was raised from 19 to 21.

          • fortaleza84 says:

            My examples are mostly on a personal level — a promise I’d get privilege or benefit X after N years, but during year N-1, said privilege or benefit was abolished.

            Can you give an example or two? If it’s not a matter of government policy, it’s seems to me it’s just random misfortune and not taking advantage of a group’s demographic weakness.

            As for policy, along with Social Security, older X’ers in the US got hit by the higher drinking age.

            Seems to me that’s not a situation where you paid your dues for years and then got the rug pulled out from under you. Besides, when we (Gen Xers) went to college, the alcohol flowed freely; there was easy access to alcohol from age 17 to 21, most grocery stores and even a lot of bars didn’t check ID very carefully. Nowadays these laws are actually enforced.

    • A1987dM says:

      Repeal the law immediately, but give all people currently aged between 19 and 39 (i.e. all the people who were punched and enslaved but won’t get to punch and enslave one themselves) monetary compensation. Kaldor–Hicks improvements FTW!

      • Anonymous says:

        Compensated change is a good idea. Tested at least as far back as the Meiji Restoration.

      • ADifferentAnonymous says:

        Yeah, this.

        The application to Social Security is that we should replace the payroll tax ASAP with a normal tax that’s designed to be the optimal way of raising revenue and doesn’t have weird distortions from pretending to involve an individual account.

    • Fossegrimen says:

      This is the retirement pension system, right? If so, yes it is unsustainable barring some sort of infinite growth/post-scarcity scenario and you’ll probably get shafted just as bad as you think and we might as well start winding it down now.

    • Brad says:

      @Andrew Hunter

      But the 39-year-olds do have at least a bit of a point: it isn’t fair that they’ve paid the cost for a system that they now don’t get to use. One could expect that the sudden expected influx of cash at 40 would warp purchasing power–you’d traditionally buy a house on your birthday using the kid’s labor as a down payment (work with me here…) and now these people would be locked out of that system. Phaseouts make this marginally less painful, but not by all that much–they really just smear the unfairness onto new people.

      Tax all the people that are still alive and benefited from the prior system and use the money to compensate their victims. To the extent there is still a shortfall because some of the beneficiaries are already dead, that’s life.

    • skef says:

      You might be right that the system we’ve cobbled together to try to keep present and future old people above the cat food level won’t be there for you. And that system includes a lot of inefficiencies (in terms of who the benefits go to) and rhetoric related to its political viability. And it is “unsustainable” in the narrow sense that it would very likely need money from some other source to avoid adjusting the benefit or taxation levels currently written into law.

      But none of that implies that letting old people slip down to the cat food level (or below) isn’t a travesty. Or that the arguments that preventing it does more harm than good are not themselves based on rhetoric grounded in bullshit psychology invented by non-psychologists (that is, economists). Or that it is unsustainable in the sense that the U.S can’t afford to pay to keep older people out of poverty.

      If the political winds shift to eliminate social security for Gen X in a way that needlessly condemns many people of that generation to grinding poverty, those responsible for that change will be at much greater moral fault than those that cobbled together the present system.

      If you disagree with that, just say you think redistribution is wrong and don’t bother with the political and generational epicycles.

      • Brad says:

        Why are elderly poor people such a compelling moral issue that a program whose main impact is to transfer money from poorer people to richer people is justified because a minor side effect is to benefit poor elderly people? Why, on the other hand, are poor non-elderly such a non-compelling moral issue that it is fine if they have to eat cat food?

        Frankly, I don’t buy it. There aren’t just “a lot of inefficiencies”. It isn’t an anti-poverty program in the first place.

        I’m perfectly fine with redistribution. From the rich to the poor, not the other way around.

        • skef says:

          Why are elderly poor people such a compelling moral issue that a program whose main impact is to transfer money from poorer people to richer people is justified because a minor side effect is to benefit poor elderly people?

          I’ve always considered this particular “instantaneous” argument to be beside the point and at least borderline disingenuous. It’s also claimed that the system is unsustainable in part because people get back more than they put in. If that happens with people paying in now, why is the fact of instantaneous transfer so important? If I save money for my own retirement, is it a tragedy that I can’t spend that money now while my old self does even though they also have a big pile of savings from all the other months I saved? (That jerk!)

          Why, on the other hand, are poor non-elderly such a non-compelling moral issue that it is fine if they have to eat cat food?

          This is a good actual question, but not as effective a rhetorical one as many people think. The rough answer seems to be that that many people don’t give the suffering of people they don’t know much epistemic respect. That is, it isn’t that they don’t care that others suffer, but they tend to stipulate on poor evidence that they aren’t suffering, or that they counter-factually wouldn’t be suffering given some change.

          Frankly, I don’t buy it. There aren’t just “a lot of inefficiencies”. It isn’t an anti-poverty program in the first place.

          My reading of the history is that it was a program put in place to reduce poverty that was justified on a different basis. This was a conscious decision motivated by the understanding that a targeted anti-poverty program would be much more likely to be dismantled than what they put together. That seems quite plausible to me.

          There are a lot of ways we trick ourselves into being a bit better. (The Constitution is another example, or at least I think that on more optimistic days.) Hypocrisy shouldn’t be a big deal to a consequentialist when it serves a purpose*.

          * [This is one key thing that our host, who is close to a consequentialist absolutist when it comes to action, and a deontic absolutist when it comes to knowledge, doesn’t seem to have grappled with. Utterances involve both but lean towards the action side. However, he seems virtually certain, with little apparent evidence, that picking apart this or that article will push the outlet that publishes them to do better, and push more people to read and trust those outlets, rather than (for example) just pushing people towards more dubious and polarized sources. I occasionally imagine him sitting in a post-apocalyptic bunker paging through old print issues of The Atlantic or whatever, looking for more articles to pick apart.

          In that light, maybe his use of statistics mostly about men calling other men faggots at work to argue that gay men are particularly letchy marks an improvement … ]

          • Brad says:

            @skef

            I’ve always considered this particular “instantaneous” argument to be beside the point and at least borderline disingenuous. It’s also claimed that the system is unsustainable in part because people get back more than they put in. If that happens with people paying in now, why is the fact of instantaneous transfer so important? If I save money for my own retirement, is it a tragedy that I can’t spend that money now while my old self does even though they also have a big pile of savings from all the other months I saved? (That jerk!)

            This is the switch part of the bait and switch, which is absolutely endemic when talking about social security. Sorry, I’m not buying it. You said it it is an anti-poverty program, let’s talk about it as an anti-poverty program. I’m not going to let you switch “it’s really just saving your own money” just because it is convenient for you to do so when I point that a regressive anti-poverty program is bizarre.

            This is a good actual question, but not as effective a rhetorical one as many people think. The rough answer seems to be that that many people don’t give the suffering of people they don’t know much epistemic respect. That is, it isn’t that they don’t care that others suffer, but they tend to stipulate on poor evidence that they aren’t suffering, or that they counter-factually wouldn’t be suffering given some change.

            I don’t know exactly what you expect me to do with this. If you have so much empathy for people that choose to support gold plated welfare for old people most of whom aren’t even poor, but not one cent for poor people that aren’t old, you should be overflowing with understanding for my position. But instead you call it “borderline disingenuous”.

            My reading of the history is that it was a program put in place to reduce poverty that was justified on a different basis. This was a conscious decision motivated by the understanding that a targeted anti-poverty program would be much more likely to be dismantled than what they put together. That seems quite plausible to me.

            There are a lot of ways we trick ourselves into being a bit better. (The Constitution is another example, or at least I think that on more optimistic days.) Hypocrisy shouldn’t be a big deal to a consequentialist when it serves a purpose*.

            We are talking about a program that’s 3.5% of GDP. Justifying it on consequentialist groups requires taking into account the enormous opportunity costs in the majority of the money not spent on the poor, but instead on your “trick”.

            Furthermore the program is structurally speaking a regressive spending program (larger checks go to wealthier elderly people) with a highly regressive tax. Outside of maybe the cigarettes excise tax, the single most regressive tax in the federal system. And one that accounts for almost 1/3 of federal revenue.

            Look at all the left wing politicians, media outlets, bloggers, and randoms on social media screaming their head off about the distribution of benefits from the tax cut. Yet somehow social security is not only immune from this analysis but bringing it up is “borderline disingenuous”?

          • On the question of the distributional effect of social security … .

            If we think in terms of lifetime rather than current income, it has features that go in both directions, some that redistribute down the income distribution, some up.

            The actual formula for taxes and benefits tends to redistribute from richer to poorer. If you pay very small amounts or for few years you get less money back, but not proportionally less. If you pay maximum amounts for the maximum number of years, you get no more back than if you had paid much less for a much shorter length of time.

            On the other hand, because full benefits do not require a full working life worth of payments, the system favors people who start earning money later, which makes it a better deal for people who go to college and perhaps schooling beyond that than for people who start working when they graduate from high school, which favors higher income people. That’s complicated by the fact that some low income people will spend a fair amount of time unemployed and so not paying in.

            The amount you get back depends on how many years you live past sixty-five. Wealthier people have longer life expectancies, so that also tends to transfer up the income ladder.

            I did some very rough calculations on this forty-some years ago for the first edition of Machinery. I believe others since have done a more detailed version. I don’t know whether, under current circumstances, the net effect is to redistribute up or down.

          • Jaskologist says:

            The Elder Friedman (not to be confused with the Young Friedman who comments here) had some thoughts on how SS redistributes which are also worth sharing.

            (tldw: it tends to redistribute from the poor to the middle.)

          • Brad says:

            On the question of the distributional effect of social security …

            If we think in terms of lifetime rather than current income,

            While I appricate the extra information, I don’t see why social security should get a different form of analysis than every other government program. When the distributional consequences of the tax bill were and are being condemned by everyone to the left of Ronald Reagan, there were no attempts to look at lifetimes.

            Take the grad student waiver issue, for example. We were constantly fed emotional appeals on the basis that they had so little money already and how could we expect them to pay any more in taxes. I didn’t see a single analysis that calculated lifetime expected earnings of those with tertiary degrees and suggested that it’s okay to tax the currently poor because they’ll be well off later.

          • skef says:

            This is the switch part of the bait and switch, which is absolutely endemic when talking about social security. Sorry, I’m not buying it. You said it it is an anti-poverty program, let’s talk about it as an anti-poverty program. I’m not going to let you switch “it’s really just saving your own money” just because it is convenient for you to do so when I point that a regressive anti-poverty program is bizarre.

            My point is that it is an anti-poverty program, even if it isn’t a carefully targeted one. The instantaneous analysis is another form of cherry-picking.

            At the bottom-end of earnings, we do have the earned income tax credit, which does counter-act the regressive nature of the social security tax somewhat.

            I don’t know exactly what you expect me to do with this. If you have so much empathy for people that choose to support gold plated welfare for old people most of whom aren’t even poor, but not one cent for poor people that aren’t old, you should be overflowing with understanding for my position. But instead you call it “borderline disingenuous”.

            It seems to be that this comment is an example of the sort of stipulation I’m talking about.

            I just looked for some statistics and two sites indicated that roughly 40% of households of people ages 51-64 have no savings and another 20% have between $1k and $50k of savings. The median household has between $10k and $20k. You make it sound as if maybe 5-10% of people or households at retirement age really need support and everyone else would be fine. That’s just not true.

            I agree with you that the funding system as it stands is regressive. I disagree about the political viability of the alternatives you favor and about the impact of the current system with all its faults.

          • Brad says:

            @skef

            The instantaneous analysis is another form of cherry-picking.

            I strengously disagree. It is how we analyze every other policy. It is insisting on a idiosyncratic analysis for social security that is cherry picking.

            At the bottom-end of earnings, we do have the earned income tax credit, which does counter-act the regressive nature of the social security tax somewhat.

            First, the seems like double counting. The EITC is its own anti-povery measure and a decent one at that (thought see two). If it is instead an offset to FICA then it isn’t an anti-poverty program at all but a way to make a highly regressive tax that brings in about 1/3 of revenue slightly less regressive.

            Second, the EITC, despite some rhetoric to the contrary, is essentially a refundable child tax credit. Households without children mostly get nothing and a small number of them get a tiny amount.

            I just looked for some statistics and two sites indicated that roughly 40% of households of people ages 51-64 have no savings and another 20% have between $1k and $50k of savings.

            “Savings” is a slippery concept. Look at net worth by age of householder.

            I agree with you that the funding system as it stands is regressive. I disagree about the political viability of the alternatives you favor and about the impact of the current system with all its faults.

            I don’t disagree that there is some anti-poverty benefits to the program. Any program that huge is bound to have some. The military-industrial complex (same order of magnitude) undoubtedly pulls some Americans out of poverty. But it isn’t just a little inefficient and maybe could be better it is wildly, grotesquely inefficient as a anti-poverty program and is regressive both on the funding side (FICA) and on the payout formula side.

            Status quo bias is the only thing it has going for it from a left of center point of view. But apparently that’s incredibly powerful.

          • and is regressive both on the funding side (FICA) and on the payout formula side.

            That’s a somewhat odd way of putting it. The formula for funding and paying out is, looking only at that part of it, a better deal for poor than rich (putting aside the further complications I mentioned elsewhere). Presumably you would describe a system that took ten dollars from everyone and then gave back ten dollars to everyone as regressive.

          • Brad says:

            Regressive is generally defined in percentile terms, not absolute dollar terms. A head tax is certainly a regressive tax and I don’t think that’s an odd way of putting it.

            Calling a spending program regressive, on the other hand, is I admit a little unusual. But I think a program that pays richer people more money than poorer people straightforwardly fits into any reasonable definition. SNAP or the ACA subsidy would examples of spending programs that are progressive — the poorer a person is the more they get (at least up to a floor in the latter case).

          • Regressive is generally defined in percentile terms, not absolute dollar terms. A head tax is certainly a regressive tax and I don’t think that’s an odd way of putting it.

            But social security is completely different from government programs where the funding is divorced from the payouts. Social security is essentially an involuntary pension plan. People get paid based on what they paid in, except the poor are subsidized to certain degree. David Friedman pointed out that the rich live longer than the poor, which offsets the subsidy, so the net subsidy might go the other way. But it isn’t clear which way the subsidy goes.

            When you go to the grocery store, the poor pay the same amount as the rich for the same food, not a percentage of their income. But I would find it odd to call this arrangement of paying for food “regressive.” In the same way I would call it odd to label social security “regressive.”

            If we need to have an involuntary pension plan for the masses because most are too dumb to save for themselves (a dubious but possible proposition), then I think we need to phase out the current payout method and phase in a defined benefit plan, where each person “owns” the funds that will eventually fund their retirement ( I put “owns” in quotes because this is an involuntary plan, so they don’t have full ownership rights).

            Of course the US will have to build up some capital so that there are actually funds out there to be owned. Some of it might be available if the country sold off a bunch of the land it owned, but that’s a segue. As well as making benefits more fair, it would also make explicit the true net worth of the poor and middle class, instead of the implicit amount of social security each person is due to earn. This would improve analysis of wealth by quintile, so that books like Piketty’s couldn’t exclude this wealth.

          • Brad says:

            @Mark V Anderson

            But social security is completely different from government programs where the funding is divorced from the payouts. Social security is essentially an involuntary pension plan. People get paid based on what they paid in, except the poor are subsidized to certain degree.

            I acknowledge that this is a common way of characterizing but I disagree that it is accurate even with the essentially.

            It is marketed that way, as well as as insurance, as well as a poverty elimination program — the basic marketing plan seems to have been throw everything against the wall and see what sticks. But when you get right down to it is two programs: a tax program and a spending program and they are only loosely connected.

            Like all taxes paying FICA isn’t a choice you make and, as you go into further in your comment, paying it doesn’t actually entitle you to anything. Your property rights in that money are extinguished and you get no title, even unvested or contingently, to any property in return. Thus FICA is just another tax that can and should be analyzed the same as any other tax, not compared to purchasing food in a grocery store.

            Likewise the social security welfare program is just another government spending program. Yes, the payout formula has as one of many factors the amount paid in, but there isn’t any linear or even injective relationship between the two. Some people pay in tons of money and receive nothing. Some people pay in nothing and receive lots. Some people pay in moderate amounts but collect under a formula that would have outputted the same number had they paid in nothing. So it makes no sense to analyze it as any kind of savings program forced or otherwise, and instead it should just be seen as a very bizarre and inefficient welfare program loosely tied to a highly regressive tax scheme.

            If we were to ever put in place a mandatory retirement savings scheme, I disagree partly with you as to what it should look like. Defined benefit schemes are collapsing all over the world. Better to have a simple mandatory set aside scheme with property rights maintained in contributions. If so desired by the worker and/or required by Congress those contributions can be invested in an actuarially sound deferred annuity with the standard right of return of principle and interest if the principal dies before the annuitization date. But that would not be defined benefit as the market, rather the government, would determine what kind of deferred annuity could be purchased at a given price in a given year, albeit under the influence of insurance regulators.

          • Calling a spending program regressive, on the other hand, is I admit a little unusual. But I think a program that pays richer people more money than poorer people straightforwardly fits into any reasonable definition.

            You have just told us that the definition on the taxing side is in terms of percent of income not dollars. So the reasonable definition on the spending side would be the same.

            Your way of doing it switches the definition in the middle, to let you describe a program whose net effect is to transfer from the rich to the poor as regressive.

          • pontifex says:

            Social Security was and is marketed to the American public as an involuntary pension program. That’s the reason why you pay more, and save more, when your income is higher.

            Yes, it is technically true that the courts have ruled that Social Security benefits are not legally considered a property right (see Flemming v. Nestor). But so what? There’s also no legal requirement that the government not hyper-inflate the currency and leave the economy in the toilet. It is a moral obligation, even if it is not a legal one.

            Likewise the social security welfare program is just another government spending program. Yes, the payout formula has as one of many factors the amount paid in, but there isn’t any linear or even injective relationship between the two.

            Some people pay in tons of money and receive nothing. Some people pay in nothing and receive lots. Some people pay in moderate amounts but collect under a formula that would have outputted the same number had they paid in nothing.

            This is true for every pension plan, though. Some people work for a big company with a generous defined-benefit pension plan their whole lives– and then have a heart attack before they ever collect a dime.

            It’s also common for defined-benefit pension plans to be extremely generous to people who joined the plan late in life. For example, in many school districts you can become a school teacher at 50 and then retire with full benefits in a few years. Surely that is not that fair to the people who started teaching at 25. But I don’t hear you arguing that those pension plans should be called something other than “pensions.”

            If we were to ever put in place a mandatory retirement savings scheme, I disagree partly with you as to what it should look like. Defined benefit schemes are collapsing all over the world. Better to have a simple mandatory set aside scheme with property rights maintained in contributions.

            Social Security already is a mandatory retirement savings program. If I can’t trust politicians not to raid the Social Security money, how can I trust them not to raid the extra-special-really-a-pension-this-time-guys money?

            The argument you’re making is incoherent. On one hand, you want Social Security to soak the rich, to be more Social Justicey. On the other hand, you complain that Social Security lets people get a free ride by starting work late in life and pulling out more than they’re “entitled” to based on the amount of work they did.

          • cassander says:

            @pontifex

            Social Security already is a mandatory retirement savings program.

            No, it pretends to be that, by your own admission. Pensions have a real fund with real assets in it them, SS does not.

            If I can’t trust politicians not to raid the Social Security money, how can I trust them not to raid the extra-special-really-a-pension-this-time-guys money?

            Again, there’s nothing to raid with SS, that’s a talking point. You could, however, set up SS to have individual retirement accounts with actual property rights rights and actual assets in them, it would be a VERY different thing if the government started raiding them.

          • pontifex says:

            Pensions have a real fund with real assets in it them, SS does not.

            Social Security has a real fund with real assets in it. Our brilliant political caste has not managed it very responsibly. Hard to believe, I know. 🙂

            You could, however, set up SS to have individual retirement accounts with actual property rights rights and actual assets in them, it would be a VERY different thing if the government started raiding them.

            Charmingly naive.

          • Again, there’s nothing to raid with SS, that’s a talking point.

            That’s a little tricky. The SS trust fund has assets–but those assets are government IOU’s. Raiding them would mean failing to honor them. That doesn’t give the government money, but it relieves the government from having to pay out money.

          • cassander says:

            @pontifex says:

            Social Security has a real fund with real assets in it. Our brilliant political caste has not managed it very responsibly. Hard to believe, I know. 🙂

            To elaborate on what Davidfriedman said, SS’s assets are kept entirely in Tbills. Tbills are real assets for everyone in the world EXCEPT the US government, because no one can invest in their own debt. Doing that is like having your right pocket write a million dollar promissory note to your left pocket, then claiming you own a million dollars in assets.

            Charmingly naive.

            How is pointing out that this sort of behavior is only undertaken as a desperate bid by a quasi-bankrupt governments after years of financial mismanagement disproving my point? I did not say that looting such funds is impossible, I said it would be harder than it is to divert funds that are entirely owned and controlled by the government. And since the US government is currently diverting SS revenue, and is not trying to nationalize all the pension funds in the country, this assertion seems almost tautological.

          • pontifex says:

            That’s a little tricky. The SS trust fund has assets–but those assets are government IOU’s. Raiding them would mean failing to honor them. That doesn’t give the government money, but it relieves the government from having to pay out money.

            Interesting. I wasn’t aware that the SS trust fund was all T-bills. I agree that the government owing money to itself is a little different than the government owing money to others. It’s not entirely clear what this means in practice, though.

            How is pointing out that this sort of behavior is only undertaken as a desperate bid by a quasi-bankrupt governments after years of financial mismanagement disproving my point? I did not say that looting such funds is impossible, I said it would be harder than it is to divert funds that are entirely owned and controlled by the government. And since the US government is currently diverting SS revenue, and is not trying to nationalize all the pension funds in the country, this assertion seems almost tautological.

            My point is simply that all saving schemes rely on you to trust the people in power not to grab the assets that you are saving. Just because SS lacks some of the legal protections that we would like doesn’t make it totally worthless. Just because other pension funds have those protections doesn’t mean they won’t be raided.

            How many people got to enjoy their pension funds in countries where the communists took power after WWII? Those savings were confiscated, in the name of social justice (under the assumption that it is unfair for some people to have more than other people.)

        • Randy M says:

          Pretty sure I’m with Brad on this, although I don’t know what I have to add. It’ll be interesting to see how discussion of SS changes as the demographic profile changes (to speak coldly of lots of people dying, but everyone goes eventually). Age is correlated with likelihood to vote, so this is a case where even though the tax is quite noticeable, the beneficiaries hold the clout.

      • Whether or not redistribution is wrong, your “cat food level” rhetoric wildly underestimates how rich modern societies are.

        To convince yourself of that, try calculating the actual cost of a minimal cost full nutrition diet made from human foods. As a first cut, have the diet be all bread made from flour bought at Costco. That’s an oversimplification, because you will need other things to get all the nutrients and some of them might cost more per calorie, and it assumes access to an oven, but it’s a calculation you can do for yourself in something under half an hour online.

        I predict that the cost will turn out to be a small fraction of what Social Security pays.

        The results of a more detailed calculation of a minimal cost full nutrition diet.

        • Brad says:

          My sister receives about $200/month in SNAP (food stamps). That represents the overwhelming majority of the food she eats in a month (a handful of times a month someone takes her out to eat at a restaurant). Not only does that afford her sufficient calories, her diet includes meat, fish, and fresh fruits and vegetables. I know because I’ve taken her food shopping.

          The contemporary supermarket is truly a marvel.

          • By the calculation I just added a link to above, $200/month is about four times what the minimal cost full nutrition diet costs.

            The original version of this calculation, incidentally, was by George Stigler and is famous as an early example of constrained maximization.

          • A Definite Beta Guy says:

            $200/month? Is that for a single person or a family? That’s incredible. $200/month is the average spend for a single woman, per the USDA numbers. A “thrifty” plan is $40 less.

            My wife and I tend to spend around $600-$650/month, but that’s inclusive of non-grocery items like paper towels, drain cleaner, etc.

          • Evan Þ says:

            My roommate and I (who have a shared grocery budget) generally spend ~$200-250 a month between the two of us. We don’t feel it’s a tight budget at all.

          • albatross11 says:

            Mark Atwood:

            One thing that strikes me about your comment: the practical cost of living for someone who knows how to cook at even a basic level is much lower than the practical cost of living for someone who doesn’t.

            More years ago than I like to think about, I was a college student living on his own in an apartment for the first time. I had no idea how to cook anything more complicated than a grilled cheese sandwich. I bought a lot of prepackaged convenience foods because they came with instructions.

            Many years later, I started learning how to cook well enough to make a decent basic set of meals. I’ve tried to make sure my kids know at least that level of cooking now, because it probably cuts your cost of food by 50% and raises the healthiness of your diet by 50%, just to know how to do things more complicated than make a box of mac and cheese according to the directions on the box.

          • Randy M says:

            I assume that is for a single adult. We have a $200 per week budget (including non-food consumables, but excluding gasoline), but that is for five. Of course, some of the five do more of the consuming than others.

          • Lillian says:

            Important detail: It’s pretty much necessary to give people a bit more than they need in terms of SNAP benefits so they have something to trade for all the various necessities not covered by said benefits. People need soap and toilet paper, but you can’t buy those with SNAP, so instead you share some of your food with your friends, and they share some of their soap and toilet paper with you. Yes this is literally welfare fraud, but following the law is often times not a realistic option for people.

          • Brad says:

            I don’t see how that makes sense. Yes you need money to buy toilet paper and laundry detergent, but you also need money to pay rent. And it’s unlikely you are going to be able to trade a gallon of milk for that. I don’t want to say it’s not possible, but I don’t see how one could survive on only SNAP.

        • Matt M says:

          Whether or not redistribution is wrong, your “cat food level” rhetoric wildly underestimates how rich modern societies are.

          And not just how rich society is, but how rich old people in society are specifically.

          Statistically speaking, the elderly are the least likely to end up having to eat cat food. The elderly are, quite literally, the wealthiest among us. Not all of them, to be sure, but on average.

          Edited to improve analogy: Having massive government programs to “prevent the elderly from eating cat food” is like having a massive societal intervention to ensure that obese men do not suffer from unwanted sexual advances. Like yeah, it’s possible that some obese men may suffer from sexual harassment – and that would be bad, and we should in fact discourage it, but come on people. There are definitely higher priorities, and other groups that deserve a greater focus…

          • albatross11 says:

            Matt M:

            One thing worth pointing out: the fact that the elderly are doing so well in our society now is largely a function of the fact that social security and medicare have been extremely successful programs. They worked, and perhaps overshot a bit. (There was also this decades-long real-estate bubble fueled by a lot of different factors that meant that my people in my parents’ generation who could afford to buy houses often ended up with an extra million or so dollars of lifetime wealth that landed in their laps as a side-effect of buying a big house for their family when they were 30 and living in it till they retired. That’s probably not available for most of us in my generation.)

          • Incurian says:

            How do we know it’s worked well? What’s our control group? Or, what’s the rationale for saying it’s worked better than the alternatives? Not meant to be snarky, serious question.

          • albatross11 says:

            We don’t know it worked well compared to the alternatives, but we do know that decades later, the problem we notice is that old people are *richer* than everyone else, rather than *poorer* than everyone else, which suggests that the goal of transferring wealth to old people to keep them from being poor was probably accomplished.

          • Matt M says:

            but we do know that decades later, the problem we notice is that old people are *richer* than everyone else, rather than *poorer* than everyone else, which suggests that the goal of transferring wealth to old people to keep them from being poor was probably accomplished.

            So at what point do we hang the MISSION ACCOMPLISHED banner and consider a change in strategy to address the new problem.

            And by “new” I mean a couple decades old.

          • baconbacon says:

            We don’t know it worked well compared to the alternatives, but we do know that decades later, the problem we notice is that old people are *richer* than everyone else, rather than *poorer* than everyone else, which suggests that the goal of transferring wealth to old people to keep them from being poor was probably accomplished.

            You have assumed that the point of SS was to make seniors richer, and that they were poorer. At least one of the justifications for SS was that it would push experienced and higher payed workers out and open up jobs for the unemployed, and to shrink the labor force driving up wages and inflation. These run counter to the idea that the elderly were poorer than those who were younger.

        • JulieK says:

          Not everyone is capable of making their own bread from scratch. It would be more realistic to calculate the cost of buying bread.

          More importantly, just focusing on food is a bit of a distraction. It’s very common in these discussions to get bogged down on the question of whether poor people are really in danger of starving, or eating cat food. I think there are two reasons for this: it’s a vivid image that arouses sympathy, and it’s a habit left over from an earlier era, when poor families really did spend the majority of their budget on food. Whereas nowadays, it might be more like 30%. If poor people have enough food, but they can’t pay their rent, or their electric bills, they still have a big problem.

          • It’s very common in these discussions to get bogged down on the question of whether poor people are really in danger of starving, or eating cat food.

            And putting the problem in terms of people starving, as the post I responded to did, is either ignorance or deliberately dishonest demagoguery, which is why I was objecting to it.

          • 1soru1 says:

            > And putting the problem in terms of people starving …. deliberately dishonest demagoguery

            Maybe it would be, if the context was a minor tweak to the current arrangements that make society rich, food affordable, etc. It’s hardly clear that it is when the context is radical change to, or abolition of, those mechanisms.

          • Maybe it would be, if the context was a minor tweak to the current arrangements that make society rich, food affordable, etc. It’s hardly clear that it is when the context is radical change to, or abolition of, those mechanisms.

            Perhaps I missed something. I didn’t realize that the context was a proposal to revive Hugo Chavez and put him in charge of the U.S. economy.

          • 1soru1 says:

            It’s really scary that you apparently think, a priori, that literally the _only_ way to break a developed economy is to follow Chavez’s playbook. There is a certain historical pattern that applies to those who think that radical changes cannot possibly have bad outcomes, because those proposing them are not the bad guys…

          • @1soru1:

            Perhaps I was being too subtle. My point was that the reason we have the sort of economy in which even poor people can be fat and “if you don’t do X, Y, and Z old people will be eating cat food” is demagoguery is not because we have a welfare system. It is because we have a market economy plus the benefits of several centuries of technological progress. Looking around the world, it isn’t government redistribution that is the reason that extreme poverty has fallen sharply over recent decades.

            Chavez was the most convenient example of the fact that what brings a modern economy back to starvation as a real threat is not the lack of government intervention but its presence to a degree sufficient to prevent the market from functioning.

            Would you like a larger example? Will thirty million dead be sufficient? How about the fact that, in the decades after Mao’s death, per capita real income in China increased about twenty-fold? That was the transition from a system where the government provided a guarantee that nobody would starve–the “iron rice bowl” system–to one where the market is, on the whole and, as here, with some exceptions, permitted to function.

        • skef says:

          To convince yourself of that, try calculating the actual cost of a minimal cost full nutrition diet made from human foods. As a first cut, have the diet be all bread made from flour bought at Costco. That’s an oversimplification, because you will need other things to get all the nutrients and some of them might cost more per calorie, and it assumes access to an oven, but it’s a calculation you can do for yourself in something under half an hour online.

          I would say that it’s more an oversimplification because: a) Costco has an annual membership fee and b) in a given area will often be located out in the boonies with questionable access to public transportation. Are we assuming this person has a car, and can drive safely?

          So as a first approximation let’s stick with what can be had at the convenience store on the corner.

          • You are welcome to do that calculation. Costco has its prices on the web and the convenience store on the corner probably doesn’t.

            My guess is that most poor elderly people either have a car, as I think a majority of poor households do, or knows someone who will give them an occasional ride. I agree that it is legitimate to add the cost of Costco membership to the calculation.

          • Brad says:

            There are rural poor, but they don’t have a convenience store on the corner or a supermarket. For the urban and suburban poor a taxi back and forth to the supermarket is very likely to pay for itself if the shopping trips are at least every other week or less frequently. For bigger households I suppose refrigerator and freezer space becomes the limiting factor.

          • Garrett says:

            Silly idea, but how about working off of non-Prime prices at Amazon. They’ve historically offered free slow shipping if you are willing to wait about a week for your stuff to arrive. Assuming that someone can build up enough of a food reserve to last those few days, this allows us to handwave away the shipping/driving costs/inconveniences, and have a citable set of products and prices which can be referred to.

      • albatross11 says:

        Moving over to IS instead of OUGHT type questions, I’d say that cutting social security is probably a guaranteed vote-loser, and so is unlikely to be done anytime soon. Even cutting it in the sense of “we’ll keep paying the planned beneficiaries and screw people who are forced to pay into it up until time T out of any benefits” looks like just another way of saying “please vote my party out of congress, the white house, and all the state legislatures for the next couple decades.”

        Also, I don’t actually think Social Security is unsustainable. It’s a bit underfunded in a pretty easy-to-correct way, but it’s entirely sustainable for the forseeable future. On the other hand, as I understand it, Medicare is unsustainable long-term because medical costs keep growing faster than the economy which must be taxed to pay for medicare.

        • cassander says:

          Also, I don’t actually think Social Security is unsustainable. It’s a bit underfunded in a pretty easy-to-correct way, but it’s entirely sustainable for the forseeable future. On the other hand, as I understand it, Medicare is unsustainable long-term because medical costs keep growing faster than the economy which must be taxed to pay for medicare.

          If by sustainable you mean “could be paid for with increased taxes” then even medicare is “sustainable”. The trouble isn’t any one program, it’s all of them put together. you can fix any one, but only by making it harder to fix all the others. You can’t consider each in isolation because, ultimately, they’re all paid for by the same entity, Uncle Sam, and his pockets, while deep, are not bottomless.

        • shakeddown says:

          Can’t republicans do it? They’re already doing all the unpopular policies, might as well add cancelling SS to the list.

          • Matt M says:

            That’s not really Trump’s brand of conservatism. He’s never even pretended to oppose social welfare programs, with the sole exception of Obamacare (which itself is reasonably unpopular)

    • Deiseach says:

      you’d traditionally buy a house on your birthday using the kid’s labor as a down payment

      A week’s worth of the labour of a twenty year old with no completed qualifications? Either that is going to be a dollhouse or the work that a young person without a manual trade but also no academic credentials can do is hugely valuable in this alternate universe world! In which case why bother with a college student only half-way through a degree, just enslave a high school graduate the minute they finish their education and have them working for you for a week/month/year.

      I see the point you want to make, but the way it’s put is not hugely convincing.

  8. Jaskologist says:

    Riffing off this thread on Joan of Arc and the recent Surfing Uncertainty post:

    The interesting part to me of Surfing Uncertainty was about how our brains edit the information we take in to make it conform to our expectations. Obviously a big concern for Rationalists. I notice this kind of thing when I go back and read primary historical documents and compare them with the more widely disseminated accounts. The latter has always been cleansed and stripped of a lot of the rough edges to better fit our modern sensibilities. One type of cleansing in particular has stood out to me recently, because I encountered 2 examples of it in rapid succession.

    Anecdote 1: The life of Phillip K. Dick (short twitter thread at the link). He sure seems like a crazy person, and most of what I’ve read about him takes that view. The part they tend to omit:

    One vision told Philip K Dick his child had an undiagnosed life-threatening hernia, which turned out to be true. His night-time murmurings turned out to be Koine Greek.

    Well… that’s weird.

    Anecdote 2: Julian the Apostate, ~360AD, decided to rebuild the temple in order to invalidate a Christian prophecy. The work was met with violent storms, which then escalated to earthquakes, and finally fireballs bursting out from the foundations. This is attested to by at least 6 writers, including a Pagan historian with no love of Christianity. Generally, the story is omitted entirely when moderns teach about Julian.

    And now that I think to notice them, I can remember supernatural events in many other primary historical documents. Josephus contains at least 2 prophecies that come true. Joan of Arc apparently has some weird stuff going on. I bet if you read through the primary sources from America’s founding you’d see similar.

    You guys, what if miracles happen all the time, and the reason we think they don’t is because we filter them out of data because we think they don’t?

    • Le Maistre Chat says:

      Don’t we filter them out because Hume said so? 😛

      Anecdote 2 seems mendacious of Hume to ignore if he was aware of the 6 writers.

      “[T]here is not to be found, in all history, any miracle attested by a sufficient number of men, of such unquestioned good sense, education, and learning, as to secure us against all delusion in themselves; of such undoubted integrity, as to place them beyond all suspicion of any design to deceive others; of such credit and reputation in the eyes of mankind, as to have a great deal to lose in case of their being detected in any falsehood; and at the same time attesting facts, performed in such a public manner, and in so celebrated a part of the world, as to render the detection unavoidable: All which circumstances are requisite to give us a full assurance in the testimony of men.”

      1. Anyone whose scrolls or codices were copied in perpetuity, or cited by others respected enough to copy in perpetuity, were of such unquestioned education and learning, and “of such credit and reputation in the eyes of mankind, as to have a great deal to lose in case of their being detected in any falsehood.”
      2. Given the corpus of Classical texts, setting “a sufficient number of men” above 6 means you have to be a true, Pyrrhonian skeptic about all such history.
      3. The Jerusalem anecdote was very public and happened in the midst of the Greek-speaking Roman Empire.

    • The best way I can judge miracles is in my own life. I don’t see them happen. Could I be ignoring them? Possibly, but I’d think some other more credulous types would point them out. In my experience, those who do see miracles are the ones who are seeing things they want to see.

      It is very hard to judge very far off events such as the anecdotes you list, so I go by my own life, and greatly discount them. Of course these anecdotes could also simply be extremely unlikely events that become known precisely because they happen so rarely. I think a world without miracles but filled with billions of people will include some really crazy events that appear to be miracles. So a few events that appear miraculous don’t add much evidence that miracles do occur.

    • Conrad Honcho says:

      From the Manual for Marion Devotion:

      Many stories have been told of miracles accomplished though Mary’s intervention. But we must note that some people make it a boast that they are free of such notions. They pride themselves on believing that no other miracles have ever taken place than those recorded in the Sacred Scriptures, looking up on all others as tales and old fables. In this matter it is well to repeat a reasonable observation made by the devout scholar Fr. John Crasset. He notes: “Just as good people easily believe miracles, so are the wicked always ready to ridicule them.” He goes on to say: “It is a weakness to give credit to everything we hear. But on the other hand, the rejection of the possibility of miracles, when they are attested by serious and devout people, suggests one of two attitudes: either lack of faith, because such things are thought to be impossible to God; or presumption, because the credibility of such reliable witnesses is denied.” We normally give credit to a pagan historian from ancient times such as Tacitus or Suetonius. So how can we escape the charge of presumption if we refuse to give credit to writers who are Christian, learned, and proven reliable. St. Peter Canisius says: “If respectable authors have reported an event with reasonable evidence, and it has not been rejected by learned men, and it is also an account that spiritually edifies our neighbor, then there is less danger in believing and accepting it as true than in rejecting it with a disdainful and presumptuous spirit”

      — The Glories of Mary by St. Alphonsus Ligouri

      There are many stories of fires quenched by a devout person flinging their scapular at them (1599 in Salerno, Italy; the Abbey of Git at Toulon in 1639; the Convent of Mont de Piete at Charite sur Loire in 1639, etc). Can I say for sure these believers who claim they threw their scapular at a fire and extinguished it are telling the truth? No, but I wear a green scapular, and if I ever encounter a building on fire or something I’ll throw it at it and report back what happens.

    • Randy M says:

      I bet if you read through the primary sources from America’s founding you’d see similar.

      Consequentially, George Washington was shot at on several occasions and survived intact to later lead the colonial army.

      Sentimentally, two famous rivals, John Adams and Thomas Jefferson, became close friends via correspondence in old age, with Adams dying at 90 on July 4th, 1826, apparently consoled in passing that Jefferson still lived to watch the Republic. Jefferson, though, had passed hours earlier that same day. It’s not exactly a miracle to have two old guys die, but the coincidences are notable and make for a great story.

      • Jaskologist says:

        I did indeed have the Indian Prophecy in mind:

        At dinner time, a sachem recalled in great detail to Washington what transpired from his point of view on the day of Braddock’s Defeat fifteen years earlier. The records do not state the sachem’s nation, but he may have been Shawnee, Mingo, or Delaware. Since the sachem and his fellow warriors were allied with the French, they had intended to kill Washington, who took over command of the British army’s attempt to retreat after Braddock was injured. Multiple shots were fired directly at Washington, but none touched him. Washington recalled to his mother Mary Ball Washington a few days after the battle that “I luckily escap’d with’t a wound, tho’ I had four Bullets through my Coat, and two Horses shot under me.” Convinced that the Great Spirit had preserved Washington, the chief prophesized that “he will become the chief of nations, and a people yet unborn, will hail him as the father of a mighty empire!” According to Custis, who related the story in his Recollections and Private Memoirs of Washington, Washington did not let the stirring words inflate his ego.

    • Jiro says:

      Well… that’s weird.

      So’s using Twitter as a primary source. I’d expect that this is recent enough that there should be a source for this other than a Twitter thread if it’s actually true.

    • Deiseach says:

      the reason we think they don’t is because we filter them out of data because we think they don’t?

      Yes? I mean, I imagine many modern accounts of St Paul (even if including the more liberal/progressive Christians) tend to go along the lines of “Paul probably suffered from epilepsy – the thorn in his flesh he mentions – which explains the falling off his horse on the road to Damascus; he suffered a fit and then the alleged vision of Christ speaking to him can easily be explained in terms of neurological damage”. Though I kind of like the “freak lightning bolt” one myself; a carefully targeted zap! of lightning which didn’t do any lasting damage, how fortunate!

      There’s a cottage industry in psychoanalysing St Paul to explain why he was a racist, sexist homophobe 🙂

      A few years back (in 2013) there was a doozy of a sermon given by the then-Presiding Bishop of the Episcopal Church, the Reverend Katharine Jefferts Schori, about the curing of the possessed slave girl by St Paul, in which she characterised demonic possession as “a gift of spiritual awareness” and the healing by St Paul as an act of spiteful jealousy:

      But Paul is annoyed, perhaps for being put in his place, and he responds by depriving her of her gift of spiritual awareness. Paul can’t abide something he won’t see as beautiful or holy, so he tries to destroy it. It gets him thrown in prison. That’s pretty much where he’s put himself by his own refusal to recognize that she, too, shares in God’s nature, just as much as he does – maybe more so!

      When you get to the stage of saying curing someone possessed by demons and being exploited by their owners as a money-making racket is bad, mmkay? then maybe you should just throw your mitre at it and give up religion altogether?

      • Randy M says:

        Do they consider the implications of from whence Paul would have the power to deprive a believer of a gift of the Spirit? Or is she implying that other gift givers share in God’s nature?

      • Nick says:

        Though I kind of like the “freak lightning bolt” one myself; a carefully targeted zap! of lightning which didn’t do any lasting damage, how fortunate!

        In fairness, not all lightning strikes are fatal. But (setting aside divine intervention) epilepsy is surely a more probable explanation than nonfatal lightning….

      • bean says:

        A few years back (in 2013) there was a doozy of a sermon given by the then-Presiding Bishop of the Episcopal Church, the Reverend Katharine Jefferts Schori, about the curing of the possessed slave girl by St Paul, in which she characterised demonic possession as “a gift of spiritual awareness” and the healing by St Paul as an act of spiteful jealousy

        Wow. Has anyone else suddenly found themselves wondering how hard it would be to form an inquisition?

        • Le Maistre Chat says:

          About as hard as getting the Pope to declare a Crusade, I fear.
          DEUS VULT!

        • Deiseach says:

          Her term of office is up and the Episcopalians have elected a straight African-American guy who is traditionally (for the Episcopalians) religious (don’t worry, he has All The Right Opinions on marriage equality etc.) as her replacement, so that probably tells a lot about the reaction to the First Female Presiding Bishop 🙂

          Things appear to have quietened down considerably under Bishop Curry (he’s even leading a series of ‘revivals’ to promote evangelism!) which is probably good news for TEC but does make it a lot less entertaining for outsiders to watch (I was one such interested outsider observing during the hey-days of the Anglican Communion Wars and the string of lawsuits TEC brought against conservative/orthodox parishes wanting to leave).

          • Le Maistre Chat says:

            Phil Jenkins did some great journalism on the Anglican Communion Wars. I particularly liked his anecdotes about white Southern clergy proudly saying “I’m a Canon of Rwanda” and gay Episcopalians saying Africans “should go back to the jungle instead of monkeying with the church!”

    • maintain says:

      If miracles happened all the time, we would expect to have historical records of many miracles, and then when video cameras were invented, we would also have video of many miracles.

      Instead miracles seem to have stopped as soon as video cameras were invented.

      • Wrong Species says:

        It’s worse than that. Basically the further back in history you go(so the more unreliable reports are), the more miracles there are. Compare the amount of miracles 2000 years ago to 1000 years ago.

        • John Schilling says:

          Not sure I agree that reports become monotonically more unreliable as they recede into the past. If we’re talking about western civilization, 2000 years ago puts you in the Roman Empire, with a bureaucracy and professional historians. 1000 years ago puts you at the tail end of something that might or might not be a Dark Age and you’re lucky if you get a monk to write down a story he heard from a friend of a friend.

          • JulieK says:

            Since the Roman bureaucrats and historians of 2000 years ago don’t mention such miracles, that doesn’t really help your case.

          • The original Mr. X says:

            Since the Roman bureaucrats and historians of 2000 years ago don’t mention such miracles, that doesn’t really help your case.

            Actually, plenty of ancient histories mention miracles, portents, omens, and the like; indeed, it was almost a cliché for historians to describe the portentous events that occurred before a big battle or change of fortune. And then you have stuff like Ammianus’ mysterious balls of fire which stopped Julian rebuilding the temple of Jerusalem.

        • Jaskologist says:

          Citation needed. I know Big Yud likes to claim this, but I note that he never actually ran the numbers on it. Where are your numbers?

          • Wrong Species says:

            What kind of evidence are you expecting?

          • Jaskologist says:

            You made what was clearly an empirical claim: the further back in history you go, the more miracles. That’s a standard numerical claim which we have many statistical tools to handle. So how many miraculous claims were there 3k, 2k, 1.5k, 200, etc years ago? What’s the r value?

        • S_J says:

          If you set your baseline at about 2000 years ago, you see a huge spike of reports for miracles surrounding an itinerant Jewish teacher who grew up in Nazareth.

          It’s not exactly the kind of thing Scott wrote about here, but it does represent a similar statistical problem.

        • JulieK says:

          In the Hebrew Bible, there are relatively few miracles during the era of the Patriarchs, big miracles during the time of the Exodus and the revelation at Sinai, and again few miracles after the lifetime of Joshua. This variation has nothing to do with the availability of recording devices.

  9. Kevin C. says:

    Does anyone else, upon each visit to SSC, go first to the “Comments” register of bans to see if they’ve been banned yet before viewing anything else, or is that just me?

    • Well... says:

      No, I’ve never done that. I’d prefer my “OMG I’m banned” experience to include a lot of confusion and surprise.

      And now to pick up from the last thread about your life worth/dating problems, etc.:

      OK fair enough. I personally think “never say never, damn the statistics” but I won’t push the point (about you personally marrying and having kids) further. But I still think a bunch of my other advice holds (stop reading all-trite websites, avoid the news, etc.) plus I also think you could probably do something to have postiive influence on kids. (Your nieces and nephews? Children of friends? Etc.) How exactly, I don’t know enough about your situation to say.

      BTW when I say influence kids to be the kind of people you like, I don’t mean make them into little edgelords who go around using racial epithets and beating women, I mean influence them to be the kind of people you would want to populate the world. Polite? Bright? Inquisitive? Skeptical? Rational? Self-aware? Respectful of old things? Able to focus on stuff for more than 10 minutes? Etc.

      If you can’t personally interact with any kids for some reason, you can contribute to the efforts of those who do. OK, so you don’t have any money. You clearly are able to sit in front of a computer and write. Maybe you could use that capability somehow?

      • Kevin C. says:

        Your nieces and nephews?

        Both my younger brothers are single and pretty unlikely to ever be married of have kids (hence another reason I’m lacking in virtue, for, as the eldest son, my failure to prevent the family line from dying out.)

        Children of friends?

        Practically nobody in my age/peer group has kids. I have only one friend with kids, my TradCath fellow “edgelord” who has five, and who I consequently talk to maybe once every few months because he’s busy supporting that family of seven.

        I mean influence them to be the kind of people you would want to populate the world. Polite? Bright? Inquisitive? Skeptical? Rational? Self-aware? Respectful of old things? Able to focus on stuff for more than 10 minutes? Etc.

        Isn’t most of that a product of, first, genetics, and second “non-shared environment” i.e. luck/chance? And probably item number one on the list of “kind of people you would want to populate the world” would be, per Albion’s Seed, “Borderer.”

        OK, so you don’t have any money. You clearly are able to sit in front of a computer and write. Maybe you could use that capability somehow?

        I don’t see how.

        Edit: as an aside, but relevant to my mood, it’s the Winter Solstice in Alaska; I should probably be spending more time in front of my SAD light.

        • Evan Þ says:

          I have only one friend with kids, my TradCath fellow “edgelord” who has five, and who I consequently talk to maybe once every few months because he’s busy supporting that family of seven.

          Does he live in the Anchorage area? If so, maybe you could volunteer to help him and his wife with their kids?

          If not, maybe look for some other TradCaths around Anchorage?

        • Anonymous says:

          OK, so you don’t have any money. You clearly are able to sit in front of a computer and write. Maybe you could use that capability somehow?

          I don’t see how.

          Do what Multiheaded did to get money.

          Making money is really easy. You just have to be non-picky about what you do. I’ve painted houses, cleaned construction sites, interpreted/translated in welfare proceedings, translated manuals, mowed lawns, administered websites, demolished bathrooms, operated concrete cutting vehicles, taught computer use and programming, taught a language, procured business equipment, changed car wheels, developed web pages from scratch, operated audio-video in religious functions, led a team of volunteers, developed a government-critical application for a large corporation and cut trees. All for money. My current income comes mostly from renting apartments and scaring animals in the woods.

          >in b4 “but muh welfare”

          Screw welfare. Your attitude should be that eating out of the garbage and sleeping in a tent should be preferable to taking a dime in welfare that requires you to stay in the metaphorical hole.

          • Kevin C. says:

            See my comments about my unemployability here and here on the previous open thread.

            Edit:

            Your attitude should be that eating out of the garbage and sleeping in a tent should be preferable

            Easy to say if you don’t consider Alaskan winters (which kill plenty of homeless people each year), our various recent laws and crackdowns to deal with our homeless problem (read: chronic inebriate problem), and dealing with other homeless people (see chronic inebriates). Not to mention wildlife.

          • Anonymous says:

            @Kevin C.

            I am well-acquainted with you unemployability, via earlier threads on the same topic. The only item of any particular seriousness on it is “how do I keep paying for my anti-psychotics”, which appear to keep you sane enough.

            Even if you lack the will to make the attempt to claw your way out of the hole you’re in through methods that would make further progress easier, by way of gathering experience and repute as a can-do kind of guy, you might avail yourself of Multiheaded’s method of pecuniary acquisition. Taking alms is not immoral. Do you have an e-begging account yet?

          • Kevin C. says:

            @Anonymous

            you might avail yourself of Multiheaded’s method of pecuniary acquisition.

            I don’t know what that is.

            Taking alms is not immoral.

            But does have good odds of triggering that “welfare trap” to render me worse off materially than not taking them.

            Do you have an e-begging account yet?

            No, because I’m not the “starving artist” type that gets donations on those things. (Plus, didn’t Patreon go on a big purge of far-right types, enough to prompt the creation of “Hatreon” as the (laugable) “free speech” alternative?)

          • Evan Þ says:

            @Kevin, why not try writing up your sociopolitical theory in blog posts? You might get some money for it on Patreon – or maybe not, but I think it’s a worthy project in itself. I know I for one would read it and comment.

          • Anonymous says:

            @Kevin C.

            But does have good odds of triggering that “welfare trap” to render me worse off materially than not taking them.

            Why do you put any value on preserving the status quo? It guarantees that your life goals will not be met. Risking everything does, naturally, involve risking losing everything, but at least then you have a chance at achieving what you want.

            Do nothing -> Die childless in poverty
            Do something and fail -> Die childless in poverty
            Do something and succeed -> Reproduce and then die

            No, because I’m not the “starving artist” type that gets donations on those things. (Plus, didn’t Patreon go on a big purge of far-right types, enough to prompt the creation of “Hatreon” as the (laugable) “free speech” alternative?)

            The aforementioned communist isn’t a starving artist, either, but rather a beggar, which is why I called it e-begging. Look, go make a GoFundMe and I’ll give you 10bux. You can even ask Scott to signal-boost your charity case.

          • Mark says:

            Kevin C. – start doing some agriculture/foraging. First of all volunteer for a local gardening club or whatever, get a cash in hand job working on a trawler or something, then when you’ve built up some knowledge and experience, build something out in the wilderness where nobody can find it.
            You get to keep your welfare, gain some experience, and build something cool at the same time.

            The guy here suggested eating out of the garbage – is there any way you can save your welfare money and try to eat off your foraging skills? Building up the bank balance is a good mini-game to play.

            [Probably not best to do those things on your own, at least not at first. Do it in some kind of social context with a community – where I live there is a community of people who find ways to get food for free – like ‘this supermarket throws out perfectly good food at this time” etc. maybe see if there is one of those.]

          • Kevin C. says:

            @Mark

            First of all volunteer for a local gardening club or whatever

            Alaska, remember.

            get a cash in hand job working on a trawler or something

            Lack the physical strength, youth, and skills they look for on those — plus, issues with being out at sea for months versus my benefits… it means I pretty much lose my SSI forever. Not to mention my apartment (and probably most my possessions, with no place to store them without said apartment).

            build something out in the wilderness where nobody can find it.

            Which is rather more expensive than you’d think. Yes, Alaska has a lot of wilderness space, but over 90% is state park land, federal park land, or protected wetlands. What land is purchasable is actually rather expensive. Second, how would I get my medications, then? Particularly when I don’t drive.

            You get to keep your welfare,

            No, I wouldn’t. Because as a condition of receiving SSI for mental issues, I have to have a representative payee, who is the one who actually receives my benefits (in a special representative payee account that they and not I can access), and who controls my funds for me — that’s my mom specifically. But it means frequent contact to cover funds and purchases and, well, dole out my money to me. I can’t really do that if living hundreds of miles away.

            The guy here suggested eating out of the garbage – is there any way you can save your welfare money and try to eat off your foraging skills? Building up the bank balance is a good mini-game to play.

            No, I’m forbidden to save money. Specifically, if my total assets — all bank accounts, stocks, bonds, or other investment vehicles — exceed $2000 in any month, I receive no benefits for that month, and until they drop back below that $2000 limit. After my grandmother died, I had to spend each of the three installments of my inheritance from her estate in the month it was paid out for exactly that reason, to minimize the lack of benefits to just those months. I use the words “welfare trap” for a reason.

          • johan_larson says:

            @Kevin C.

            You’ve gotten a lot of advice from people in this forum. Have you found any of it useful?

          • Mark says:

            I hate it when they punish saving.

          • Kevin C. says:

            @johan_larson

            Actually, someone on SSC did give me a couple of pointers that improved my search for local volunteering activities.

          • Mark says:

            I may be misremembering, but I think that I gave the one good suggestion. And, it was something like, “there are activities a, b, c at this location, at this time, here is a link to them”

            So, advice to Kevin C. advice givers – specificity. He isn’t looking for general ideas.

            If you want to give your advice on work, I would suggest researching Alaskan welfare system and potential job opportunities, first.

          • So, advice to Kevin C. advice givers – specificity.

            I mentioned the possibility of exam grading for pay. Judging by a quick Google, this looks like one starting point for investigating it. One of the options is “scoring at home.”

            From a page about it:

            To qualify for Pearson employment, applicants must have at least a Bachelor’s degree from a university or college located in the United States. It’s also necessary to have a stable phone and Internet connection as well as a quiet environment to work.

            This page is about ways of earning money from home.

          • Orpheus says:

            @Kevin
            If I am not mistaken, one of the frequent commenters here is an accounting wiz (unfortunately I don’t remember who). Maybe you can try to catch his attention and ask if he knows any ways to work around your saving problem?

          • Winter Shaker says:

            one of the frequent commenters here is an accounting wiz

            You’re probably thinking of A Definite Beta Guy. So… paging A Definite Beta Guy, I guess 🙂

          • Anonymous says:

            @A Definite Beta Guy

            This might work better.

          • Kevin C. says:

            @Orpheus,

            And here I thought I was the crazy one. Do the words “Social Security fraud” mean anything to you? How about “under penalty of perjury”? Or, how about “Federal prison time”? Though, of course, it wouldn’t be you facing the last one if I got caught trying to circumvent the law.

            And as for fully legal, non-“fraudulent” loopholes, there simply aren’t any. Do you really think that a random person, however much a “wiz”, could find a significant loophole in a fairly simple rule, that has somehow evaded the notice of the entire Social Security Administation?

          • A1987dM says:

            if my total assets — all bank accounts, stocks, bonds, or other investment vehicles — exceed $2000 in any month, I receive no benefits for that month, and until they drop back below that $2000 limit.

            That’s f***ing insane.

          • Anonymous says:

            @A1987dM

            That’s f***ing insane.

            Yup. At least you can move up from eating out of garbage.

        • Well... says:

          Sounds like your TradCath friend could use some help with his kids. Hint hint.

          Isn’t most of that a product of, first, genetics, and second “non-shared environment” i.e. luck/chance?

          Did any older males you weren’t related to influence you as a kid? It happens. That kind of experience can be very momentous.

          I don’t see how.

          I wonder if there are pen-pal-style mentorship programs. Don’t talk politics with kids (that would be bad even if your politics weren’t verboten) but you can certainly talk about high-level values (e.g. defending the notion of traditionalism or organized religion) or teach kids to think critically, and as an exercise have them try to steelman opposing arguments to whatever PC thing you think opposing would be both eye-opening for them but not call your motives into question. Meanwhile you could also open their eyes to what it’s like to live your life, tell them about your story, and sort of pass on a legacy of some kind in that form.

    • powerfuller says:

      If this is coming from the same place as “my ideal society would reject me as a member,” then I’m glad you enjoy it around here? But seriously, if you’re doing this, perhaps you’re overestimating the amount of animosity you get or would be getting from other quarters well.

      • Kevin C. says:

        then I’m glad you enjoy it around here?

        Not so much “enjoy” as that this is one of the few places that I can even mention my politics without being insta-banned. (And most the rest are the sort of places full of low-quality commenters where I get called a “paid Jewish infiltrator” and “Grima Wormtongue” for criticism and told to “go kill yourself” on a regular basis.)

        As for overestimating hostility, there’s at least one forum, non-political and related to certain hobbies, I post on, under a very separate identity, that has, as a matter of policy, banning not merely the expression of certain views on their page, but simply holding those views; in other words, if they ever connect my identity there with here or elsewhere, I’m perma-banned.

    • johan_larson says:

      What? No. Heck no.

      The last time the admins of a forum let it be known in various ways that some of my comments were near the line or over it, I raised the issue in some public posts. When it became clear their standards were (to state the matter politely) incompatible with my own, I left.

      • Kevin C. says:

        the admins of a forum let it be known in various ways that some of my comments were near the line or over it

        Except as I’ve seen here, Scott tends to go straight to permaban far more often than he gives warnings.

        • bean says:

          Virtually every ban I’ve seen falls into one of two categories:
          1. People who say things blatantly over every line of civility and the rather loose discussion norms. Almost always new people, usually attracted by a major post.
          2. People who have long been dancing close to the edge in terms of making discussion worse, and have been admonished about it. The most recent one is HFA, but Jill/Moon and EK also spring to mind. Scott usually warns them before they get banned, and the rest of us have usually pointed out that they’re at risk, too.

          Actually, there’s a third category:
          3. John Sidles

          • Anonymous says:

            HFA got banned? What’s the story?

          • Nick says:

            He got banned for bringing up again the stuff Scott told him not to. This was after his warning.

          • John Schilling says:

            Extreme and persistent lack of tact in raising discussion topics like “Didn’t the Nazis make a lot of sense about almost everything?” and “Maybe we should not have stupid people breed so much or at least not let them vote?”, basically auditioning for the job of our very own James Damore but dialing it up to seventeen or eighteen, and violating an explicit warning from the Rightful Caliph to keep that sort of thing to the Culture War subreddit.

          • Matt M says:

            I’m still not entirely convinced HFA wasn’t an elaborate troll. Always came across to me as “This is exactly how people who hate ‘rationalists’ see rationalists and would choose to caricature them”

          • achenx says:

            I rarely post here but I lurk pretty thoroughly. I was pretty sure HFA was trolling you all; after a little while I just minimized and skipped over any threads that they started.

            Or maybe not, I don’t know anything. But I agree with the ban from a quality-of-comment-section standard.

          • Nick says:

            I’m actually pretty sure (90% or so) he wasn’t a troll. If he was, it was a very thorough and impressive act, and I have really got to hand it to him.

          • AnonYEmous says:

            HFA seemed like a pretty classic autist, not a troll as such. Still a pretty terrible poster though.

            Kevin C, you aren’t quite that bad, even if you are a bit one-note (and that note not particularly pleasant at that). Keep on keeping on.

    • Deiseach says:

      What, and spoil the surprise and reaction of vindictive malice? 🙂

    • ManyCookies says:

      Dude Scott didn’t ban an already warned poster for going “Scott is just trying to distract from the (((real sexual predators))) in Hollywood”. If you were even on the ban radar, you’d get a warning first.

      • Kevin C. says:

        When was Balrog, the most recently banned commenter, warned? When did Scott warn the fairly-frequent GregQ before permanently banning him? And I don’t recall, say, TenMinute getting any warnings, either.

        • Brad says:

          I don’t like your posts but nothing I’ve seen you post has ever come close to being as flagrantly banworthy as what balrog posted. I can’t imagine what is wrong with a person who would think it was appropriate to make such a post.

        • bean says:

          I’d say that the banning offense in all three cases was being unnecessarily mean to the other side. Balrog was just completely out of line, and GregQ and TenMinute were both much more insulting than could be justified. I admit to not being a huge fan of yours, but I’ve also never seen you insult people. I can’t imagine you foaming at the mouth in anger or cackling about the clever insult you just deployed.

    • Anonymous says:

      That would be pointless. If I got banned, I expect I would very quickly notice – by dint of not being able to post. Nevermind that I receive notifications of replies to my posts, so I’d probably be informed pretty quickly even if I were not posting.

    • Douglas Knight says:

      If you’ve been regularly commenting here for 700 days, then the doomsday argument says that you are checking way too often.

      On a separate note, you should automate repetitive tasks. Checking web pages is a bad habit. My computer checks the “comments” page once a week. (Mainly to notice if he quietly changes the policy again, but also out of morbid curiosity about the bans.)

    • Creutzer says:

      I don’t think you should be worried. You have beliefs and values that many don’t share and may find abhorrent, but you’re not very offensive in your behaviour towards others people directly, which I think is what mostly gets people banned. I think there are at least two reasons for this perception of relative inoffensiveness I get. First, rather than being an asshole to others, you’re basically screwing yourself over by having some of the beliefs and values that you have, which makes you seem, at the very least, non-threatening. And second, my impression is that you are very explicit about these beliefs and values of yours being… just that; and not trying hard to push them on others or presupposing that everyone who doesn’t share them is an idiot.

    • Mark says:

      You don’t do political posts, that I’ve seen. You do posts about the experience of having certain political beliefs.

      I’m not actually too sure, in detail, what those political beliefs are, except that in your ideal society, political agitation and speech would result in harsh punishments.

      So, it makes sense that you’d enjoy fantasising about being banned.

      • Kevin C. says:

        I’m not actually too sure, in detail, what those political beliefs are

        That the political and cultural “progress” of the past 500 years was a mistake that needs rolled back as much as possible, that “equality” is a myth and Confucian-flavored hierarchy the key to orderly human society, that hereditary monarchy checked by an established church and a powerful hereditary aristocracy is a far better system of government than democracy, that “the Enlightenment” is a highly-infectious mind-virus that is utterly destructive to its host humans in the long run, and that what Scott calls the “universal culture” is indeed, in his words, “an alien entity from beyond the void which devoured its summoner and is proceeding to eat the rest of the world.”

        • Rick Hull says:

          How did you arrive at these beliefs? What is the good in life that is satisfied by your vision of society? For example, I’m a defender of “the Enlightenment” on ethical grounds — there is much less room for oppression in the world when individuals are allowed to flourish according to their own values, rather than be subject to some arbitrary ruler’s preferences. It’s not a dichotomy, but liberty is pretty close to a terminal value for me, so the Enlightenment is a huge step in the right direction.

          • Kevin C. says:

            How did you arrive at these beliefs?

            Basically, over the course of my entire life. Unlike most here, my background is dirt-poor, rural-ish, not-religious but otherwise deeply Red Tribe, with a functionally-illiterate high-school-dropout father and a homemaker mother who married right out of high school. My many fights with the Anchorage School District over the course of my public education left me with a deep and abiding hatred of bureaucracy, bureaucrats (I barely made it through OotP because I’ve known to many real-life Umbridges), and particularly bureaucratic diffusion of responsibility. At Caltech, and having significant access to the Internet for the first time, I was reading things like WorldNetDaily — this was when Vox Day was writing for it. Plus, college also gave me my first real first-hand tast of full-blown Blue Tribe culture — and I’ve wanted it utterly destroyed ever since. And lots and lots of reading, old books and new blogs. Aristotle. Xunzi. De Maistre. Nietzsche. Aquinas. Hobbes. The Anti-Federalist Papers. Loyalists like Peter Oliver and Thomas Hutchinson. Thomas Carlyle. Dune. Warhammer 40,000. Moldbug. The dreaded Jim. Hayek. Victor Davis Hanson. Martin van Creveld. The Collapse of American Criminal Justice by William J. Stuntz. Winnifred F. Sullivan’s The Impossibility of Religious Freedom.

            It might just be down to being an autistic introvert with sensory processing issues, but whereas you value liberty, my priority is order. Harmony, predictability. Solid rules, including of etiquette and manners. So that I know what to say and when to say it, and am not foundering “without a net” in the wide open chaos that most call “freedom.” A solid hierarchy which gives rules and “guardrails” to every relationship, and leaves no uncertainty as to who is in charge. A place for everyone, and everyone in their place.
            And I have concentric loyalties: family first, before friends; friends before acquaintances, acquaintances before strangers. Local before state, state before country — I’m an Alaskan before an American. My race and tribe before those outside them. I just don’t care about distant strangers.

            In short, my experience, innate tendencies, cultural background, mindset, and core values are just rather alien from yours (and vice versa). And probably not capable of long-term coexistence.

          • Rick Hull says:

            That makes a good deal of sense and sheds some light on the situation. I agree almost entirely on the concentric concerns. I do see some tension between desire for order and hatred of bureaucracy. In my mind, liberty (and the satisfaction of individual preferences) provides an important check on large power structures misbehaving or acting dysfunctionally. Monarchies can easily go off the rails and it’s not pretty when they do. That government which governs best governs least. Keep governance as local as possible, and resist the temptation to centralize power. Distributed systems fail more often but much less catastrophically. Tolerance of failure and learning from it and how to deal with it is antifragile.

          • Anonymous says:

            @Rick Hull

            Monarchies can easily go off the rails and it’s not pretty when they do.

            I question the implied difference in ease of going off the rails in republics.

            According to de Jouvenel, the monarchic lifecycle looks like this:
            1. Strong landed aristocracy with weak monarch, first among equals at best. Nearly all power is local.
            2. Successive monarchs centralize power, strip privileges off the aristocracy, and install bureaucrats to replace them.
            3. Eventually, the monarch’s power is near-absolute, the aristocrats having no power, and the bureaucrats do the will of the king.
            4. At some point, the bureaucrats realize that they hold all the power, and overthrow the king, establishing a republic, but keep the massive centralization of power, and ramp up the oppression.
            5. After a while, due to various factors – such as lack of personal central authority that can fix systemic issues gone wild – a strongman appears and turns the republic into a dictatorship.

            Details may vary. At any point, but particularly at step 5, the state may collapse due to overall degeneracy and barbarian invasions. Then it’s back to step 1.

          • Kevin C. says:

            @Rick Hull

            I do see some tension between desire for order and hatred of bureaucracy.

            Yes, but the problem of bureaucracy is the lack of personal responsibility. A lord’s decisions belong to him. A bureaucrat portrays their exercises of power, however petty, as just being a cog in a greater machine, and there is no person whom is utimately answerable for the decisions of that machine; the proberbial “buck” stops nowhere.

            That government which governs best governs least. Keep governance as local as possible, and resist the temptation to centralize power.

            First, what Anonymous said. Second, I favor a more feudal monarchy, not the strong centralized monarchies of the Early Modern period. “The mountains are high, and the Emperor is far away,” as the Chinese saying goes. Third, see “Throne, Altar, and freehold.

          • @Kevin C.:

            In considering alternative models, you might think about the Confucian model as it existed in theory, much less in practice. The Emperor controls the kingdom not by force but by example. There is no need of punishment to enforce the rules, because if the Emperor does a good enough job nobody will want to break them. That’s in contrast to the Legalist approach, at least as the Confucianists described it after there were no Legalists left to argue–severe punishment to make violating the rules no longer in the interest of potential violators.

            I speculate, in the chapter on Imperial China in my current book project, that that attitude may help explain the puzzle of the Imperial Exam system. Why make people spend a couple of decades studying the Confucian classics and poetry and the like in order to pass a series of ferociously difficult exams in order to get a position in government service? In order to make sure that all the civil servants, plus the much larger number who had tried hard to get through the exams and failed, had been indoctrinated in the Confucian approach and so would want to act virtuously.

        • Mark says:

          How about the Roman Republic? Is the hereditary monarchy a necessary element?

          • Anonymous says:

            @Mark

            That’s a good point. The RR certainly had a lot of the features Kevin would want.

            The disadvantage is that republics like these have their own lifecycle, they tend to loop faster than monarchies, and the United States is just a late-stage specimen – and Kevin happens to live there. After all, the early post-revolt Thirteen Colonies were hardly the democratic-bureaucratic mess they are now. If I recall my Maine, the United States system was modeled closely after the contemporary British (monarchical) system, with relatively minor tweaks.

  10. Nick says:

    We don’t get nearly enough linguistics around here.

    What’s your favorite weird or obscure word, SSC? How about crazy etymologies? Do you ever get to use the words? I have a certain fondness for catachresis, a rhetorical device I use in speech sometimes—deliberately using a word in an unusual (though sometimes denotatively correct) way. But the word “catachresis” itself can mean anything from the strictly incorrect to strained use to merely more ambiguous terms—while the Greek just means “wrong use.” 😀

    • johan_larson says:

      I have a trio of words that should not be used, because they sound like they mean something quite different.

      nacreous – means pearly, but sounds like it should mean rotten
      restive – means agitated, but sounds like it should mean relaxed
      limpid – means clear, but sounds like it should mean flaccid

      • j1000000 says:

        Great list! Pulchritude is another good one that I’ve heard Steven Pinker mention before — it may not sound like “ugly” but it certainly doesn’t sound like beauty.

        • Randy M says:

          That reminds me of homely, and how I never recall whether it is complimentary or insulting, and, checking google, I understand the confusion.

          1. NORTH AMERICAN (of a person) unattractive in appearance.
          2. BRITISH (of a place or surroundings) simple but cozy and comfortable, as in one’s own home.

          Such great and terrible words make me think of the quote “Do I contradict myself? Very well, then I contradict myself, I am large, I contain multitudes.”

          • dodrian says:

            Having lived on both sides of the pond, many words have caught me out:

            Pants: Trousers or underwear
            Non-plussed: Not surprised or very surprised
            Jumper: A sweater [unisex term], or a small girl’s dress
            Fanny: this line intentionally left blank
            Public school: Public as in publicly funded, or public as in ‘anyone [who can afford it] can attend’
            First floor: zero-indexed ground floor or not
            [edited to add] Fag: cigarette or homophobic slur
            chips: french fries or crisps

            Let’s not get started on the pronunciation of tomato, or the whole turnip/parsnip/swede debacle.

          • Tarpitz says:

            Just in case: if and only if, or in order to be prepared for an eventuality.

        • Lillian says:

          Pulchritude sounds like the kind of clean and orderly the headmistress at the boarding school likes. Not coincidentally, this is nearly always how i see it used.

      • Nick says:

        Oh man, I hate the word limpid. I was just looking up its meaning again a few days ago.

        • The original Mr. X says:

          I don’t know why, but the word always makes me think of slimly water creatures, like lampreys or leeches.

        • Winter Shaker says:

          In Portuguese, ‘limpo/limpa’ means clean. Probably a lot of hastle learning a whole ‘nother language just to make it easier to remember a word in English, but if you can remember just that one word… 🙂

          • Winter Shaker says:

            *Hassle, not hastle. My brain knows the spelling of that word, but apparently my fingers don’t.

      • Careen — Sounds like it should mean “swerve wildly”, but means to turn a boat on its side and scrape the barnacles off.

        Noisesome — sounds like it should mean loud, actually means smelly.

        • Deiseach says:

          Noisesome — sounds like it should mean loud, actually means smelly.

          Because it is spelled noisome? Without the extra “es”? Sorry to be a pedant here!

        • The Nybbler says:

          Careen — Sounds like it should mean “swerve wildly”, but means to turn a boat on its side and scrape the barnacles off.

          In “careening around the corner”, it means to turn a car so fast you induce noticeable body roll (or actually tip it up on two wheels). The error would be to assume it has something to do with the sound.

        • Careen means both:

          1.
          turn (a ship) on its side for cleaning, caulking, or repair.

          (of a ship) tilt; lean over.
          “a heavy flood tide caused my vessel to careen dizzily”

          2.
          North American
          move swiftly and in an uncontrolled way in a specified direction.

    • Well... says:

      About a decade ago my brother pointed out to me that the word “take” has a ridiculously huge number of meanings, way more than I’d considered before then.

      • Nick says:

        A Series of Unfortunate Events book 10 remarks that “set” has the most meanings of any English word.

    • j1000000 says:

      This isn’t nearly as etymologically rich as yours, but: The library in the small town I live in once had a spelling bee as a fundraising event. At one point they asked the word “callipygian,” which apparently refers to the Venus Callipyge statue and means “having well-shaped buttocks.” I’d never heard it before and it has always stuck with me as something David Foster Wallace would’ve used if he were a rapper.

      • The original Mr. X says:

        Since, if my Greek doesn’t fail me, “callipyge” literally means “beautiful buttocks”, I suspect the term callipygian comes direct from callipyge, without the Venus getting in the way.

        • Creutzer says:

          Yes, kallípygos* meaning “having beautiful buttocks” is encountered not exclusively as an epithet for Aphrodite

          It is, indeed, a marvellous word. I’m rather fond of it.

          *It actually has the ending -os even in the feminine form. These bahuvrihi adjectives do that. I don’t know what language “Venus Callipyge” is supposed to be or why it’s got an -e, perhaps it’s latinised Greek and the Romans didn’t know Greek well enough, thinking the feminine had to end in -e.

          • The original Mr. X says:

            *It actually has the ending -os even in the feminine form. These bahuvrihi adjectives do that. I don’t know what language “Venus Callipyge” is supposed to be or why it’s got an -e, perhaps it’s latinised Greek and the Romans didn’t know Greek well enough, thinking the feminine had to end in -e.

            It might be an earlier/later version of Greek thing: the trend in Koine was for irregular forms to become regularised, so it wouldn’t surprise me is -os feminines became replaced by more usual feminine forms.

    • dodrian says:

      It’s not weird or obscure, my favorite word and etymology is ‘cacophony’, which always makes me laugh.

      Similarly, the ‘euphonium’ is the best-named musical instrument.

      • Randy M says:

        Pretty sure cacophony is an example of an onomatopoeia.

        • Kevin C. says:

          Actually, isn’t it from Greek, as a compound of kakós “bad” (as in kákistos “worst”, kakistocracy “rule by the worst”, etc.) and phōnḗ “sound”?

          • Randy M says:

            Convergent evolution, then.
            In other words, yes, I’m technically wrong, which is really the only kind of wrong, but the word does make quite the ruckus when said aloud. (To clarify, I was wrong about the meaning of onomatopoeia, which denotes a causal link between the sound and the etymology, rather than just an associative one)

          • Nick says:

            It is, but I assume Randy was joking that it’s a terrible-sounding word. I can’t say I disagree. 🙂

            Euphony, cacophony’s opposite, which as dodrian alludes to is the origin of an instrument’s name, is also a linguistic term referring to the “smoothing out” of harsh vowel combinations.

        • The original Mr. X says:

          Incidentally, kakos comes from the Proto-Indo-European word meaning excrement, which (in Greek) evolved into a general term of disapprobation. So when you say something is cacophonous, you’re literally saying that it sounds crappy.

          • Well... says:

            Hm. Another word for poop is “kaka”. Maybe kaka came first?

            And I think in Hebrew they call it (transliterating freely here) khaki. (Rhymes more with “rocky” than with “tacky”.)

          • Nick says:

            Apparently caca in various Romance languages comes from the Latin verb caco, -are meaning to defecate. That in turn derives from the same PIE root.

          • Randy M says:

            I have to ask, any connection to cacao?

          • Winter Shaker says:

            Not unless the Aztecs had contact with the Old World a lot earlier than we thought.

          • The original Mr. X says:

            *Apparently caca in various Romance languages comes from the Latin verb caco, -are meaning to defecate. That in turn derives from the same PIE root.

            Incidentally, archaeologists found a graffito in the toilets at Pompeii reading “Apollodorus medicus Titi imperatoris hic bene cacavit” (“Apollodorus, physician to the Emperor Titus, here had a good shit”).

    • Winter Shaker says:

      Not a word in English (and not an obscure one in its own language) but one of my favourite discoveries recently: zuurstof, which is the Dutch word for oxygen, cognate with ‘sour stuff’ in English (but presumably is intended to have the connotations of ‘acid material’) … which lead to me looking up ‘oxygen’, to discover that our Greek-derived term means roughly the same thing.

      Fancy sciencey terms just seem so wonderfully unpretentious in languages that don’t insist on building them out of Greek/Latin roots.

      • Dutch also has “lucifer” for match and “bioskoop” for cinema.

        • Aapje says:

          ‘Bioscoop’, actually, literally meaning: watching life.

          German brothers Skladanowsky developed a movie projector in 1895 which they called the Bioscop. These brothers traveled to The Netherlands and Scandinavia*. It’s quite likely that this was the origin of the name.

          * Google translate tells me that the Swedish word for a movie theater is ‘bio.’

          As for ‘lucifer’, this name actually seems to come from an English inventor, Samuel Jones, who sold his matches under the brand name ‘lucifer.’ This then seemed to have become used as a generic name for matches in Britain, where other manufacturers also sold their matches as ‘lucifers’. A popular 1915 marching song also referred to matches as lucifers. The term eventually fell out of use in Britain. It almost certainly was adopted by the Dutch from the British when it was still popular and retained when the British stopped using it.

          • Winter Shaker says:

            Thus both English and Dutch have fancy Greek-derived words for a building where you go to watch moving pictures, but completely different Greek-derived words. That must be pretty unusual.

    • hlynkacg says:

      My personal favorite would probably be “Anagnorisis”, a moment realization or recognition. It’s typically used but by literature nerds to describe that moment in a story where Oedipus figures out who his mother is, or Bruce Willis realizes he’s dead, but I also use it to describe that feeling when a pattern or relationship suddenly “clicks” or snaps into focus.

    • Not obscure, but I was long ago struck by the fact that “feud” and “feudal” sound as though they are closely connected but in fact are entirely independent in both meaning and etymology.

      • johan_larson says:

        I thought it was interesting that GRRM introduced the word “bannerman” for feudal lords who properly could have been called “vassals.” It would be interesting to know why. Perhaps the word suggests too subservient a relationship for GRRM’s purposes.

        • The original Mr. X says:

          Probably just using slightly different terminology to make his world seem more alien. Kinda like how he uses “sellswords” instead of “mercenaries”.

          • Nornagest says:

            In ten years it’ll probably be the standard fantasy term, just by osmosis. Kind of like how Robert Jordan used “balefire” (an old word which means “bonfire”) for an attack spell in his magic system, and pretty soon everyone else was using it to mean magical flame.

            It does sound cool, I’ll admit.

          • Lillian says:

            It’s not because it makes the world seem more alien, it’s because it makes it seem more Anglo-Saxon. “Banner” and “man” are both Germanic words, while “vassal” comes from the Medieval Latin “vassalus” itself derived from Celtic. “Sell” and “sword” are Germanic words, “mercenary” comes from the Latin word for wages. To English speakers there seems to be an inherent timeless appeal to Anglo-Saxon words which Latin-derived words lack. It’s why for example he still calls his knights “knights” and not some other fancy word like “cavaliers”, and the king’s guards are the “kingsguard” and not the “paladins”. Those changes would make it sound more exotic, but also step away from the Anglo-Saxon roots, so he didn’t do them.

          • actinide meta says:

            I didn’t think this smeerp was original to GRRM, so I looked it up and I was right, it dates back at least to the late 70s. (A Game of Thrones was published in 1996). And if “sell-sword” counts there’s a book using it in 1975.

            I can’t tell you about “bannerman” because it seems to be a relatively common surname; you get a graph of when people with that name were active in politics or something.

          • Douglas Knight says:

            Try the plural, bannermen, to avoid the surname. The usage seems to mean vassal, but the context is always East Asia. (It is used to translate the Japanese term Hatamoto, which sounds like a kind of vassal, at first just knight-ish, later lord-ish. In Chinese context it seems to be about Manchu military organization, not personal loyalty.)

          • actinide meta says:

            Much prettier, thank you!

          • johan_larson says:

            The results may make more sense in context. Here’s the comparison with “halberdiers”.

            Hmm, why was “halberdiers” so popular in the 19th century? Those weapons were way the heck out of date by then. And why is the word on an upswing since about 2000? I must consider how to monetize this discovery.

          • Incurian says:

            Maybe it has something to do with the Vatican.

        • psmith says:

          George Macdonald Fraser uses “bannermen” to refer to some kind of Qing Dynasty military leaders.

          • The Manchu military structure when they conquered China had, if I remember correctly, ten banners, five of Manchu, five of allied Chinese. Those ended up as, in theory, permanent elite military units, in practice more like a permanent minor aristocracy, with special legal rules applying to them.

        • Well... says:

          I thought GoT was basically high-budget garbage but it had a word I just love: grumpkins (as in “grumpkins and snarks” — shorthand for ridiculous mythological critters that only children believe in).

          I’ve appropriated it to refer to my kids.

    • Nick says:

      Another one from me: the range of meanings of “lay” is surprising. I’m familiar with the meaning “to put” and the related noun, I’m familiar with idioms like “lay of the land,” I’m familiar with the ecclesiastical meaning of “not clergy.” But the one that escaped me until recently was “narrative poem,” as in The Lay of Sir Savien Traliard. Better yet, the “narrative poem” meaning is actually a false friendly in the Old French with the “not clergy” meaning—both were lai, but “not clergy” is from the Greek λαός (laos) for “people” while the other is unknown.

      I also have to link Stephen Chrisomalis’ site Phrontistery, which has lots and lots of great words and etymologies as well as speculative coinings.

      • Brad says:

        I was really surprised the other day when I came across someone that didn’t have any idea what “draw lots” meant (in the context of the tied election in Virginia). I don’t know why I should have been so surprised, but I was. Hopefully I wasn’t rude about it.

        In any event, that points out that draw is a pretty versatile word.

        • Nick says:

          (in the context of the tied election in Virginia)

          Does that make it a draw draw? 😀

          • Brad says:

            My guess is that the one meaning proceeds from the other. That is a tie became known as a draw because it meant that it was time to draw lots.

          • Douglas Knight says:

            See here. The origin of a drawn game is unknown. Some suspect from withdrawn. The original use of draw was pull, as in draw a sword or draw water. Pictures are made by drawing a pen. He doesn’t give a date for draw lots, but here he seems to say that it is the sense of pull

      • johan_larson says:

        So you thought the “Lay of Leithian” referred to a seduction, perhaps? 🙂

        • achenx says:

          The Lay of Leithian is subtitled “Release from Bondage”, but we never once get to see her tied up.

          • Deiseach says:

            we never once get to see her tied up

            Tsk, tsk! Reverse your gendered role expectations! Beren is the one who is the damsel in distress needing to be rescued by Lúthien and he gets chained up (in a dungeon full of werewolves) until she throws it down.

          • Evan Þ says:

            I thought it was about Lúthien’s release through death (the real death-of-mortals) from the bondage to the Circles of the World. As Tolkien says in On Fairy-Stories, the fairy-stories of us mortals are about release from death; the human-stories of fairies must be about release from deathlessness.

        • Nick says:

          I wasn’t familiar with that poem, but if it’s really about Beren and Luthien then that sounds plausible to me. 😉

      • S_J says:

        Connected with “lay” as “non-clergy”, is the plural form “laity”, which is also “non-clergy”.

        I’m not sure whether that is also from Greek-via-French, or whether than an invention in English.

        Related: Is “clergy” singular or plural? I know that “Clergyman” or “Clergywoman” are singular. I know that “laity-vs-clergy” is correct usage, but I can’t recall if “lay-vs-clergy” is considered correct.

        • johan_larson says:

          “clergy” is singular, but it refers to the entire class of people who are ministers or officials of a faith. One person from this class is a “clergyman” or a “member of clergy”, if you want a less gender-suggestive form, but definitely not a “clergy”.

        • Nick says:

          “Lay vs clerical” is probably the best for opposed adjectives. Clergy is the collective noun, while cleric is the regular noun. Although use cleric in the wrong circles and the assumption may be the D&D class. 😀

    • Kevin C. says:

      Well, in English I like “invigilator,” “beadle,” and “retromingent.” For words from outside English, German “Backpfeifengesicht” is a favorite for a very useful concept.

      • Well... says:

        One of my personal in-jokes is a sense of certainty that the Germans have a word for any specific concept or experience I can describe. This culminates in “I’m sure the Germans have a word for the certainty one has that the Germans have a word for some concept or experience.”

        • Creutzer says:

          I predict that at some point there will be more English speaker who know the word Backpfeifengesicht than there will German speakers who do. Backpfeife is a very dated word that nobody uses anymore, and the compound is dying out likewise. Fortunately, there are more up-to-date variants (at least in the Souther German speaking regions), so German speakers’ attribute of having a word for this concept is not in danger.

    • Anonymous says:

      “Supererogatory” is pretty much the most obscure word I semi-regularly use.

    • rahien.din says:

      “Pyknolepsy” is the antiquated term for absence seizures. It basically means “density or compaction seizure.”

      The old name for Parkinson disease is “paralysis agitans.”

      The adjective form of “arch” always throws me.

      To me, the word has always seemed to mean aloof and emotionally distant in a slightly arrogant or bloodlessly magesterial or overtly refined manner. Coolly and humorlessly Victorian, sort of.

      But it means practically the opposite : impish, puckish, coyly humorous, roguishly playful

      The word always occurs to me and I have to look it up. every. single. time.

    • The original Mr. X says:

      I can’t find it at the moment, but there’s a Proto-Indo-European word whose descendants mean both “dark” and “bright”. The theory is that the word itself originally meant “burning”, and different daughter languages applied it to either the colour of something that’s been burnt or to the colour of something that’s in the process of being burnt.

    • SamChevre says:

      The word I most miss in English is the Pennsylvania Dutch Freundschaft; in standard German, it seems to be used just to mean “relatives”, but I’m used to it meaning close connections that aren’t family. There’s not a good English equivalent for “the group of people who if something happened to you, they’d know, and if you needed help, you could ask them, and vice versa.”

      • Aapje says:

        The term seems to mean either friendship or relatives, which is a peculiar combination, but I don’t see how it adds something beyond the English words ‘friendship’ and ‘relatives.’

        • SamChevre says:

          I have no idea, beyond the definition you link, of the standard German meaning; PA Dutch has developed independently for 300+ years and the meaning could be different.

          But as I’m used to the meaning of Freundschaft, it includes friends who are “like family”, and people who are “like family” but not exactly friends. (For example, the woman who was my mother’s maid; I haven’t spoken to her in 30 years, but if I were still Mennonite I’d have some obligation to help her children out if I could–it would be utterly typical for my daughter to maid for one of hers.) Those relationships are family-like in three ways: they are effectively permanent, and they are inheritable, and they have a component of obligation.

      • Incurian says:

        Family friends?

      • Well... says:

        That’s kinda interesting. Pennsylvania Dutch a.k.a. Amish live in very tight-knit communities, and they do so deliberately. When everyone in your community is either family or like family, it probably helps to be able to distinguish who actually isn’t family, but in a way that reminds them they’re still very close to you.

        • SamChevre says:

          Oh–that makes me realize I mis-described Freundschaft: I should have said “meaning including close connections that aren’t family”.

          You are right that Plain communities are very tight, but not everyone in your church would be part of your Freundschaft. Your obligations/connections to them would be general, because they are part of your community–“…helfen raten und arbeiten…” Your Freundschaft would be those people to whom your obligations/connections are personal; they might well be in a different church community.

          (This is one reason it’s such a useful word: there’s a whole world structure inherent in it.)

        • Pennsylvania Dutch a.k.a. Amish

          A minor quibble. “Pennsylvania Dutch” is the name of the language, a German dialect that developed in America, spoken by some but not all Amish–the descendants of the earlier wave of immigration. The descendants of the later immigration speak a variant of Schweizdeutch.

      • Douglas Knight says:

        kith

    • Vermillion says:

      I love weird words like a philatelist loves stamps, my favorite probably being defenestration, i.e The Defenestrations of Prague that precipitated the 30 years war.

      • cassander says:

        I love most that there were 2 defenestrations of prague, both of which led to 30 years of religious war. If you’re ever in prague, never throw a catholic priest out a window.

        • johan_larson says:

          Two? Well, that’s why there’s a word for it.

          Now if someone would just coin a word for dumping a body in the Hudson river, we’d be all set.

  11. j1000000 says:

    Why isn’t the speed limit much, much lower everywhere, including highways? If the highway speed limits were 25 MPH, wouldn’t that save tens of thousands of lives every year?

    I understand why America would never vote to drastically lower highway speed limits at this point, I’m just wondering how we originally decided this many deaths was an acceptable level. (Especially since deaths per capita/per mile were so much higher in the earlier days of the automobile.)

    • Well... says:

      Interesting question. I wonder if it has anything to do with the speed of horse-powered travel (as perceived by the traveler) becoming a Schelling point for “slow.”

      I usually cruise on the highway at 60 (in the right lane of course!), though for fuel efficiency rather than safety reasons. But in either case it would be because I value “getting there quickly” less than something else.

    • The Nybbler says:

      Like Sammy Hagar, I can’t drive 55, never mind 25.

      The current rules were developed mostly when life was cheaper and speed was more valued. Nobody explicitly decided on an acceptable level of deaths; instead, speed limits were imposed ad-hoc for a while (especially in small towns along major highways, the purpose not being safety). Then along came Nixon and the 55 speed limit, and it turned out the revealed preference of drivers when they were actually out on the road was to burn the gas, accept the risks (including the increased risk of a ticket), and go faster. Twenty years later the limit was repealed, and now it’s a cautionary tale like Prohibition. Were alcohol or the automobile to be invented today, they’d never be allowed. We’re far too risk averse — I think Positive Train Control, mandated by Congress to stop certain train crashes (like the one in Washington state recently), has a price tag of hundreds of millions per life.

      • skef says:

        I occasionally look at the glowing burner on my stove after I take the tea kettle off it and think “How is it that we’re still allowed to have these?”

        • Well... says:

          Remind me never to go camping with you.

          • skef says:

            Open wood fires seem to be on, rather than past, the regulatory frontier these days. Beach fire pits keep getting removed, and I wouldn’t be surprised if California’s wildfire season this year prompts new campground rules.

            Anyway, my own last burn mishap was with a fireplace implement when I was something like five or seven, so I’ve mostly been keeping it together.

          • Well... says:

            I was more thinking about it in terms of how humans have been living around fire for hundreds of thousands of years. We’ve had fire under our roofs (rooves?) probably for at least tens of thousands of years. The idea that we maybe can’t handle it struck me as funny, although I honestly do see your point.

          • pontifex says:

            In our old age, we will no doubt look back nostalgically to the days when humans were still allowed to drive cars without automated supervision, make fires, and carry pointy objects without a special permit.

        • Zeno of Citium says:

          Do you have an electric stove? If so, amusingly, the safer solution (at least to the issue of the burners remaining hot enough to injure you long after the stove is off) is also a better one for cooking: running a pipe full of flammable gas through your house so you can light it on fire to cook, also known as a gas stove. They have other problems – like, you can leave the gas on without it being on fire and unknowingly suffocate yourself, although several things make that unlikely – but if you turn off the fire and touch the burner itself on a gas stove, you’ll get a superficial burn at worst before you reflexively pull back.

          Anecdotally, everyone I know who’s had an electric stove has burned themselves (not seriously) on coils that were “off”.

          • The still safer (electric) solution is an induction burner–assuming you are not yourself ferromagnetic.

          • Well... says:

            I’ve never burned myself on a stove, gas or electric. (Where “burn” means anything more serious than touching it and pulling back reflexively just in time for your brain to register that it’s still really hot.)

            The worst burn I ever got on or near my fingertips was from a hot drill bit (I was making a wooden toy, with adult supervision). I was 7 and it hasn’t happened since.

            The induction burner stoves are very expensive and might require you to also replace your pots and pans.

          • The induction burner stoves are very expensive and might require you to also replace your pots and pans.

            The latter, plus the fact that I like gadgets and my wife doesn’t, is why our kitchen stove is half induction, half flat surface electric. If she had her druthers, her half would be the old coil electric which she prefers, but that wasn’t an option.

            A pot will work on induction if a magnet sticks to it.

            I’ve burned myself on the edge of an oven repeatedly, when putting something or out and not paying careful enough attention. Off hand I don’t remember having burned myself on a stove top, electric or gas, or a campfire, but I won’t swear it’s never happened.

          • CatCube says:

            I can’t say I’ve ever burned myself on a stovetop. Of course, I’ve not shared a kitchen with anybody so I know if a burner has been used recently and I make a habit of not touching the coils unless I need to, both of which reduce the likelihood of burns.

            For an induction stove, I know that it doesn’t heat directly, but won’t the hot pan heat the surface below it and still give you the possibility of burning yourself on a hot surface?

          • @Catcube:

            In principle that’s possible, but I don’t think I have ever gotten burned that way. An ordinary electric burner has to be substantially hotter than the pot on it in order that enough heat will flow into the pot.

    • Incurian says:

      If I could only do 25 instead of 80, the range of places I could regularly visit would shrink drastically, cargo trucks would take two or three times as long to reach their destinations… there’s a lot of economic activity that only works because distance isn’t as important as it used to be. Think of all the malaria nets we wouldn’t be able to buy!

    • dodrian says:

      I have to drive at least 40 miles if I want to shop anywhere that isn’t groceries or Walmart. Or if I want to see a movie, or eat out anywhere that isn’t BBQ or fast food. It’s 180 miles if I want to visit a proper city, and that’s where many people round here go to see their specialist doctors, and for certain cancer treatments. This ‘ruralness’ is the norm for a significant minority of Americans.

    • John Schilling says:

      The average American adult spends ~300 hours per year driving, per the AAA. If we could completely eliminate traffic-accident fatalities by reducing driving speeds by 10% across the board, this would save ~1E6 QALYs per year from traffic fatalities but cost us ~1E6 QALYs per year from time spent in traffic.

      This assumes that time spent driving in traffic has zero life quality, which is unduly pessimistic, but also counts only drivers and not passengers as suffering this loss, which is unduly optimistic. Since we cannot come close to wholly eliminating traffic accidents with a mere 10% speed reduction, reduced speed limits cannot be justified by first-order utilitarian calculations based on traffic deaths. Indeed, we should on that basis be raising speed limits.

      But fear of traffic deaths is another, possibly more consequential matter. If we double traffic speed and find that this doubles the traffic death rate, then people spend only 150 hours per year on the road and only 0.025% of them die from it, which is probably a net QALY gain except that maybe now they are spending those 150 hours in abject terror, and having nightmares about it and taking many shaky hours recovering after their daily commute and maybe ODing on alcohol and valium instead of dying in fiery car crashes.

      So, reduce speed limits to the point where people aren’t afraid to drive, and don’t go out of your way to make people afraid of driving, and that’s probably as good as it gets. And we’re probably already there.

      • Matt M says:

        But fear of traffic deaths is another, possibly more consequential matter. If we double traffic speed and find that this doubles the traffic death rate, then people spend only 150 hours per year on the road and only 0.025% of them die from it, which is probably a net QALY gain except that maybe now they are spending those 150 hours in abject terror, and having nightmares about it and taking many shaky hours recovering after their daily commute and maybe ODing on alcohol and valium instead of dying in fiery car crashes.

        I’d actually expect something of an opposing effect. People generally:

        1. Assume themselves to be safe drivers, at whatever speed they choose to travel
        2. Assume, therefore, that any risk they have of being in an accident would be entirely due to some other guy being a bad driver
        3. Assume that their awesome driving skill will likely allow them to potentially mitigate and avoid the effects of the other bad drivers on the road

        I think most people consider routine driving, even at high speeds on crowded freeways, to be relatively risk free. That’s why they don’t fear it, and do fear things like terrorist attacks, statistically 1000x less likely to kill you, but seen as random and impossible-to-avoid.

        I predict that a significant increase in the speed limit wouldn’t do much to change that perception.

        • John Schilling says:

          If this were the case, Americans would never feel apprehensive about driving at high speed on Germany’s Autobahns. From the first-person reporting, they quite often are. I think you are overestimating the effect of Dunning-Kruger as applied to driving skills. Most people don’t actually consider themselves to be NASCAR-quality or otherwise infallible drivers, and very much do have a comfort zone outside of which they are, well, very uncomfortable.

          This comfort zone is probably calibrated to the speeds they experience now on the roads they drive on now. Changing that, in either direction, is likely to do more harm than good for the first decade or two.

          • Matt M says:

            The autobahn occupies a unique cultural space in that it’s basically a meme that it’s fast and dangerous and unsafe (even though I don’t think this holds up statistically).

            The more relevant question would be, are people significantly afraid of driving on highways in Montana, where there is no legal speed limit at all? When people cross the border from a 55mph state to a 65mph state does their terror level measurably increase? Because mine sure doesn’t, my reaction is usually just “oh thank goodness, I can get to where I’m going faster now.”

          • bean says:

            The more relevant question would be, are people significantly afraid of driving on highways in Montana, where there is no legal speed limit at all?

            This isn’t true any more. They’ve had speed limits for at least the past 7 years (which is when I first drove there).

          • John Schilling says:

            When people cross the border from a 55mph state to a 65mph state does their terror level measurably increase? Because mine sure doesn’t,

            There are no 55mph states any more, and aside from the CA/OR to NV/ID borders, no place where crossing a state line results in more than a 5 mph change in the speed limit on rural highways. And the changes which do exist, mostly correlate with the average openness and straightness of the various states’ highways, which will be a much stronger influence on perceptions of automotive safety than the posted limits themselves.

          • Aapje says:

            @Matt M

            When people cross the border from a 55mph state to a 65mph state does their terror level measurably increase? Because mine sure doesn’t, my reaction is usually just “oh thank goodness, I can get to where I’m going faster now.”

            That merely suggests that the comfort zone doesn’t immediately end when you go a bit above the maximum speed you are used to, which seems logical.

            For me personally, my terror level rapidly goes up near 135 km/h (84 mp/h). Driving much faster on the autobahn was an interesting mix of exhilaration, terror, frustration (the autobahns I drove on were two-lane, which meant frequent slow traffic across both lanes) and pleasure (at getting somewhere fast).

            PS. I believe that the statistics show that a road that is designed for high speed doesn’t have substantially more accidents than if you have a lower speed limit, but that the consequences of the accidents that do happen are more serious.

        • Nornagest says:

          Driving is safest not when it’s unusually fast or slow, but when it’s maximally predictable to the other drivers on the road. Those Nascar-wannabe types you see weaving between the lanes at 90 MPH in STIs or low-end Audis or BMWs are dangerous, not mainly because of raw speed, but because their driving style’s so different from everyone else (and because people sometimes panic and do something stupid when they see someone coming up on them at 30 MPH relative). Higher speeds make crashes worse when they happen, but if everyone was going that fast we probably wouldn’t see many more crashes than we do (as per the German experience). Conversely, Google self-driving cars get rear-ended a lot when they’re going the speed limit, if that happens to be 15 MPH lower than what people actually drive in some sections of highway.

          tl;dr go the speed of traffic, and use your damn turn signals.

          • Randy M says:

            Raw speed makes crashes worse when they happen, but if everyone was going that fast we probably wouldn’t see many more crashes than we do

            Agreed but as you imply, there would be more fatalities unless the number of accidents actually decreased.

            Which it possibly might, in the hypothetical that the presence or fear of traffic enforcement leads to erratic or inattentive driving, rather than the intended, opposite effect. Doubtful, imo.
            I’d still be in favor of increased highway speeds, though, at least in the empty stretches. I, uh, might have been testing the safety of a 15-20 mph increase in central CA highway speed limits last summer, and anecdotally it seemed to go fine, provided one pays attention to the notices of impending ghost towns.

          • Well... says:

            I would guess that on today’s freeways, if cars were going the same speed they normally go but ALL of them were weaving in and out of lanes–even with full use of indicators–the fact that an individual’s driving style was not too different from those around him would not help reduce his odds of a crash.

            PS. What about the impact of various in-car technologies? I’m kind of shocked at how often I look over and see people obviously looking at their phones while driving. But car radios probably aren’t too much better!

          • Nornagest says:

            I, uh, might have been testing the safety of a 15-20 mph increase in central CA highway speed limits last summer, and anecdotally it seemed to go fine, provided one pays attention to the notices of impending ghost towns.

            Well, there’s not much else to do along most of I-5.

          • Randy M says:

            Well, there’s not much else to do along most of I-5.

            Wyoming is worse. Nothing piled upon nothing for hours. Luckily their legislature’s highway safety committee realized this and set speed limits accordingly.

          • Incurian says:

            You should drive the correct speed, not what the sign says, and you should stay to the right except to pass – people wouldn’t need to weave if idiots didn’t drive slow on the left.

          • Well... says:

            people wouldn’t need to weave if idiots didn’t drive slow on the left.

            This is dangerous, crap thinking. I agree you should always stay to the right except to pass. BUT:

            – Not all who don’t stay to the right are idiots: some are momentarily distracted, some are made nervous by something, some simply never learned this principle. And some are in fact in the act of passing but not at Ludicrous Speed or within hair-width tolerances.

            – There is never an excuse for weaving. You never “need” to do it.

          • Mark Atwood says:

            What he said. You never need to weave, unless you are properly driving an official or emergency services vehicle with emergency lights and siren, which are on. Or maybe if there is either someone giving birth or bleeding to death in your back seat, and even then, your weaving is putting them at greater risk.

            Slow the F down, suppress the type-A “I’m an important person on their way to important meetings” road rage driving habits, and chill. If you can’t do that, you should not be allowed to drive, at all.

          • Incurian says:

            Nope. Just pay attention and we’ll all be fine.

          • The Nybbler says:

            (epistemic status: Klingon)

            Need? Need is thin soup to live life on. Of what purpose is a half-life lived at 55mph, waiting for what seems like interminable periods for the person in front of you to pass the car next to them doing 54.9mph, and then move back over? No. You have not lived until you have, through skillfully placed lane changes, moved through clump after clump of traffic, accelerating through the open road between them. There is no legal way to match the visceral satisfaction of blowing by, on the shoulder, two cars riding side-by-side at 55mph on an otherwise empty Midwestern interstate near the start of a 700-mile journey.

            Important person on the way to a meeting? No, no; the time reduced is important but ultimately secondary, a way of keeping score. Speed itself is its own reward. There is no “chill”, and those who don’t understand that should get the “F” out of the way.

          • CatCube says:

            The best part of the weavers is that if you happen to take the same exit as them, they’re usually like the second car ahead of you at the traffic signal.

            “Glad that all the chaos you caused saved you a whole 2.5 seconds there, Ricky Bobby.”

          • A Definite Beta Guy says:

            How many states have regular “SLOWER TRAFFIC KEEP RIGHT” signs? I see these all the time, and I assume they are in most populated locations. Slower drivers who do not keep right are, IME, committing acts of willful malice, not ignorance.

            I rarely see people driving slow in the left-hand lane. Typically someone is puttering at or below the speed limit in one of the center lanes, which forces everyone else to drive around them. Perhaps drivers don’t realize that they are impeding traffic flow around them by driving slowly in the middle lane.

            The difference between 55 and 75 is a big difference to us Midwesterners driving hundreds of miles to visit family or whatever. It’s not 25 seconds at a traffic light.

          • Gobbobobble says:

            There is no legal way to match the visceral satisfaction of blowing by, on the shoulder, two cars riding side-by-side at 55mph on an otherwise empty Midwestern interstate near the start of a 700-mile journey.

            Except, perhaps, the visceral satisfaction of blocking the shoulder in a traffic jam and watching the douchemobiles in the rearview go into conniptions. Hard to pull off since it generally requires a confederate to save your space in the lane in case of emergency vehicles, but when it works out…

          • bean says:

            @The Nybbler
            I’m pretty much with you. People passing slowly really annoys me. You have an accelerator pedal. Use it.
            (The worst are semi trucks overtaking with a .5 mph speed difference uphill, and no speed difference at all downhill.)

            @ADBG

            How many states have regular “SLOWER TRAFFIC KEEP RIGHT” signs? I see these all the time, and I assume they are in most populated locations. Slower drivers who do not keep right are, IME, committing acts of willful malice, not ignorance.

            Never saw them in California. Didn’t see them in Arizona, New Mexico, or Missouri. See them all the time in Oklahoma, and Texas has something similar.
            Actually, OK passed a law that took effect in November making being in the left lane while not passing a crime of some sort. Doesn’t seem to have done much. Lane discipline is OK, but not great.

          • Matt M says:

            I rarely see people driving slow in the left-hand lane. Typically someone is puttering at or below the speed limit in one of the center lanes, which forces everyone else to drive around them. Perhaps drivers don’t realize that they are impeding traffic flow around them by driving slowly in the middle lane.

            Right, people who think “I’m not in the far left lane therefore I can go at whatever speed I want” are the problem. The lanes should be traveling such that the fastest are on the left, and the slowest on the right. If you notice that people are gaining on you from behind and then passing you (on either side) then you’re going too slow and need to move right. Period.

          • Incurian says:

            There is no legal way to match the visceral satisfaction of blowing by, on the shoulder, two cars riding side-by-side at 55mph on an otherwise empty Midwestern interstate near the start of a 700-mile journey.

            I think I’m in love.

          • Nornagest says:

            I’ve seen “Slower Traffic Keep Right” signs in California, but only for the passing lanes you see on the rare straight and flat sections of twisty mountain roads. They seem more common in Oregon. Never seen one in Washington or Nevada.

            And oh man, I hate semi trucks passing each other when there’s only two lanes in each direction. I used to drive to work along a highway that was also used by a chemical plant and a couple of quarries, and there was one stretch of road where I’d consistently have to slow from 70 to 35 because there was always some trucker trying to pass another trucker going uphill.

          • Evan Þ says:

            @Nornagest, I saw several of those signs in Washington when I was crossing the mountains a few weeks ago.

          • CatCube says:

            I should point out that driving on the right and passing on the left has been the law for many, many years, and applies in your state whether or not there are signs to that effect.

            They just don’t pull people over for violating it. The presence or absence of the signs won’t change compliance without enforcement.

            It’s something they *should* pull people over for violating, since an orderly flow of traffic is important for road safety. Same reason they should pull people over for passing on the right.

          • Incurian says:

            It would not be possible to pass on the right if it were full of all the slow cars. In Arkansas, I heard, the law they have is that it’s illegal to be passed from the right.

          • CatCube says:

            @Incurian

            I’d have to look at the exact working of the Arkansas law, but that functionally means that you must keep right except to pass. It also doesn’t necessarily mean that because the guy ahead of you is violating the law that you’re allowed to pass him on the right, either. I agree that if someone won’t get over, yeah, you’re going to have to violate that law. However, if you give it a minute, most people will get over when they see you behind them. If after a minute they haven’t moved over, a flash from your headlights will remind them. Then it’s OK to pass them on the right.

            I’ve seen plenty of trucks (who were passing much slower traffic) get stuck in the left lane because impatient assholes zipped over to the right to pass them well before they had a chance to safely move back over.

            Both people who go the exact speed limit in the left lane and people who zip around from lane to lane in congested traffic are bad drivers, just in different ways.

      • A1987dM says:

        But fear of traffic deaths is another, possibly more consequential matter. If we double traffic speed and find that this doubles the traffic death rate, then people spend only 150 hours per year on the road and only 0.025% of them die from it, which is probably a net QALY gain except that maybe now they are spending those 150 hours in abject terror, and having nightmares about it and taking many shaky hours recovering after their daily commute and maybe ODing on alcohol and valium instead of dying in fiery car crashes.

        There’s no way doubling the actual risk would increase the perceived risk an order of magnitude. If anything I would expect that doubling the actual risk would slightly less than double the perceived risk.

    • AnonYEmous says:

      Why isn’t the speed limit much, much lower everywhere, including highways?

      the much better question is why it isn’t much higher

      it’s not like anyone obeys it anyways. The alternative to this level of deaths is taking forever to get anywhere, and fuck that. The dead will be (not me) anyways because I’m not bad at driving, and I say that not in a meta sense but in a personal sense.

      • Well... says:

        Bad drivers often kill good drivers, especially at high speeds.

        • AnonYEmous says:

          I calculate my odds of getting a bullshit speeding ticket at a lot higher than getting into an accident. I’d rather be able to go reasonably fast, and to hell with bad drivers – I’m good enough to avoid them.

          • Well... says:

            That’s what everyone thinks.

          • Deiseach says:

            I’m good enough to avoid them.

            The problem is that they are not good enough to avoid you.

            So you’re speeding up coming to the lights just before they change so you can get through and you know you’re a good enough driver to judge the time and handle the car to get through the junction without running into anyone. So does the idiot coming from your left think, and he slams right into you as you’re half-way across.

            Or the guy trying to overtake four cars at once against oncoming traffic.

            Or the guy driving too fast in a car that isn’t meant for those speeds. Or in bad weather. Or is drunk but thinks he’s fine to drive. Or insists on driving right on your bumper and so when you need to slam on the brakes for an unexpected stop he runs into the back of you. And so on and on and on.

            You may indeed be as good a driver as you think you are. The question is: are all the other people who think they’re good drivers as good as they think they are?

    • rahien.din says:

      I would echo everyone above.

      According to the interwebs, most accidents seem to occur in neighborhoods, parking lots, and during rush hour commutes.

      Velocity raises the stakes of a collision, or, makes a collision more dangerous.

      But the primary driver* of the probability of collision is inhomogeneity (disparate travel velocities, sudden change in velocity or direction by other drivers, etc.)

      The most effective way to prevent highway driving deaths is to reduce the speed limit to zero. But this fails everyone’s risk-benefit calculations. Any feasible speed limit will be accompanied by a non-zero risk of death, all other things being equal.

      It might be better to figure out what we can do to increase highway safety, independent of or concurrent to maintaining a safe speed limit.

      * Ha!

    • Nabil ad Dajjal says:

      If the highway speed limits were 25 MPH, wouldn’t that save tens of thousands of lives every year?

      de Blasio pls go.

      This is basically his “Vision Zero” plan, and while traffic deaths may be down it’s much harder to actually get anywhere aboveground than it used to be. And that’s for a 5 mph speed reduction in a city where people generally don’t drive to begin with: imagine what would happen to a city like LA.

      Paralyzing the roads will definitely prevent high-speed accidents but at that point you’re throwing the baby out with the bathwater.

  12. johan_larson says:

    Slate has an article about Wesleyan College (a classic SLAC) and its efforts to improve its football team:

    As a Wesleyan undergrad in the 1970s, Michael Roth lived in a co-ed literary fraternity and didn’t typically go to football games. His interest was piqued, though, when he heard that a dean had complained about students using profane language in the stands. Roth went to the next game, spending the afternoon “cursing vigorously.” The school president told me it “was a free speech issue. That’s the kind of place Wesleyan is.”

    Roth didn’t return to his alma mater with a great sports background—prior to Wesleyan, he was the president of the California College of the Arts. Just after he was hired, the football team hit rock bottom, winning a single game in 2008. “We had graduates who worked in the same offices as Williams and Amherst grads,” John Biddiscombe, the athletic director at the time, says. “And they told us they were tired of losing.”

    • j1000000 says:

      Sort of amazing the article waits that long to mention that Bill Belichick went to Wesleyan, and then only mentions in passing his current involvement. (I realize he didn’t play football there.)

      Pretty interesting article, though. I often wonder how explicitly admissions officers and higher-ups talk about such matters behind closed doors, because I suspect even in the most private settings they’re too proper to openly say things like “We need more bros whose families don’t need financial aid and who’ll eventually give us their Wall Street trader money for our endowment.” I’d imagine they’d only ever say things like “football helps foster community” even if they have more cynical underlying aims.

      • johan_larson says:

        I suspect the discussions get very explicit, because admissions decisions require tradeoffs among finances, social justice, and prestige, all three of which are very important to university administrators.

  13. Matt M says:

    Man, even Batman hates me because of my race and gender…

    • Well... says:

      I liked Christian Bale better when he was running back and forth cheering airplanes, and sailing around on pirate ships eating Long John Silver’s.

    • Mark says:

      I’d like to see his working.

      Edit:

      Actually, he says:

      Our culture will be so much richer the day that we stop saying ‘Hey it’s all white dudes who are running things,’

      ‘Stop saying’ – my faith in batman is restored.

      • Matt M says:

        I get your point, but it’s pretty clear that’s not what he means 🙁

        • Philosophisticat says:

          It’s also clear he doesn’t mean that he hates your race and gender. He’s saying that culture would be richer if those in positions of power were more diverse. But by all means, don’t let me get in the way of the local favorite persecution complex.

          • Matt M says:

            He’s saying that my race and gender should disqualify me from positions of power that I would be otherwise well suited for. And that this would make all of us better off. Maybe this isn’t a 1:1 map to “hatred” but it’s pretty clear that if anyone ever said this about literally any other group besides white males they would be denounced as villainous bigots.

          • Quiet Lurker says:

            The problem, as I see it, is that he’s singled out a superficial distinction (skin color).

            If he had said that the entertainment industry would be richer if more of the people in charge were from different cultural, economic and political backgrounds I don’t think this thread would exist.

            Edit: Also, by talking like this he’s contributing to the myth that the principal difference between humans is their race.

          • Philosophisticat says:

            @Matt M

            No, he didn’t say that at all.

            @Quiet Lurker

            I also think that if people read what was said with the slightest bit of comprehension and charity instead of immediately pattern matching to the most extreme bogeyman version of their tribal enemy this thread wouldn’t exist. And yet here we are.

          • Matt M says:

            No, he didn’t say that at all.

            Yes, he did.

            He outlined one goal and one goal only: Fewer white men in positions of power. No other qualifiers.

            He didn’t say, “We need to identify the white men who are unqualified and replace them with minorities who might be better qualified but are only held back because of structural racism.” That wouldn’t, necessarily, be an attack on me, because I consider myself well qualified for whatever position I do, or may in the future, hold.

            If qualifications or ability were relevant to his thought process, he would have mentioned them. He didn’t. Skin color and gender, that’s all that matters to him. And I happen to have the genes and organs that make him want to keep me out of power, completely irregardless of my economic background, cultural background, skills, abilities, experiences, etc.

          • Evan Þ says:

            “And I happen to have the genes and organs that make him want to keep me out of power”

            No. He’s perfectly fine with some white men being in positions of power. He just wants some other people in those positions too.

            On the margins, it’s an attack, but let’s not exaggerate it.

          • Gobbobobble says:

            Skin color and gender, that’s all that matters to him.

            All that his PR team says should matter to his public persona.* Man’s an actor.

          • Chalid says:

            If someone said that academia would be better off if it had more conservatives and fewer liberals, would you be reading them as saying that they hate liberals, or that qualifications or ability were not relevant to their thought process?

          • Matt M says:

            A proposed plan to quickly and immediately remove liberals and replace them with conservatives would strike me as unconcerned with qualifications, yes.

            Isn’t that the left’s usual response when someone points out the imbalanced ratio? That conservatives are simply unqualified to be professors?

            If I were an aspiring liberal academic and saw someone advocating a plan that started and stopped at “fewer liberals and more conservatives” then I would take that as a pretty direct and personal attack, yes.

          • AnonYEmous says:

            If someone said that academia would be better off if it had more conservatives and fewer liberals, would you be reading them as saying that they hate liberals, or that qualifications or ability were not relevant to their thought process?

            I’d read them as saying that they feel as though conservatives have essential qualities that liberals do not. (And, it is to be assumed, the reverse is naturally true.)

            If someone is saying that about a race, then usually racial hatred is assumed (and present). I suppose there are a few benign people who believe that the races think differently but that this is good or at least OK, but even those who exist are usually on the right and are highly unlikely to be advocating against the interests of white men. It’s not always hatred, but it has the same basic effect and sometimes is hatred.

          • Chalid says:

            proposed plan to quickly and immediately remove liberals and replace them with conservatives would strike me as unconcerned with qualifications

            This is in no way analogous to anything Bale said in the article you linked.

          • rlms says:

            @Matt M
            “If I were an aspiring liberal academic and saw someone advocating a plan that started and stopped at “fewer liberals and more conservatives” then I would take that as a pretty direct and personal attack, yes.”
            If that’s so, then you’re oversensitive. I think the obvious interpretation is that the person advocating that plan thinks there is unjust discrimination against conservatives in academia which should be stopped. I would not take that as an attack unless I actually was a irrationally prejudiced against conservatives. Likewise, I don’t think you should take Bale’s statements as an attack unless you are actually a racist.

          • Matt M says:

            I think the obvious interpretation is that the person advocating that plan thinks there is unjust discrimination against conservatives in academia which should be stopped.

            Then why not say that? It’s a lot more palatable and less likely to offend anyone. Also far closer to what you actually mean, assuming that is what you actually mean. If Bale said “There is unfair and unjust discrimination and that should be stopped,” then maybe I might quibble over the existence of discrimination, but I’d agree wholeheartedly that to any extent it exists, it should be stopped.

            If he just says “Fewer white men plz” then that, in and of itself, does not address discrimination at all. It implies that he supports the goal of reducing white men with access to power whether discrimination is actually happening or not.

            I think the left needs to realize that “discrimination is unfair and must be stopped” and “diversity is our strength” aren’t complimentary arguments, rather, they are mutually exclusive. Either you favor fairness, OR you favor diversity for its own sake. You have to pick one, and you can’t say one and then imply you mean the other when someone calls you out for preaching a blatantly racist message.

            Likewise, I don’t think you should take Bale’s statements as an attack unless you are actually a racist.

            I am a white male who has decided that I, too, can participate in identity politics (and that I probably have to in order to survive). Does that make me a racist?

          • Chalid says:

            I think the obvious interpretation is that the person advocating that plan thinks there is unjust discrimination against conservatives in academia which should be stopped.

            Agreed that this is the most obvious interpretation. The second most obvious interpretation is that in some fields such as media (and parts of academia), perspectives become more valuable in proportion to their rarity. “I hate white dudes” is a very distant third.

          • herbert herberson says:

            Christian Bale works in a field where the ratio of people qualified to do the job compared to the jobs available is exceptionally high.

            I’d bet good money that when he says “we should have fewer white men” instead of “we should boot unqualified white men in favor of qualified minorities” it’s because he doesn’t really think there are a significant number of unqualified white men out there–just that the ratios of how many perfectly qualified people get screwed are not as they should be.

            (also, what effectively makes an actor “qualified” is often totally divorced from the actual skill of acting; see, e.g., the many many action/blockbuster stars who pretty much always just play slightly different versions of themselves)

          • anonymousskimmer says:

            @Matt M

            He outlined one goal and one goal only: Fewer white men in positions of power. No other qualifiers.

            He didn’t say that. His exact quote is actually in the article. A charitable interpretation is: More positions of power + same number of white men in positions of power = greater diversity in power.

  14. I would like to explore the facts of corporate taxation in an international context, in the hopes that someone who knows more about it than I do can correct my errors and fill in some of the holes.

    The situation as I understand it, prior to the new tax bill, was that most countries taxed income on the basis of where it was earned, the U.S. taxed it on the basis of where the corporation earning the income was incorporated. To avoid double taxation, the U.S. credited tax paid to other countries by U.S. corporations against U.S. corporate taxes.

    Imagine, for simplicity, that France taxes income (actually profits) earned in France at 20%, the U.S. taxes income by U.S. corporations at 30%. A U.S. corporation makes $100 in France. It owes France $20 and the U.S. $10.

    One result, given that the U.S. corporate tax rate was considerably higher than that in other developed countries, was to make it in the interest of corporations to be incorporated outside the U.S. Another was that, if a corporation was for some reason incorporated in the U.S., it was in its interest to create a subsidiary incorporated elsewhere and funnel as much of its profit as possible to that subsidiary.

    Suppose the U.S. now lowers its rate to 20%. There is no longer a cost, other than some additional paperwork, to being incorporated in the U.S., so more companies choose U.S. incorporation. But this doesn’t bring in any extra tax income, because the foreign operations of those countries get a credit for taxes paid abroad that cancels their U.S. liability.

    One thing I don’t know is how the U.S. treats income earned in the U.S. by foreign corporations. Logically one would think it would be untaxed, but that would leave foreign corporations operating in the U.S. with a big advantage over U.S. corporations operating in the U.S. which strikes me as unlikely, so there is probably some mechanism to tax it.

    Do I have it right so far? If so, what are the advantages to the U.S. of making it more attractive for corporations to be incorporated in the U.S., assuming our corporate tax rate is now similar to that of other developed countries?

    • Steven J says:

      The US taxes the income earned by foreign corporations from US activities:
      26 US Code Section 882
      https://www.law.cornell.edu/uscode/text/26/882

    • SamChevre says:

      I am not an expert: real experts on this make a lot more than I do. But there’s a key point you are missing by thinking of France and the US: that’s the subsidiaries designed to capture income in a low-tax jurisdiction–by definition, they have few tax credits to apply. There’s a fair bit of income that might end up in the US and might not (how much US income does an iPhone made in China and sold in a Malaysian airport generate?), and it’s significantly more likely to end up in the country of incorporation eventually than to end up elsewhere: more will end up in the country of incorporation and more quickly, all else equal, if the tax rate is lower.

    • I am a US corporate tax accountant, so I think I know most of the answers to your questions.

      1) For US corporations, all income is taxed in the US, no matter where in the world the income is earned. As you say, the foreign tax credit is supposed to eliminate double taxation, as your example indicates. Although it is also true that the foreign tax credit rules can be very complicated, and sometimes don’t mitigate double taxation very well.

      2) The US of course does not have the right to do this for non-US corporations. As Steven pointed out, foreign corporations are taxed in the US to extent they have earnings in the US. If the corporation mostly does business in the US, then it doesn’t help to be incorporated elsewhere. But if business is done overseas, then the high US rate can be avoided by foreign incorporation. In your example above, if the $100 income earned in France was earned by a French company, then only the $20 tax would be owed.

      3) Thus you are correct that US corporations will create foreign subsidiaries partly for tax reasons. There is so-called subpart F income, where the US parent must pay tax on certain kinds of income earned by its foreign subs, but in my experience, that is usually only 10-30% of most foreign subs’ income of large US companies.

      4) When the foreign sub pays a dividend to the US parent, the earnings represented by those dividends are then subject to US taxes. Thus you hear about the high tax rates in the US result in US corps. not being able to remit earnings back to the US without a tax penalty. That’s why they had a temporary rule about 10-15 years ago, that allowed US corps to pay tax on only 15% of these remitted dividends as long as these dividends were spent in the US or some such. IF you had a permanent rule like that, US corps would bring back much more cash, but the Feds would collect less tax, so it was only temporary.

      5) I have heard that other countries have territorial rules, so that income outside one’s home country is totally free of tax from the home country. But when I’ve done research on this, it appears to be much more complicated than that, so I don’t really understand how other countries’ rules work. I really only know US rules.

      International tax is extremely complicated, so I’ve only scratched the surface. And I do not specialize in international, so there are many details I don’t understand. From what I’ve heard of the new GOP law just passed, it sounds like they are making the international rules even more complicated. That’s a very bad thing IMO. And I don’t think it will help revenues either, because as the rules get more complicated, the difference in pay scale between IRS lawyers and corporate lawyers becomes more significant. The big firms will eat the IRS’s lunch.

      • Incurian says:

        The big firms will eat the IRS’s lunch.

        This sounds like a good thing. What I’d be worried about is the small companies without a harem of lawyers and accountants.

  15. rlms says:

    Possibly the weirdest consequence yet of the Bitcoin bubble: Long Island Iced Tea renames itself to Long Blockchain.

    • A1987dM says:

      I saw that on my Facebook feed and I didn’t even follow the link because I assumed it was satire. OMG.

    • Randy M says:

      0.o That seems pretty close to fraud.

      • Deiseach says:

        Seems awfully like a scam. How do you switch from selling soft drinks to brokering a cryptocurrency? Sounds like the first business was failing and dying, so they shifted to “A fool and his money are soon parted” mode and that seems to be bearing fruit.

        I wonder if the Stock Market Crash of 1929 was preceded by the same kind of signs? Smiths Widgets is doing nothing, so they change the name to Smiths Gold Mines and the man on the street can’t buy the stock fast enough and in large enough quantities? This is honestly the kind of thing that makes me think Bitcoin is due to crash for real this time.

    • Matt M says:

      On the one hand, this is pretty clearly a scam.

      On the other hand, when idiots fall for a transparently obvious scam, I find it hard to sympathize with the idiots, and my main reaction is something like “I wish I would have thought of this first.”

  16. Le Maistre Chat says:

    I’m in an airport waiting for my flight to Cancun, paid for by family. Here’s hoping I can find a Latin Christmas Mass when I get there, seeing as the vernacular isn’t mine. 🙂

    • Nick says:

      I think you’d have better luck just finding a missal in their vernacular. 😛 There’s also no doubt an app, but using one’s phone in church is not recommended.

    • The original Mr. X says:

      It’s sadly ironic that the Church abandoned the universal language of Mass at just the time when frequent international travel would have made a universal language more useful.

  17. Mark says:

    I would guess that many banned posters return to post under a different name, as Sidles does.

    When we punish Sidles, we aren’t standing up for principle or obedience. We are punishing the decision to write in a distinctive style.

    It’s punishing a form of honesty.

    So, perhaps, what we want is conformity to our fantasy. The appearance is more important than the reality.

    • Aapje says:

      He got banned for that posting style, so it seems quite reasonable to punish him for coming back without changing that style, but not others who change their style to one that is more acceptable.

      • Mark says:

        He wasn’t banned for his current writing style, he was banned for use of bold and for using too many bullet points.

        I mean that was bad writing in the technical sense- like if I posted everything in font-size 1 or something.
        Or posted vast quantities of text like this.

        Sidles hunting serves a purpose as a useful fantasy.

        If you can pretend that you knocked the policeman’s hat off by accident, that’s almost as good as it actually being an accident.

        • Whatever Happened To Anonymous says:

          He wasn’t banned for his current writing style, he was banned for use of bold and for using too many bullet points.

          I mean, only in the most literal interpretation. The intent of the ban was to ban his overall writing style, of which bold and bullet points were just the most egregious.

    • Brad says:

      I would guess that the moon is made of green cheese.

    • Nick says:

      “Punishing a form of honesty” is kind of an odd way to put it. Superficially that sounds bad, but Scott’s commenting policy already balances true against necessary and kind.

      • Mark says:

        Right – I think the balance principle applies to the enforcement of the rules as well.

        The appearance that the rules work must be necessary, because it isn’t true. I don’t suppose the enforcement of an unenforceable rule can be unkind.
        But actually, is it kind? To always target Sidles?

    • Nabil ad Dajjal says:

      John Sidles is hard to read on his best days with his typical post being completely incoherent. I’m still not sure whether or not it’s ok to make fun of his writing style because it sounds like he has some kind of thought disorder.

      If the Time Cube guy started commenting here I’d expect him to get banned too. Rambling nonsense which has the structure of language but not the content is really really obnoxious to read.

      • Matt M says:

        Temba, his arms wide 🙂

      • hlynkacg says:

        Rambling nonsense which has the structure of language but not the content is really really obnoxious to read.

        I don’t think that’s a fair description. Sidles’ problem is not lack of content, his problem is that he really is obnoxious. His whole shtick consists of seeing how many veiled insults and passive aggressive jabs he can slip in before the audience gets wise.

      • Tarpitz says:

        I immediately found Sidles’ posts much more tolerable when I started reading them in Mordin Solus’/Michael Beattie’s voice. Pro tip.

    • Anonymous says:

      You may as well say that locking Al Capone up for tax evasion is unjust.

      • Jiro says:

        I would agree that it is unjust.

        • Anonymous says:

          I can’t wait to see the reasoning behind this.

          • CatCube says:

            It’s pretty straightforward. We can’t prove that you did murder, so we’re going to cast around for some other crime to hammer you with. They probably wouldn’t have spent much time looking at his taxes if they weren’t using it as a cudgel for something else.

            This is a process question, and larger justice concerns mean that sometimes people who deserve to go to jail don’t. I don’t lose sleep over Capone himself, but the fact that this is how prosecutors have learned to approach justice is scary as hell. See Popehat on 18 USC 1001 (lying to investigators) to see how this mentality is used in normal cases, rather than exceptionally against gangsters.

            This ties back to the recent Net Neutrality debate. I actually don’t have an opinion one way or the other about the wisdom of the regulations promulgated by the FCC (I just haven’t had the time to read them myself, and the utter frothing at the mouth lunacy over the whole thing means that I probably can’t trust any media report to be honest) However, the way that Net Neutrality was implemented scares the shit out of me. A bunch of unelected bureaucrats used a strained reading of a law to increase their own power. That’s unbelievably dangerous, and if it doesn’t scare the shit out of you, you’re probably making the mistaken assumption that this tactic will only be used to do things you like. If you think that’s the case, remember that Donald Trump is president right now, and he gets to appoint people that will stick their fingers in pies using this method.

            Repeal of this is one of the few things that I can say that the Trump administration is doing a good thing without reservation.

    • Garrett says:

      My experience with Sidles posts is that they were:
      * Very, very long
      * Poor grammar
      * A meandering style, leaving this reader uncertain of the conclusion being argued.
      * A desire to provide a huge number of citations. That may be useful for a formal academic paper, but is overwhelming, especially when the point is unspecified.
      * A sense of smugness and certainty reminiscent of a freshman who’s just read their first philosophical paper, ever.
      * A hostility to attempting to improve.

      The last two are pretty much fatal when it comes to useful discussion. Combined, it comes across more as graffiti-like vandalism of the site rather than a poor attempt at communication. Other may have a different experience.

  18. Nancy Lebovitz says:

    There’s no link to the survey at either LW site. Should there be?

    I’ve got an economics question about The Last Jedi. It would be a mild spoiler. Too soon to post?

  19. sunnydestroy says:

    Can any economics experts help me understand the validity of this research on the impact of Net Neutrality?
    https://link.springer.com/content/pdf/10.1007%2Fs11151-016-9554-8.pdf

    It seems like a pretty convincing argument against Net Neutrality in terms of discouraging investment/making prices higher for rural residents. But I wanted to sanity check against anyone with an actual good base of economic knowledge because I’m not an expert.

    • pontifex says:

      I am not an economics expert. But I do have some comments.

      So, the crux of the argument seems to be this:

      …with current net neutrality regulation as specified in the 2015 OIO, ISPs are not allowed to charge any fees to CSPs. Hence, ISPs are forced to generate revenue solely from last-mile fees, which is likely to result in an increase in the average price that is paid by end consumers.

      This doesn’t really make sense, though. If the ISP charges Netflix $20 to connect to each customer, Netflix has to increase its prices by $20 per customer, in the long run. The content companies are not magical unlimited pots of money. Price increases eventually have to be passed on to consumers.

      Perhaps the argument here is that if the ISPs capture more of the wealth associated with using each connection, they will have more of an incentive to connect people. However, in my experience, ISPs already capture most of the value of the connection. For example, I pay about $60 a month for internet, and Netflix is $10 a month. Would an extra $10 a month to the ISP rather than Netflix greatly change the economics of what neighborhoods are “worth” connecting? I doubt it. And locking me into their crappy services might well make the connection more than $10 a month less useful to me, reducing my willingness to pay in the first place.

      Also, why does South Korea, the country with the fastest internet in the world,
      have net neutrality? There has been surprisingly little discussion of how the rest of the world does internet.

      • sunnydestroy says:

        One thing that someone has cited to me while talking about Net Neutrality issues is the difference of high speed internet availability in rural areas of the EU vs the US. His position was that the EU’s net neutrality friendly laws suppressed investment by ISPs in rural areas.

        I checked some stats and there does seem to be a gap between the EU and US rural broadband access for networks with speeds equal to or greater than 25 Mbps.

        2014 report using 2011-2012 data:

        A far greater percentage of U.S. households had access to Next Generation Networks (NGA) (25 Mbps) than in Europe. This was true whether one considered coverage for the entire nation (82% vs. 54%) or restricted the analysis to rural areas (48% vs. 12%),

        (Pg 1)

        I thought that data was kind of old so I went looking for newer data since the stats may have changed coming closer to current times.

        2016 European Commission report on broadband access (emphasis mine):

        At the end of June 2016, 99.2% of rural households across the EU28 had access to at least one broadband technology. However, only 39.2% (12.0 million rural households) could benefit from NGA broadband.
        Nevertheless, rural NGA coverage growth is increasing at a faster rate than in the first half of 2015, suggesting NGA deployment is shifting towards rural areas. Rural NGA coverage increased by 9.5 percentage points in the last year compared with 2.7 percentage points in the six-month period to mid-2015. In total 2.9 million additional rural households gained access to next generation broadband services between the end of June 2015 and 2016.

        (Pg 21-22)

        2016 FCC broadband report

        For the US, this FCC report says 39% of rural Americans lack access to 25Mbps broadband, which would mean 61% of rural Americans have access to broadband of at least 25Mbps speed (100 – 39).

        So according to 2016 data, the difference is 39.2% (EU) vs 61% (US) for rural high speed (>25Mbps) net access.

        Are EU net neutrality regulations causing this? I haven’t really looked into how the EU regulations are structured so maybe another expert there can chime in on how the EU vs US regulations differ.

        • pontifex says:

          There are tons of complicating factors here. The US had government programs and policies specifically designed to incentivize internet service providers to provide service to rural and semi-rural areas. Some of them were quite expensive and controversial. What is considered “broadband” in one place might not be in another. And the way that internet service is provided is drastically different in some countries. The US model of private entities owning the last mile is not universal.

          It would be interesting to see a relatively unbiased deep dive on this topic. I’m not sure where you would find such a thing, though.

  20. CatCube says:

    I am unbelievably angry right now, with Windows Fucking Update.

    I’m at my parents place, who have HughesNet. They used to have a MyFi, but the signal was so terrible that it was a 50/50 shot if you’d actually get a connection. Anyway, the data allotment is 20GB/month.

    I checked on it early yesterday, to find 92% remaining after 25 days (my mom checks e-mail and that’s about it, so that seems about right). Of course, this morning, I got the “Your computer has been updated” message, which was eye-rolling as always, because it comes with a forced restart at some point today.

    I surfed here and some other places yesterday, and today decided to use up a portion of the remaining allotment. So I watched two videos, an 8-bit Guy video on YouTube and the new Zero Punctuation. To get a sense of what that took, I just checked the data. You know what I found? I’m at the bottom because the fucking update took 15 GB! 15! The fucking OS decided on its own to use 75 fucking percent of a monthly data allotment! I’ve been pissed about the unbelievable arrogance of the people who designed the new update system, but this really takes the cake.

    I’m just about ready to burn my OS to the ground and try Linux again. I stopped playing with it 15 years ago after I decided I’d rather just use my computer, instead of spending all my time trying to get it to work, but this is really dancing on my last nerve. And I’m saying this as a guy who has a Windows Phone!

    (I *do* know about using gpedit to get around the forced update. I don’t use this computer much, so I had forgotten I didn’t set it. It’s also a pain in the ass to use the “metered connection” setting because I have to log out and log back in as an administrator, so I hadn’t done that before it dropped the fucking update on me. Of course, now I’ve done those things. I’m pissed because I shouldn’t have to–those should be available to regular users, not people who’ve learned to grovel in Group Policy.)

    • Incurian says:

      Ubuntu is basically fire and forget these days.

      • Anonymous says:

        @CatCube

        This, basically. Linux UX tech has made great strides. I don’t particularly like Ubuntu, but Mint with XFCE has been quite good.

    • Gobbobobble says:

      Well, see, your first mistake was using a Windows newer than 7. 8 and 10 are unmitigated* garbage. Microsoft is at the “Elderly” phase of software maturity – they had a wonderful system but now they’re just going to shove anti-feature after anti-feature into it until it falls over.

      *open to possible niche examples, my interactions have been universally negative

      • Deiseach says:

        Ah, I’ve been using Windows as long as I’ve been introduced to computers, and Windows 10 is okay (Vista was a nightmare, 7 was a huge improvement, then they went crazy with 8 and started chasing the whole “everyone is on a smartphone so design this for smartphones with touch screens” rainbow, which is why they needed 8.1 and then 10).

        They’re still trying to carve out a slice of the new pie about the Internet of Things and to imitate Apple in creating a whole hardware brand to go along with their OS (I don’t know if they think that Apple is now vulnerable). What that means is that I wouldn’t touch a Surface anything they put out (I’m sticking with my desktop), Bing is a mess (honestly, Bing is awful), Edge may be okay but I’ve moved on to Chrome now that I’ve finally abandoned Internet Explorer, and all the snazzy “we’re doing Fall Creators Updates for the creative types who design the graphics for webpages on their tablets while they sip their [latest modish coffee] in a coffeshop” malarkey goes unused by me.

        Apple were the Cool Kids. Microsoft never were, they were bland, beige dependability. Now Apple is the aging trendy uncle of the last generation trying desperately to appear hip before the new generation of Cool Kids who have moved on to other devices, and Microsoft is still trying to shake off its “used in the cubicle at work for paper-pushing” image instead of embracing the dullness and being Zen about “we are timeless, we don’t chase fads or fashions”.

        • beleester says:

          I’m in the same boat as you. Windows 10 works perfectly fine for me, but I really don’t use the new features much. I’m already used to Chrome, I don’t see the use of Cortana, and I always use type-ahead on the start menu so I don’t care whether it covers the screen or not, so long as it can find something when I type its name. There are a couple of “under the hood” improvements – it boots faster than 7 did, and it has a nicer task manager, but for the most part it feels like the old Windows with new wallpaper.

          I did have to use Edge for browser compatibility testing, and it seems alright.

    • pontifex says:

      Windows update truly sucks.

      I have heard tales of computers upgrading to Windows 10 without the owners’ consent. Tales of huge download sizes, and buggy updates that break the OS. Microsoft apparently fired their entire QA team, so I expect this to get worse in the future.

      I have a computer that somehow got itself stuck in an infinite upgrade/downgrade cycle in Windows 7. Every time I reboot it, it tries to upgrade itself, fails, unapplies the updates, and then reboots again. The whole process takes about 5 minutes. Fortunately, I almost never use Windows– I mostly use Linux.

      Linux really does work pretty well these days. If you’re at the point where you’re editing registry entries, you already are skilled enough to use Linux. Just move on to the grown-up operating system already. On the other hand, if you’re buying a computer for your parents, I’d get them a Mac or a Chromebook. You really don’t want to have to help them with tech support.

    • toastengineer says:

      Of course, this morning, I got the “Your computer has been updated” message, which was eye-rolling as always, because it comes with a forced restart at some point today.

      You can disable those; the old registry/GPE switch that used to work doesn’t anymore, but (unless they patched this…) you can flag the scheduled task Windows makes for itself to reboot as “cancelled,” find the file that represents it deep inside your Windows folder, and then delete the system account’s permission to alter it (it always has permission to change its permissions, but you can take away its permission to alter a file.)

      I do understand why they do all this stuff though; they don’t want the world getting annihilated by something like Mirai, and the only way to do that is to keep everything updated, and the user’s “I’ll reboot later” always turns into never.

      What Microsoft needs is a button that says “no seriously, I know what I’m doing, please turn off all the user-is-a-dumbass switches like we used to be able to” that people who don’t actually know what they’re doing can’t press.

      • CatCube says:

        It’s probably not a good idea to put off a restart for too long, since per Raymond Chen, you don’t really know what state the computer is in after an update. You’ll have some new libraries already active, and some old ones still running because they can’t be updated until the restart. You don’t really know how they’ll behave together long-term, and should just finish the update with a restart.

        What I do (and did do on my main desktop computer) is set Windows 10 to only notify me when there’s an update, and I’ll choose when to install it. Then once I’ve actually got a good time for the computer to restart, I’ll give it permission to install. Sometimes I’ve got enough open on the desktop that might be a few days to a week, depending on what I want to deal with. I don’t use this computer much, only when traveling, so I had forgotten to do that. Normally, on hotel WiFi it doesn’t matter how much data I use so I didn’t make a priority of digging out the proper place on the Group Policy tree to set it. I’ve of course done that now.

        The most common updates are to Windows Defender, which are small and don’t require a restart. I just give it permission to install those immediately. Unfortunately, there’s no way to whitelist those but make it ask for permission to install ones that are likely to require a restart.

        I noticed that Microsoft has gotten more and more shitty in how they treat the user regarding updates. Once 8.1 came out, I didn’t have any huge problems with 8, except for updates. You could either have it install updates automatically without asking, or have it not notify you at all and you’d have to check manually for updates. It was like they modeled the new Windows Update system on somebody’s pissy girlfriend, who passive-aggressively makes things difficult if you don’t give them carte-blanche permission to do what they want. Then, of course, in 10 they just went ahead and made it difficult to turn off automatic updating entirely. There was a slight improvement in that they at least notify you of pending updates when you turn off silent automatic updating.

        Overall, 10 is fine. Except for this one thing, I don’t have huge complaints. The UI didn’t move most of the things that I use daily from Windows 7 and 8.1, it starts up fast, and stability is good. I’ve had a few crashes on my desktop, but I haven’t localized if that’s the OS or possibly something physical with the computer going (the computer is about 8 years old at this point)

  21. Andrew Hunter says:

    So after a hard hour of rolling BJJ, I stagger into the changing room to pull off a sweaty gi. I walk in on a full on discussion of the Jones Act.

    Of all the things I didn’t expect to hear there, shipping regulations that only economists seem to care about seem high up on the list. (Though at least the other guys were also mostly against it, despite working in the local shipping industry!)

  22. Well... says:

    For anyone who did better in physics than I did:

    When you spin a bucket of water around, centrifugal force keeps the water inside the bucket even when the bucket is inverted over your head.

    But part of what’s allowing that to happen is the sides of the bucket, which keep the water moving along with the bucket. Is that right?

    If you had a walled ring–e.g. a hollow tire–and spun it in a vertical orientation, the extent to which the water stayed pressed along the entire inside of the tire, even upside-down, and didn’t just slosh around in the bottom would depend in part on the roughness of the tire’s inner surface, since that determines its ability to pull the water along as it rotates. Am I still right so far?

    Does this mean a cylindrical spaceship like Arthur C. Clarke’s Rama, in order to have a cylindrical sea, would have to have a very rough or even notched sea-bed (with the notches oriented along the axis of spin)? If so, how would the water behave if the sea-bed was very smooth? Would it act like a big ring river?

    • Anatid says:

      The friction with the surface only matters while you are spinning up the cylinder. During the spin-up, you do need some friction with the side of the cylinder to pull the water along and spin it up too. Once the sea is rotating with the cylinder, it will stay put without any friction or notches or whatever, as long as the cylinder maintains a constant rate of rotation.

      Consider the rotating Earth: the oceans don’t stay put because of friction with the seabed; they stay put because they are spinning around at the same rate as everything.

      • actinide meta says:

        This seems true to first order, but if you are actually trying to engineer a big spinning space station, you are going to have another set of concerns around stability. I think I’ve been told that bodies of water that extend far around the axis of spin are a stability problem – presumably a tiny wobble in the rotation of the cylinder results in sloshing of the water that increases rather than decreases the wobble. But I can’t find a source.

  23. Le Maistre Chat says:

    Today I Learned: the baptismal name Jesus is common in Iberia and across Latin America, not just Mexico. A Google search suggests that no one can explain why this is considered pious in Spanish and Portuguese but taboo in other Christian nations.

    • hyperboloid says:

      Speaking as a Latinoish person (my mother is Chilean); the common folk explanation in Latin America is that, as an act of defiance during the Moorish occupation of Spain, Iberian Christians took to naming their Children Jesús in response to the Arab habit of naming children Mohamed.

      Whether or not this is literately true, it should be noted that “Isa”(عيسى), the Arabic form Jesus most commonly used by Muslims, is a fairly common name in the Islamic world; so one way or another it’s very likely due to Arab influence.

      Also, I thought you were French.

      If so, it is kind of odd that your exposure to hispanophone culture is so filtered through north American media that you associate the name Jesús with Mexicans.

      • Le Maistre Chat says:

        I am an American citizen.
        I am currently visiting family in Mexico. I never knew how widely to generalize this custom. Your explanation explains it best, but if it was limited to the Spanish New World, I’d have to wonder if it resulted from converts in the Mesoamerican civilizations wanting to continue using theophoric names after accepting Christ as superior to their pantheon.

    • Douglas Knight says:

      Sure, Jesus is unique to Iberia, but what about Christ? That is used as a surname in many languages, including English. It is used as a given name in Greece and Bulgaria. Maybe hyperboloid’s explanation applies there. What about South Africa?

      (I think the Bulgarians distinguish between Jesus Hristos and the common name Hristo, but I don’t think the difference is important.)

    • Whatever Happened To Anonymous says:

      My perception is that for most people it’s the “Christ” part that’s sacred, so Jesús is an acceptable name, in the way Mary or Joseph are.

  24. Stable Cryptocurrencies

    There are two problems with using Bitcoin as a substitute for conventional currencies. One is that transaction costs are high–I gather there are some proposals to solve that. The other is that its price is very unstable. So far that’s been a plus, since it has mostly gone up. But in general it’s a minus, since it means that if you are holding a substantial amount of currency for transactions you are also speculating in its value, whether or not you want to.

    That raises the interesting question of whether it is possible to construct a cryptocurrency with a stable value. Basecoin is a recent attempt to do so. I haven’t examined how it works carefully enough to offer any opinion on it, but I think in principle it is doable.

    Suppose you want a cryptocurrency which exchanges at one for one with the U.S. dollar. There is a real world model that worked, although it was for paper currency not cryptocurrency–the Hong Kong dollar issued by the Bank of Hong Kong and Kowloon (I think also a second bank). It maintained a constant exchange rate with the dollar (not one for one) by a simple mechanism. Any time the $HK went above the target rate the bank printed more, any time it went below they bought some and took them off the market. Basecoin works on the same principle. In theory it eventually shifts from pegged to the dollar to pegged to a market basket of goods.

    Pre-bitcoin, Chaumian digital cash was supposed to provide a currency that could be exchanged by sending messages, with neither party having to trust or know the identify of the other. The problem was that it required an issuer, a trusted bank. A digital currency makes enforcing money laundering rules hard, an anonymous digital currency, which the Chaumian version would have been, makes it impossible, so governments very much don’t want it to exist, which makes it hard to establish a trusted issuer. The beauty of Bitcoin is that there is no issuer, so the problem goes away.

    How do you maintain a stable digital currency without an issuer, a private central bank? I don’t know the details of how Basecoin proposes to do it, but here is my version:

    You start with, say, ten issuers, each a respected private firm, probably in different countries, each with a private key/public key pair that it can use to prove its identity. You have some way that the software can measure the exchange rate between your currency and the dollar–the Ethereum people, at least, refer to a mechanism for reporting real world facts to the software as an oracle and have some ingenious ideas for making one work. The software then establishes the following rules:

    1. Any time the exchange rate is above the target, one of the issuers can create one more coin, which issuer it is depending on which ones created the last four coins–they go in sequence.

    2. Any time the exchange rate is below the target, one of the issuers, again in sequence, is supposed to buy a coin and take it out of circulation.

    3. An issuer that fails to obey rule 2 is no longer in the sequence for rule 1. Each time it fails to obey it, its debt to the system goes up by one. Only when it has repaid that debt by taking the corresponding number of coins out of circulation does it rejoin the sequence of issuers.

    4. If the exchange rate is above the target and the issuer whose turn it is to issue a dollar doesn’t, after an hour it loses its turn to the next issuer in sequence.

    Assume an expanding demand for the currency, either (long term) from economic growth or from its initial growth. Issuers profit, because they are, in effect, getting an interest free loan in the form of the ability to issue the coins to hold the value down to a dollar. That gives each issuer an incentive to obey rule 2, so as to maintain its ability to issue.

    Rule 4 exists mostly to cover the possibility of an issuer going out of business or being shut down by its government. There should be a mechanism making it possible, if that happens, to transfer its status to a new issuer. In the simplest case, that is done by selling the private key that proves its identity to another firm that wants to replace it.

    What is wrong with this proposal?

    I should say that I was pointed at this problem, and at Basecoin, by my son Patri. I gather it has been discussed at some length recently, so I may well be reinventing the wheel, but I find it more interesting to think such things out for myself than to start by reading what other people have written.

    That reminds me of an anecdote about my friend and ex-colleague the late Gordon Tullock. Someone prominent wrote an article. Tullock wrote a rebuttal. The original author wrote a response, claiming that Tullock had entirely misunderstood the original article. It included the line (by memory so not verbatim):

    We have all been long impressed by how much Professor Tullock has written. It is even more impressive that he has written so much without being able to read.

    • Andrew Hunter says:

      This is an interesting mechanism for pegging a digital coin, but is pegging actually necessary for a more stable currency? That is: Bitcoin is highly volatile, but bitcoin is also new, poorly understood, and highly manipulated by a number of players, and has a reasonably small market cap (as currency goes…) That this leads to volatility doesn’t surprise me.

      What I’m asking is, do you see any reason that bitcoin’s mechanisms are particularly unstable? If this was the major world reserve currency (say, if the US took tax payments only in bitcoin…) do we see a reason we’d expect it to be particularly volatile? (THere’d be a lot of weirdnesses in that it’s functionally a non-fiat currency and there’s not a lot of control, and I don’t tend to think this is a good idea, but would it be particularly volatile?)

    • actinide meta says:

      Assuming, as you have, that there is a reliable, hard-to-manipulate source for the exchange rate, wouldn’t it be better to do away with the extra counterparty risk created by the issuers, and just change the quantity of tokens by “interest” / “helicopter drop”? Every time you get a new exchange rate data point, multiply all balances [1] by the exchange rate. I haven’t thought about this deeply, but I think it is just like your scheme except that every holder of tokens is an issuer, and that they don’t have to actually do anything. Obviously no scheme (without a taxing authority) can prevent the token from becoming worthless, but I think this should “stabilize” its value. Or if it doesn’t, I think your scheme has the same problem.

      You could use the same approach to target NGDP growth, or a price basket, instead of USD, and cut out the (very real) risk that the central bank controlling USD does something dumb.

      I actually think the more interesting question is whether you actually “need” a stable currency. There are two reasonable arguments that I’m aware of, but they both seem to ignore the adaptations that people could make in a world with a fixed-supply currency. The first is that volatility plays havoc with debts, other contracts that specify nominal amounts of money in the future, and expectations. But why can’t people learn to just include CPI adjustments or whatever in contracts? The second is that no one will want to invest in real projects if they can get a better return by just holding money. But this seems to me like it rests on a nonsensical assumption, that a currency can rise in value forever in a world with no real investment. If so, great, we’ll all be rich! In reality it seems like there would be an equilibrium, where the rate of deflation corresponds to the natural risk free interest rate, and nominal interest rates are pure risk premia. Holding money defers consumption, freeing present resources to be used for capital accumulation, and earns you the risk free (real) rate. Lending money gives you a higher return, at a risk.

      [1] You would do this on-chain in O(1) time, by keeping the timestamp associated with each balance and a price index history, and combining these whenever balances are displayed or updated. You might also want a more complicated control algorithm than just multiplying directly by the exchange rate.