Open Thread 127.5

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

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

647 Responses to Open Thread 127.5

  1. Conrad Honcho says:

    In “literally the most America thing possible” news, McDonalds restaurants in Austria now function as mini US embassies.

  2. Aapje says:

    Hypothesis: ‘acting white’ is to many black people what ‘nerd’ is to many white people.

    • Faza (TCM) says:

      Given that both may plausibly be interpreted as disloyalty to one’s tribe – seems legit.

      • DinoNerd says:

        It’s bizarre to think of “being a nerd” as disloyalty of any kind. It looks to me more like behaving naturally, for someone on the autistic spectrum.

        But OTOH, plenty of children have been punished for being on the autism spectrum (= “willful misbehaviour”), and see also the discussion of flat earthers below.

        And heavens knows, people enforcing tribal membership shibboleths have no interest in truth, or fairness, or kindness, AFAICT. (But that’s a rant for another OT thread.)

        • Faza (TCM) says:

          It’s not that hard really, if we consider what tribe is under examination, specifically: school kids.

          I’ll defer to Paul Graham for the ethnography.

          What we should take note of is that if you evidently don’t share your putative tribe’s values, you are an outsider and hence – suspect.

          Tribalism’s added value is trust. Well-adjusted members of the tribe can be expected to behave in predictable ways. Outsiders can’t.

          • baconbits9 says:

            Tribalism is more in line with rent seeking than trust building. When exit costs are high, as is the case with small tribes or geographical monopolies like schools, then high status/power individuals can enforce their norms easily.

          • Faza (TCM) says:

            I’d disagree.

            Tribalism, as I see it, is about strength in numbers: essentially, an extension of familial relations (depending on size of tribe, there’s gonna be blood ties all over anyway).

            Sure, tribes impose norms and status is important, but that’s just a normal dynamic when you have individuals working towards a common goal (advancement of the tribe).

          • baconbits9 says:

            If it was about strength in numbers then you wouldn’t expect to see enforced conformity and ostracism. Which makes your group larger, hard specific norms or broad inclusive norms? Which is economically stronger, a narrow range of skills in a group or a broad range and comparative advantage?

            If tribalism was about strength in numbers the ostracizing aspects would have been out-competed right after agriculture was developed.

          • Faza (TCM) says:

            In a word: no.

            Would you consider strength in numbers to be important for an army? I’d say it’s supremely important.

            This being the case, we can predict with 100% certainty that an army would never execute one of its soldiers!

            Er…

            It’s not about simply having a lot of people; it’s about having a lot of people acting as one.

            “Broad inclusive norms” means that what you have is a bunch of individuals who may or may not share your goals and values. I expand on this idea further downthread but, in short, diversity isn’t what you want any time you need to trust the next man to have your back.

            Regarding outcompeting, would your money be on the larger group that’s aimless and arguing at cross-purposes or a smaller one that is high-trust, where all members share the same goal?

            Now, consider two high-trust, cohesive groups where one is larger. Who’s likely to win now?

          • baconbits9 says:

            You have assumed all kinds of things on your way to your conclusion, such as the idea that needing someone to have your back is independent of how you enforce norms and how accepting you are. Weather or not two countries have open trade is a great predictor of them NOT going to war with each other.

        • Aapje says:

          @DinoNerd

          For some people, murder may be natural behavior. That doesn’t mean that others have to accept it. So I think that we have to look at consequences.

          Furthermore, I think we should assume that human behavior has a reason.

          The consequences of overachieving can be negative for others. Imagine a single cobbler in a medieval town. It takes him 2 days to make a pair of shoes. He can justify asking two normal day wages from his customers. Now a new cobbler comes to town and only takes 1 day to make a pair & asks 1 normal day wage. The customers of the slow cobbler who never complained before, now start demanding he lower his prices. So he now has to up his game or accept a loss in income (either by lower pay or because his customers leave him).

          Another example: one man is very eager to please his wife and gives her daily back rubs. All her female friends hear of this and start demanding daily back rubs as well. The eagerness of this man shifts the equilibrium between men and women in that friend group to the benefit of women and at the expense of other men.

          So a spontaneous solidarity then makes sense between peers: don’t undercut the other by overachieving (too much). You see similar behavior with unions, who also often try to prevent undercutting by peers. Overachieving is then disloyalty to this peer agreement.

          However, this also shouldn’t be taken too far, because if the cobblers are really slow, people will start looking for alternatives for shoes or make shoes themselves. If men are too unkind, women will refuse to marry or will refuse to be kind themselves. So a balance has to be struck between solidarity with the less able and the bargaining power of the group & individual.

          In schools, having the less capable demand solidarity makes perfect sense if people are graded on a curve: the overachievers worsen the situation for the less capable. However, also when the grading is not on a curve, you see similar behavior. Presumably, people always expect to be graded on a curve, regardless of whether they actually are.

          For slaves, solidarity make even more sense. They don’t get paid for their work and have no reason to work harder than is needed to avoid punishment. So in the context of white people enslaving black people, it makes perfect sense to not act white. In other words, it makes much more sense for slaves to favor solidarity much more in the balance between solidarity and bargaining power, because slaves are not free to bargain with their labor.

          Yet in a post-slavery context keeping such norms means that black culture is ill adjusted to a situation where they do have that bargaining power.

          Of course, the most capable can benefit by not showing solidarity. Yet the very capable are few. Bullying is then a way by which the larger group of less capable people can convince the more capable to show solidarity. Five people don’t have to be very capable to be able to harass or beat up 1 person.

          So abuse of overachieving nerds, diligent students and overachieving black people is the enforcement of this solidarity.

          PS. I think that autistic people often fail at pro-social behavior and also fail to notice subtle norm enforcement mechanisms, which causes others to escalate their norm enforcement to more drastic measures. So this suggests that a solution might be to get (partially) exempted from such norms due to reasons of inability (just like children and obviously (mentally) handicapped people can be exempted) or to be held to different norms (like women are held to different norms than men). A complication is that many autistic people don’t look different and that people can only discriminate between people if they can identify to which group one belongs. Furthermore, autism is probably too diverse and ambiguous for people to be able to expect an amount of pro-social behavior that matches the ability reasonably well.

          • DinoNerd says:

            Being on the autistic spectrum myself, I have a very strange reaction to all the claims of how wonderful life was in tight little communities. I wouldn’t have been able to comply with the unspoken social rules, and people would have done just about everything possible to make me conform – except for telling me the actual rules, or teaching me how to do the pointless-to-me rituals required of community members – before finally ostracizing me, assuming of course they didn’t wind up kiling me.

            For people like me, life sucked. If they were lucky, they got out before they were too badly damaged – psychologically or physically. (And to be fair *some* of these cultural enclaves seem to have been less hostile to high functioning autistics than others. But the American small town experience I’ve heard described seems to have been especially awful. Fortunately, however, I grew up in a large city in Canada, and was spared this trauma.)

            But your reaction is telling; comparing autistic behaviours (disliking small talk; believing what people say, minus customary modifiers; telling the truth as they see it; prefering to work with things rather than people – or even some of the more disruptive behaviours) – to murder.

            I guess they are ‘murder’ – in the same sense that e.g. gay people marrying each other ‘destroys’ the institution of marriage.

            I’m so glad that all you can do is look down your nose at me and declare me uncool, not require I spend hours of every day miserably pretending to share your tastes, and to act as if I see myself as an inadequate imitation of you, rather than a person with talents and interests of my own.

            What do you mean by pro-social behaviour anyway? I suspect that when you use that term it’s the opposite of shop talk, direct frank speech, enjoying working with things, and all the other things that nerds typically are seen as doing. So from my POV – bullying, both physical and verbal; believing things based on the status of the speaker and/or one’s own desires; insisting that others (pretend to) believe things you want to be true; and lots and lots of sucking up to people who have high status because they are good at ass kissing, bullying, and other behaviours commonly despised by nerds – etc. etc..

            [Edit on rereading the above – I’d never heard being a nerd equated with doing more of whatever work was happening than others, and thereby interfering with their attempt to convince those handing out rewards that the amount the normal forks were doing was the most that could reasonably be expected. Which is why my response above misses a lot of what you were saying.

            But I think that’s probably fairly common – unless someone explicitly says “we’ve decided to all limit our output to the same level,” and gives reasons why that’s good, many folks on the autistic specturm will just believe the claim that the low level demonstrated is all the average person is capable of, and see an opportunity to be better than others (for a change, given the flack they regularly get for mind-reading failures) and go for it. And many folks on the spectrum make a real fetish of honesty, so won’t want to go along with what they’ll see as fraud.

            In my own case, I remember my first boss telling me that the way expense accounts *really* worked was that there were specific amounts you should claim for specific things, and if you claimed less – even if you’d found a bargain, or felt too tired to eat one evening, and so didn’t claim a meal – that would result in the acceptable quota being cut for everyone, some of whom might have larger appetites etc. than you – so I shouldn’t do that. She then revised my first expense account, before passing it up the chain ;-(

            So yes, that is a thing. And one that many folks on the spectrum would co-operate with if told explicitly, but never figure out from observation.

            But it’s not, in my experience, what nerds are usually dissed for – and FWIW, I see “nerd” and “geek” as insults normally directed at people on the autistic spectrum, but not so far onto it as to be ‘classically’ autistic.]

          • Faza (TCM) says:

            @DinoNerd:
            I understand that this may feel like your person is being attacked – because you are on the spectrum – but might I propose that we step back and attempt to consider the issue dispassionately?

            Also, it’s probably not a good idea to conflate “nerdy” and “on the spectrum”. While there is some overlap, it’s quite possible to be a nerd and not be autistic (at least insufficiently to be diagnosed as such) – I am one such example (I may be schizoid, however, so there’s that). To the extent that there is overlap, nerds and people on the spectrum tend to be treated in a broadly similar way by the “normies” and it’s my guess that the reasons are also broadly similar.

            To unpack my somewhat glib statements upthread: being part of a tribe is being stronger together – both against the environment and against outsiders. Without wading too deep in to the morass that is evo-psych, I would nevertheless posit that this tendency towards socialization is a major part of what makes us humans – and makes us so successful as a species.

            I like the word “tribe” as a generic term-of-art for all sorts of groups that people end up forming, because it captures the pre-institutional nature of a large part of these. Strictly speaking, however, things like nations can also be viewed as tribes.

            The main feature of a tribe is the demarcation between “us” and “them”.

            If you’re one of “us”, you can count on our support against “them” (everything and everyone external to the tribe), but in return you must support us. The strength of a tribe is in its numbers.

            Talk is cheap, so each tribe has methods of building and validating trust amongst its members. Participating in the same rituals, observing the same customs and professing the same values works to increase cohesiveness in the group. Defections are seen as much less likely if we’re all in this together. For similar reasons, tribes tend to be pretty adamant about norms getting enforced – the benefits of tribal support come at the price of obeying tribal rules.

            The end result is that members of a particular tribe end up being broadly similar – very similar when it comes to things the tribe cares about. This is by design: if everyone shares similar values, they’re more likely to work together than against one another.

            When individual members of the tribe are seen to profess a differing set of values, they immediately become suspect. How can they be trusted to work for the group, and not against it, when their goals and values are different than those of the other members? These people must either be brought back in line or ejected from the tribe, because their continued presence weakens group cohesion. The worst case scenario is the disintegration of the tribe, destroying its greatest strength – numbers.

            So what options do you have if you don’t fit in with your tribe (remember, most tribes are like families – you don’t get to pick yours in advance)? You can either depart and join a different tribe – with the danger that you may find yourself at odds with your former tribe at some future point; meaning conflict with people you might otherwise care about; or – you can be a lone-wolf, with all the problems that entails (nobody has your back, for a start).

            What is typically not an option is retaining tribal support without submitting to tribal rule and ritual.

          • The original Mr. X says:

            Being on the autistic spectrum myself, I have a very strange reaction to all the claims of how wonderful life was in tight little communities. I wouldn’t have been able to comply with the unspoken social rules, and people would have done just about everything possible to make me conform – except for telling me the actual rules, or teaching me how to do the pointless-to-me rituals required of community members – before finally ostracizing me, assuming of course they didn’t wind up kiling me.

            I’m not so sure about that. In the first place, most of the people you’d be interacting with would be people you’d known since birth, and people tend to be more understanding with such people than with relative strangers. In the second place, I think a lot of the social rules were actually more open and well-defined than they are now. A medieval lord knew he was a lord, you knew he was a lord, and there were various social rituals you were expected to follow when addressing him (like taking your hat off and addressing him as “My Lord”); his modern equivalent likely still expects deference on the part of the plebs and gets angry if it’s denied to him, but is officially in favour of equality and would deny wanting any special treatment if you asked him directly. Which one seems like it would be easier for someone on the spectrum to know how to act towards?

          • Aapje says:

            @DinoNerd

            I wasn’t comparing autism to murder. I was pointing out that natural behavior is not an accepted/acceptable defense for antisocial behavior by using a clear cut example.

            I’ve been bullied and abused a lot. When fairly large collectives did so and/or let it happen, I can’t blame individual crazies. It was normal humans that did so or stood by. Then I either have to regard most of humanity as evil or see their behaviors as producing fairly good outcomes for them, but unfortunately not for some outliers, such as you and I. The latter seems much more logical and consistent with observations.

            Making rules and obligations explicit doesn’t really work for normies because they are constantly renegotiating the rules, want robustness against rule lawyering, don’t want to hurt people’s egos (and their own) by making things explicit that hurt feelings a lot if said, etc, etc.

            In fact, most people are probably not even able to make their rules and obligations explicit. They are machine learning systems that incorporate inputs (including nudges by others) into a black box system, that doesn’t work with explicit rules. If you ask them about their rules, most people pretty obviously just make up some rules that they don’t actually live by.

            And many folks on the spectrum make a real fetish of honesty, so won’t want to go along with what they’ll see as fraud.

            This is also known as defecting. My point is that going against norms that benefit many people and especially the ingroup, harms many people and/or the ingroup. It’s no more surprising that these people will want take revenge than when someone smears shit all over the communal bathroom.

            If some autistic people are not willing to change their behavior even when told that it is not acceptable, because their morality forbids it, then people with a different morality are not just going to say: “I guess that I have to accept a shit-smeared bathroom then.”

            When you live in a society that gives you benefits, you get to accept the obligations that accompany those benefits or you have to convince people to change the obligations. You can’t just (selectively) ignore the social contract.

            As for tight little communities:

            In a small community, there is less need for signalling, as people know each other’s history. In the big city, most people you meet are relative strangers, including your coworkers. You’ve probably known them for a few years at most. So it is a lot easier to cheat in the city, which in turn makes people more cynical and wary. Want to park in the handicapped spot: get a permit for your handicapped mom and use it yourself. Who is going to challenge you? The people who see you park your car don’t know whether you have a hidden handicap.

            In a small town, you can’t get away with this as easily, as most people will know whether or not you have a handicap. This means that they can prevent defecting more easily, but also that they can more easily agree to threat people with hidden disabilities differently, as long as they agree that the disability is a good reason to be exempted from certain obligations. If they disagree that you should be exempt, but you strongly disagree, you are better off in a city, because you effectively are a cheater in people’s eyes, even if not in your own eyes.

            An issue with small communities is that for practical reasons, they can afford less diversity. There are simply not enough people to create a broad array of thriving past times. So they need to push for fairly strong conformity in behavior to be able to have events with enough attendance, sports teams with enough players, etc.

    • albatross11 says:

      Is “acting white” just the particular way that some kids in black communities refer to people being nerds/geeks/knowitalls/grinds/swots/etc.? I’m not convinced that this is really a pathology of black communities, rather than a pathology of humans or at least WEIRD humans.

      • Aapje says:

        My theory is that slavery made African-Americans associate nerdy and overachieving behaviors with white people, as I explained further in my comment above.

        Then white people try to bully nerds & overachievers in one way and black people do the same in another way, that makes them seem very different, even though the basic mechanism is the same. Although as I argued in that earlier comment, it makes sense for slaves to be much less accepting of nerds & overachievers and thus for a post-slavery culture with legacy behaviors to bully black people who ‘act white’ more than the extent to which white people bully nerds and overachievers.

        So the pathology would then be the racial element and the excessive intolerance of nerdy/overachieving behavior.

    • Well... says:

      I disagree. I’ve watched a lot of classmates, as well as some of my close black friends and at times my wife, get accused of “acting white” and there is something consistently underlying it to do with tribal betrayal, along with “Oh so you think you’re better than me because you talk properly/finished school/dress respectably/etc.?”

      Whereas working-class whites who look down on “nerds” (typified, perhaps, by hipsterish bespectacled Silicon Valley-types) tend to do so because of a combination of “you’re a weird pencil-necked egghead” and “you make a lot of money for what looks like not doing very much/difficult work”.

      • Aapje says:

        I think that a lot of nerd hate is because nerds are seen as uppity: getting rewards they shouldn’t get due to their relative lack of pro-social behavior.

        • Well... says:

          Maybe I’m not clear what you mean by “nerd”.

          • Aapje says:

            I mean what people gesture to when they accuse others of being nerdy. A belief that someone is overachieving in a certain way, is too thing-oriented, etc.

    • Urstoff says:

      Does “nerd” even have any negative cultural connotations these days? “Geek culture” has been so mainstreamed these days (for better or worse) that traditional nerd pursuits seem completely neutral, if not laudatory.

      • John Schilling says:

        The usage is not entirely consistent, but “geek” mostly means preferring for Sci-Fi/fantasy/gaming entertainment genres whereas “nerd” mostly means preferring STEMish intellectual interests to normal social interaction. The former no longer has negative cultural connotations, the latter still does, but most importantly these two concepts are no longer strongly correlated. Hollywood’s recent affinity for comic-book superheroes says absolutely nothing about society’s tolerance for introverted coders even if we aren’t 100% clear on which word refers to which. And whichever word is, at the moment, being used to refer to the latter, still has the traditional negative cultural connotations.

        • Urstoff says:

          Preferring anything (STEM or otherwise) to normal social interactions is always going to have negative cultural connotations because being anti-social is perceived as negative, but I don’t know if that’s exactly what “nerd” means. I think you’ll find plenty of normal social interactions in STEM spaces.

          It would be interesting to see what the usage of the word is among teens.

          • DinoNerd says:

            If wanting to be alone/not talk right now/read a book/etc. is “anti-social behaviour”, what do you call yelling insults, hitting people, damaging common facilities (from littering to blowing up buildings), etc.?

          • Aapje says:

            @DinoNerd

            Some pro-social behavior is the norm in some contexts. If the norm at work is to hand out cake on your birthday and everyone else does it, but you don’t, then that lack of pro-social behavior is anti-social in that context.

            Yelling insults, hitting people and damaging common facilities can actually be pro-social behavior in the context of a group of drunken louts, although it can simultaneously be anti-social behavior in the context of society at large.

            The norms/social contract varies by group/context.

  3. Two McMillion says:

    So I’ve noticed that I am confused and having trouble getting an answer from google. Democrats in the US House of Representatives have introduced a bill intended to roll back so-called “right to work” laws. Vox says, “[R]ight-to-work laws let unionized workers skip out on paying union dues if they don’t want to. Normally, every worker chips in for the cost of negotiating a labor contract, because everyone in the bargaining unit benefits from it. Giving workers the option not to pay means many won’t, which then lowers a union’s membership and political influence. Unions have lost millions of dollars in states that have passed these laws.”

    Okay… that sort of makes sense as far as it goes. But why is it the case that everyone in a workplace has to benefit from union negotiations? Why not just have two different benefit schemes for union and non-union members?

    My first thought was that there was probably a law that said you couldn’t pay union and non-union works differently or something, but I tried google and couldn’t find anything that seemed to work.

    This seems important, because it seems like a lot of conflict about unions comes from things like agency fees. When I hear people complain about unions, I often hear them say things like, “I don’t want to support an organization I disagree with.” Again, fair enough. But normally the solution to this is just… don’t. Is that not an option for some reason? Because instead of people who don’t want to pay just not paying and not getting the benefits, we apparently have a system where either everyone has to pay even if they don’t want to, or we destroy unions.

    Why? This doesn’t make any sense to me.

    • J Mann says:

      Management has an incentive to pay non-union employees the same as union, so as to avoid incenting union participation, and the union doesn’t want there to be an option for people to be hired as non-union, otherwise management might find the non-union guy more “qualified” than his union alternative.

      I assume the union contract forbids management from paying non-union employees more than union employees in return for freedom from union work rules, for obvious reasons.

      • Two McMillion says:

        Huh. That’s an interesting bit of game theory I hadn’t thought of. By making a precommitment to pay everyone the same, management can prevent a union driving wages up.

        A contract that forbids management paying non-union employees more seems good for the union, but ultimately bad for workers. If management ends up paying employees more not to join a union, it seems like the best response is, “Mission accomplished.”

      • A Definite Beta Guy says:

        In our factories, the union contract prohibits hiring employees who are paid less than union employees.

        This, for instance, discourages the use of temporary contractors, so union employees can instead get that sweet, sweet overtime to move empty boxes around for 8 hours.

    • The Nybbler says:

      The National Labor Relations Act is the law you’re looking for, specifically section 9:

      Sec. 9 [§ 159.] (a) [Exclusive representatives; employees’ adjustment of grievances directly with employer] Representatives designated or selected for the purposes of collective bargaining by the majority of the employees in a unit appropriate for such purposes, shall be the exclusive representatives of all the employees in such unit for the purposes of collective bargaining in respect to rates of pay, wages, hours of employment, or other conditions of employment: Provided, That any individual employee or a group of employees shall have the right at any time to present grievances to their employer and to have such grievances adjusted, without the intervention of the bargaining representative, as long as the adjustment is not inconsistent with the terms of a collective- bargaining contract or agreement then in effect: Provided further, That the bargaining representative has been given opportunity to be present at such adjustment.

    • greenwoodjw says:

      Unions negotiate “closed-shop” agreements where only union members can be hired, and in states where it is illegal, they negotiate requirements that non-union workers get the same pay and benefits, because they don’t want the cheaper workers out-competing them. They also negotiated requirements that non-union members have fees taken out of their paycheck to cover the costs of “representation” (generally over 75% of dues), so there was no way out. The latter was recently ruled unconstitutional.

      Basically – union power relies on establishing and having legal authorization of a monopoly on labor for a company. Losing the monopoly breaks the union (because they’re out-competed, like all would-be monopolies)

      • Murphy says:

        In Ireland it’s illegal to discriminate based on membership or non membership of a union.

        Unions can’t force you to pay dues to a union you’re not a member of.

        So union and non-union members get the same benefits.

        I remember a union in a place I used to work.

        Union dues were extremely cheap. When there was a major dispute the union sent a professional negotiator. Our pay was pretty good because they negotiated pretty well and it was a pretty pleasant workplace.

        A fair fraction of people still joined the union because humans aren’t homo-economicus and can recognize when a group is acting generally in their own interest and often want to support that group, even if they don’t have to when the cost is reasonable.

    • Deiseach says:

      Why not just have two different benefit schemes for union and non-union members?

      Because when Joe the Non-Union guy notices that Carl the Union guy gets better hourly rate/has his breaks guaranteed/gets double time for working Sundays and he doesn’t, the first thing he’s going to ask the boss is “why are we being treated differently?” and replying “He’s in the union and you’re not” won’t fly. It won’t matter that Carl gets better terms because a big organisation negotiated them for him, the non-union workers will resent the unequal and (perceived) unfair treatment, and the company will face having to treat non-union workers the same as union workers, so they might as well be in the union anyway.

      “But I want the same treatment!” “Well then join the union” is something no employer has ever said, and it’s much more likely that the resentment caused will be aimed as pressure to get the union out instead.

      I have no experience of American unions and all I’m relying on is hearing stories of how they abused their power and became more interested in enriching their officers than protecting the workers, and that’s what is behind the reluctance of workers to join unions – they don’t see any benefits for them and it’s an expense taken out of their pay.

      But that’s a global problem in that yeah, unions that got big and powerful started getting (a) courted by politicians looking for a deliverable voting bloc of thousands of votes and (b) started getting much more interested in perks for the guys running the union than doing their job representing the workers – that’s definitely happened in both Ireland and the UK. Like the US, big unions fossilised work practices in the 60s and 70s shackling home-grown industries and allowing foreign competition that was more nimble to come in and steal their market – be that the automotive industry in the US or the miners’ union in Britain.

      On the other hand, when you’re in trouble at work, having the union at your back makes a huge difference in how you’re treated. When it’s “Joe versus the Corporation”, the company can wait Joe out knowing they can afford to pay lawyers to run out the clock on a wrongful dismissal case for longer than Joe can; when it’s “the Union and Joe versus the Corporation”, it’s often (not always) a different matter because now it’s not one guy paying out of private means, it’s a large organisation with its own deep pockets and rabid lawyers going toe-to-toe with the company.

      • Clutzy says:

        Your explanation makes no sense. The person would, indeed, just join the union then. The difference is that unions often negotiate slightly higher wages for their workers, but no enough to pay for union dues. Or they have laws passed for them that forbid paying less than union rates to nonunion employees once a shop has passed a unionization referendum.

        Your situation I don’t think is a large concern for any party.

        • Faza (TCM) says:

          Getting union benefits without paying union dues seems like a much better deal than same with dues. Not to mention any other obligations union membership imposes.

          Few employers are particularly happy about union presence, so anything that might undermine the union’s position is likely to be considered. Offering non-members the same perks as members is one of the first things that comes to mind (“you don’t need to join the union, Joe, all they’ll do is take a cut of your paycheck”).

          What you won’t get is the employer telling Joe that if he wants the benefits of union membership, he should join the union, because that strengthens the union.

    • ana53294 says:

      For some things, being part of the union does give advantages that being non-union does not.

      In Spain, union members’ dismissal will get fought in arbitrage with the company. So basically, a union member has to do something criminal to get fired, because the union will defend members. Unions also fight for their members in redundancies, which are negotiated.

      I guess that in the US, union lawyers won’t represent non members for free when fighting against your employer, right?

      Salaries and vacations, AFAICT, are the same.

    • Beck says:

      Why not just have two different benefit schemes for union and non-union members?

      There are some dual shops in construction. I’m not sure how common they are and, given that most of the Google results on the subject are links to legal firms, they seem to have to jump through some legal hoops.

    • sharper13 says:

      This may be obvious… but the goal of the new legislation is to require States to force all employees at a company to join a unions if at some point a majority of the employees at a company voted to join one. Unions as organizations tend to overwhelmingly support Democratic politicians with money and volunteers, so it’s not difficult to understand the unstated motivations involved (Note, there are plenty of additional stated reasons as well, which you can believe or not based on how cynical you want to get).

      The statement “Rght-to-work laws let unionized workers skip out on paying union dues if they don’t want to” is inaccurate. Right-to-work laws give people the right to work without having to join a union, so it makes no sense to call them “unionized workers” unless you’re attempting to insert a bit of cognitive bias. They essentially forbid closed shops where you must join a Union in order to have a job.

      Typically shops with both Union and non-Union workers differ in how employees are treated, but how much different is a pretty wide range from just not having a union rep during disciplinary issues to completely different pay, workplace, and seniority rules.

      The combination of right-to-work laws in some States (which keep joining a union voluntary) and USSC decisions prohibiting forcing people who don’t join a union to pay for union political speech they disagree with has put a real damper on unions in those States, as its turned out that many previously forced-to-pay-to-keep-their-job contributors don’t actually want to contribute. That in turn has deteriorated the traditional level of support Democratic politicians have enjoyed from unions in the past, so of course they’d like to “fix” that.

      As a thought experiment, imagine that in some Republican States, to purchase a firearm you were required to pay a monthly firearm education fee to the NRA, based on the argument that you’d benefit from the firearms safety information they’d send you and the the safety classes you’d be entitled to as a “member”. It’s a tiny bit of a stretch because of the historical situations differing, but in effect it’s not that much different.

  4. dndnrsn says:

    RPG thread storygame thread!

    By some accounts, they’re a different beast from RPGs. Taking the definition that something is a storygame when the mechanics determine narrative control or outcome of some narrative process rather than success or failure of an action – what storygames do people like?

    I’ve tried Fiasco a couple of times recently, and it seems like it has potential for my group. I assume things will move more smoothly with practice.

    I’m thinking of trying to fit in Blood & Honor which seems more of a hybrid game, but looks like it has some interesting mechanics to incentivize placing the group over the individual character.

    • broblawsky says:

      I’ve had a lot of success with Fiasco. You just need to pitch it the right way to your players – they need to understand that they aren’t playing individual characters, they’re co-writing a story.

    • RDNinja says:

      My brother wrote a storygame along those lines, called Journey Away, which I’ll shamelessly plug: https://www.drivethrurpg.com/m/product/262890

      In his game, you can pretty much automatically succeed at anything you attempt, but the dice determine how many complications are caused or boons are gained as a result. Then another player narratively determines what those are.

    • J Mann says:

      I watched the We’re Alive:Frontier streamed game, and when you have a really committed group of players and a GM who wrote the game, Outbreak:Undead 2ed works pretty well.

      IMHO, the biggest challenge in story games is commitment from the players and the GM. If you have a couple people there who are emotionally invested in “winning,” it will be easy (and detrimental) for them either to shift the spotlight to their character and/or to shift narrative events in their character’s favor.

    • mendax says:

      Fiasco is fun.

      Microscope is good, if it counts as a story game. We didn’t do any scenes, so it was more of a guided brainstorming, but still recommend.

      Dialect is a unique experience. Would recommend.

      I’ve got a copy of Circles of Power. Undecided if it would be best to play with my pro-SJ friends or anti-SJ friends.

      I was really hyped to play Apocalypse World, and had prepped playbooks with the sex moves taken out (originally planned to move them to the back page, but that proved too much work) but lost the files before I printed them and the enthusiasm is gone.

      • dndnrsn says:

        There’s something in the book or maybe I read it somewhere else, that he originally wrote the game for his wife. Not sure why the sex moves weren’t made an optional thing. Just cutting them out weakens the PCs a little bit, and my players are the sort of people (or, a couple of them are) who would complain about anything that unilaterally weakened the PCs when it wasn’t obvious they were OP. So I’d have to put something else in, probably. In any case, it seems like a bad idea to put an incentive for sexytimes (and thus magical realming) into the game.

    • dndnrsn says:

      It’s about a fictionalized feudal Japan, so I doubt the reference was intentional. But it got me thinking: storygames usually can push the envelope in terms of “the characters are bad people” more than RPGs (on the face of it; “all orcs are CE” type stuff aside), it seems to me at least. What’s the furthest that envelope has been pushed?

  5. reddragon says:

    For your argument to be correct then you would have to believe that a small change in insurance premiums would be enough to make you not have locks on your doors.

  6. honoredb says:

    Does having a well-trained security guard with a firearm at a U.S. workplace increase or decrease violence on net? How do I go about figuring this out? I see some claims out there that having an armed guard increases your insurance premiums, which sounds like the market saying that the risk of violence is ultimately higher, but presumably your liability for a shooting is higher if the shooter is a guard than an intruder so there’s more than one signal here.

    This seems like something people could and would study, but maybe regional effects and (thankfully) low sample size make it too hard?

    • Conrad Honcho says:

      I see some claims out there that having an armed guard increases your insurance premiums

      Corrected for the obvious confounder that a business that thinks it needs an armed security guard is probably more in danger of attracting criminal attention than one for the which the owners do not think they need a security guard? That is, does the act itself of getting a guard increase your insurance premiums or is the effect:

      bad neighborhood -> security guard and higher premiums

      good neighborhood -> no security guard and lower premiums

      • honoredb says:

        I saw this in the context of “pros and cons of getting an armed security guard,” which probably eliminates the confounder issue unless insurance companies are being really silly with the incentives they’re creating (“You’re putting a lock on your door? Then we’d better raise your premium!”).

        • Conrad Honcho says:

          Doesn’t that do just the opposite, then? “You’re getting a guard? Then we’d better raise your premium!”

          It’s not the guard that makes the premium go up, it’s that places that need guards are at greater risk than places that don’t need guards.

          • honoredb says:

            Right, I think we agree here, I’m just pointing out that if insurance companies use the presence of a guard as a signal, they’re creating bad incentives so they probably don’t do that. So yes, I’d certainly expect a lock on the door to correlate with higher premiums. But I wouldn’t expect insurance companies to specifically raise premiums if they saw you had a lock on your door, because then people who needed locks wouldn’t install them and they’d end up making more insurance claims.

            So I’d expect insurance companies to raise premiums in response to you adding an armed guard if and only if they believed the armed guard causally increased your expected liability.

          • quanta413 says:

            It’s not necessarily true that charging more if a company uses an armed guard creates a perverse incentive. It depends on exactly what the insurer is covering and exactly what the armed guard protects against. I wouldn’t be surprised if they actually lacked overlap in such a way that the insurer could safely use the hiring of an armed guard as a sign they should increase your premium.

            Exactly what coverage cost is going up if you hire an armed guard strikes me as very important.

          • Conrad Honcho says:

            That’s a good point quanta. Are we talking the liability insurance premium or the property insurance premium? I could see the property insurance premium going down while the liability premium goes up.

    • Drew says:

      US tort law treats action & inaction very differently. This would create a difference in insurance rates, even if security guards made things safer in aggregate.

      Without the security guard, you can claim that a shooting was an unforeseeable tragedy. It’s awful that it happened. But, since you didn’t do it, it’s not directly your fault. And since employers don’t normally hire guards, you’re not negligent for failing to do so.

      With the security guard, things change. Now I can argue that the attack was foreseeable (You hired a guard!), and only happened due to your cost-cutting, and failure to provide the guard with training X, Y and Z. Also, if the guard screws up, then that’s an act of commission on your part, so obviously you’re liable.

  7. Le Maistre Chat says:

    Thoughts on the consolidation of visual entertainment, and the de-consolidation of streaming that started with Disney pulling its content from Netflix?
    The idea of Netflix was access to any movies or TV shows for one monthly fee. It was an idea that propelled it to a market cap of >$152 billion. However, media consolidation has meant that, outside of funding its own content, it only has a few companies to get flix from: Disney-Fox, Sony, AT&T’s Warner subsidiary, Comcast’s NBCUniversal, and Paramount. So Netflix is becoming a victim of its own success as several of these companies dream of a streaming subsidiary worth as much as Netflix.
    Like Netflix, Disney and Paramount are pure content companies (ignoring Disney’s parks, small retail stores, etc.) AT&T is vertically integrated through phone service, and Comcast through cable internet (& its legacy business of cable TV). This would be bad for the pure content giants and consumers who prefer them if data providers are legally allowed to discriminate in the future. Meanwhile, Sony is this odd conglomerate that manufactures semiconductors and consumer electronics, also produces content in the music and video game media, and is a financial services and insurance company in Japan.
    Disney pulled all its content from Netflix some time ago and now owns 2/3rds of competitor Hulu with full administrative control (Comcast owns the other 1/3rd) and is rolling out Disney+ later this year, which in conjunction with its sports stream ESPN+ seems to me a confused business model, but a huge blow for Netflix nonetheless. Comcast will also be pulling its content for its own streaming service. And AT&T explicitly said the same today.

    • acymetric says:

      I know Disney owns a ton of properties now, especially after giving Fox the Katamari treatment, but I just don’t see how they can support a full priced streaming service. I can also see things getting so compartmentalized that people just give up and go back to cable or the similar Internet based packages (Youtube TV, Sling, etc).

    • Conrad Honcho says:

      Yeah I’m about to have to cut the…cord cutting cord? I canceled cable 6 years ago because it was too expensive at ~$100/month and subscribed to Netflix instead. But then got a Hulu+ subscription to watch new TV shows without commercials. And then crunchyroll for anime. And then YouTube Premium for Cobra Kai. I don’t know what’s going to happen when every single content house wants $9.99/month for their streaming service.

      • Thomas Jorgensen says:

        More people will start doing the “Subscribe a month, watch the content you like, cancel until next year” rota. Except, much of the fun of media consumption is to be a part of the conversation, which is not possible under that model, unless you coordinate on a huge scale.

        Though the idea of “June and july are disney+ months” as a cultural phenomenon is bloody hilarious, I do not think it will actually happen.

        • baconbits9 says:

          Americans at least watch far more TV than just 1st showings of new episodes. Most of the popular ‘gotta watch so I can talk about it at the water cooler shows’ aren’t that popular. GoTs spent its first 6 seasons averaging fewer than 8 million viewers per episode, while non event shows (ie sitcoms) average double or triple what GoT pulled in for its early run.

      • baconbits9 says:

        US TV has been driven by old content for decades. Syndication was often the major revenue source for TV series and I bet that will become on some level the model again. These streaming services will cross platform even if a few years after the original airing, how much that works will be tbd, but only the type of person who was already paying for 500 cable channels is going to end up paying for a dozen streaming services at $10 a month.

        • AG says:

          Yep, the rest will go back to what they were doing before legal streaming took off, aka pirating.

          • acymetric says:

            I think this is going to become a problem for subscription models generally. I’m looking at you, everyone who pivoted to SaaS.

    • John Schilling says:

      Thoughts on the consolidation of visual entertainment, and the de-consolidation of streaming that started with Disney pulling its content from Netflix? The idea of Netflix was access to any movies or TV shows for one monthly fee. It was an idea that propelled it to a market cap of >$152 billion.

      Access to any movies or TV shows for one monthly fee is a workable idea if the monthly fee is on the order of $200/month. Well, probably closer to $2,000/months for any movies or TV shows, but “everything that anybody else is talking about over the water cooler and almost everything you fondly remember and want to rewatch and you won’t notice the gaps in the catalog” is probably $100-ish per month.

      Netflix got to an eleven-figure market cap on the basis of a business model that only works where Netflix is a small company that represents a tiny fraction of e.g. Disney’s distribution, so that letting Netflix distribute a large fraction of Disney’s catalog for a few dollars per month is not a big hit to Disney’s revenue and can be justified as cheap advertising – get more people talking about Disney’s stuff over the water cooler so that people who aren’t early-adopter technophiles will pay full price to join in. That clearly isn’t a good deal for Disney any more, so yeah, they’re either going to pull all their content or charge enough that Netflix can’t pay Disney’s price and maintain positive cash flow with any fee structure Netflix’s customers will stand for.

      And that logic applies to any pairing of content producers and content distributors. Providing a mass market with nigh-unlimited access to movie/TV entertainment is going to cost $200/month or thereabouts because that’s what it actually costs to produce and distribute all that content with modest profits for all involved. Whether that means that there’s One Giant Mouse-Shaped Producer of All Entertainment that charges $200/month to every American who doesn’t want to be left out of the common cultural dialog, or half a dozen distributors buying content from a dozen or so producers, each charging $40/month or so and with overlapping catalogs, is indeterminate at this point.

      If you want to pay less than $200/month, you have to consume substantially less (or cheaper-to-produce) movie/TV entertainment than the average American. Substantially less because now you’re demanding the additional overhead of a la carte pricing, and you’re probably a niche market in demanding that. Lots of people imagine that they are the sort of people who consume substantially less/cheaper TV than the average American, but most of them are wrong. Most of them are in the position of standing in front of the $20 all-you-can-eat buffet and saying “but I’m only planning to sample ~10% of the entrees here, so certainly you can let me in for $4”.

      • Edward Scizorhands says:

        I largely agree with the point about bundling, because people wrongly imagine they can save a bunch by unbundling all the useless stuff they don’t use.

        Yet, there is a difference between paying for access to everything being made right now and new, versus access to the libraries of stuff over a few years old. The latter is probably doable on $20 a month, because people already paid for upfront access to watch it when it was fresh.

        • JPNunez says:

          It’s surprising to me how few old shows Netflix does have.

          They got like Friends, and …. He-Man? And that’s it??? I think some Star Trek at some points.

          I suspect a lot it’s driven by people wanting shows to fill their whole screen, so maybe Netflix isn’t fond of non 16:9 shows.

          Despite that, how expensive can be to get a bunch of reruns of old syndicated shows? Surely a bunch of them must be cheaper than one more season of Orange is the New Black, which stopped being good at like season 2.

          • Nick says:

            I think they’ve recently (last year or two) lost quite a bit of content. I know they were pushing for greater exclusivity in a lot of these deals, as in, they wanted to be the only company streaming this content; I guess those deals didn’t work out.

          • John Schilling says:

            It isn’t a trivial expense, because even if the rights to older shows are available for say $0.99 unlimited worldwide distribution in perpetuity, there’s a substantial administrative overhead in making sure you can come up with the relevant documentation on demand when e.g. Alexander Courage’s heirs don’t think they’re getting the royalties they deserve every time the Star Trek theme is played.

            Now, how many people are there who will actually cancel their Netflix subscription just because it doesn’t have TOS or whatever?

      • AG says:

        Is it just that we’ve been spoiled on production value (and that said production value has a real prohibitive cost) that audio-visual media hasn’t gone the way of the music industry?

        I mean, yes, the whole “pivot to video” sector has gone the way of the music industry, but even the few heavy hitter music artists who actually make money (mostly via non-music endeavors womp womp) are still on Spotify.

        • Faza (TCM) says:

          (…) even the few heavy hitter music artists who actually make money (mostly via non-music endeavors womp womp) are still on Spotify.

          Where else are they gonna go?

          This isn’t a rhetorical question: the traditional physical market (CDs) is on its last legs, file download sales have imploded spectacularly and while vinyl continues to show growth, it remains a niche.

          Spotify (and streaming in general, but Spotify is bigger than anyone else) are the only mainstream recorded music market left standing.

          The difference between music and movies is that even leaving “production values” aside, movies are overwhelmingly more expensive to make – just by virtue of the necessary number of people involved (the entire cast, for a start). I can make a record on a pack o’ cigarettes, a jug o’ coffee and a lightning flash of inspiration. Movies, not so much.

          The flip-side is that the music biz cannot really breed mice the size of Disney. Even the Big Three don’t have that kind of clout.

  8. proyas says:

    Does anyone here know about lasers and laser weapons? I’m not talking about sci-fi, I’m referring to actual scientific and engineering knowledge of real lasers.

    I curious about the future potential of laser weapons, but lack basic knowledge about them. The research I’ve done seems to indicate that a laser beam’s destructive power is determined by the amount of energy it transfers into the target it strikes, and that the energy is measured in Joules or Watts.

    Moreover, based on news articles I’ve read and demonstration videos I’ve watched, it seems like a laser would need to be able to transfer at least 1 kW of energy into a target in one second to be a practical anti-personnel weapon. That assumption is based on the requirement that your weapon be able to melt through cloth uniforms or thin metal helmets, and to seriously burn the flesh underneath, and that it be able to do this too fast for the enemy to realize what is happening and jump behind an object for cover (i.e. – a 100 Watt laser would have to focus a beam on an enemy soldier’s helmet for 20 seconds to melt through it, which is too long since the soldier would probably realize after 10 seconds that he was under attack, and then duck his head down behind a wall or something). Is my 1 kW/sec estimate right?

    Second, this is probably a dumb question, but please humor me: If a laser can transfer 1 kW of energy to an object in a second, then the laser’s power source must also be capable of releasing 1 kW of energy in one second, correct? No one will ever be able to invent a future technology that lets them make 1 kW/sec laser pistols that somehow use 100 W/sec batteries, right?

    Third, I’ve heard that heat waste is a major problem for military lasers. Electricity goes into the laser gun, it is turned into a powerful laser beam that shoots out the other end, but the energy conversion is not 100% efficient, so some fraction of the electricity gets turned into heat, and the laser gun gets hot to the touch. Is my understanding of this process right, or do the laser guns get hot for other reasons? Would it be possible to build 1 kW/sec laser guns that didn’t get hot (maybe by using parts made of superconductors), or would that violate the laws of physics?

    Thanks.

    • The Nybbler says:

      Second, this is probably a dumb question, but please humor me: If a laser can transfer 1 kW of energy to an object in a second, then the laser’s power source must also be capable of releasing 1 kW of energy in one second, correct? No one will ever be able to invent a future technology that lets them make 1 kW/sec laser pistols that somehow use 100 W/sec batteries, right?

      No. Pulsed lasers take in energy over a (relatively) long period of time and emit it over a short period of time. So at perfect efficiency your 100W/sec battery could support 1 kW/sec pulses at a duty cycle of 10%. This is important because destructive potential often depends largely on peak power delivered, not average power.

      • proyas says:

        What is a “pulsed” laser?

        What is the difference between “peak power” and “average power”? Can you explain it by referencing my hypothetical 1 kW laser?

        • The Nybbler says:

          Some lasers output their light as a series of pulses rather than a continuous beam of light. I am definitely not an optical physicist so that’s about the limit of my knowledge.

          The difference between “peak power” and “average power”: Peak power is a measure of the maximum instantaneous power reached. Average power is total energy transferred over a given period of time.

          Suppose you have your 1kW laser fire continuously at an object for one second. That’s a peak power of 1kW and an average power of 1kW. Now suppose you have that 1kW laser fire 10 pulses of 1/100 second each, over the course of one second. You still have a peak power of 1kW, but now your average power is down to 100W.

    • Joseph Greenwood says:

      Disclaimer: my last physics class was AP Physic C in high school. I am an amateur, and nothing I say should be taken very seriously.

      I curious about the future potential of laser weapons, but lack basic knowledge about them. The research I’ve done seems to indicate that a laser beam’s destructive power is determined by the amount of energy it transfers into the target it strikes, and that the energy is measured in Joules or Watts.

      First thing: joules are a unit of energy. Energy is what it takes to do work, which can mean a lot of different things, but in this context it mostly means “heat things up.” 4.18 joules of energy is one calorie of energy, which is enough to heat one gram of (pure) liquid water 1 degree Celsius (making reasonable assumptions about ambient pressure etc., and assuming the water is not on the brink of freezing or boiling). Watts are a unit of power (W = J/s), which is energy over time. Power matters a lot for a laser weapon because under normal circumstances I wouldn’t expect your laser to stay in contact with its target for very long.

      Moreover, based on news articles I’ve read and demonstration videos I’ve watched, it seems like a laser would need to be able to transfer at least 1 kW of energy into a target in one second to be a practical anti-personnel weapon.

      Let’s make a little Fermi estimate. Suppose that you are able to keep your laser in contact with your target for one tenth of a second. One kW then would transfer 100 J of energy, which is enough to heat one gram of water 23 degrees celsius. According to Wikipedia (https://en.wikipedia.org/wiki/Enhanced_Combat_Helmet_(United_States)), modern military helmets in the US are made of thermoplastic and… I don’t really know what that means in terms of specific heat and meltability. One gram of water isn’t very much, although 23 degrees celsius is a decent amount. Um… I’ll come back to this later, after I’ve had a little more time to do research.

      Second, this is probably a dumb question, but please humor me: If a laser can transfer 1 kW of energy to an object in a second, then the laser’s power source must also be capable of releasing 1 kW of energy in one second, correct? No one will ever be able to invent a future technology that lets them make 1 kW/sec laser pistols that somehow use 100 W/sec batteries, right?

      It is actually possible for a laser with a 100 W battery (not W/sec–that would mean that the amount of energy it discharges is continually increasing) to fire a 1 kW energy beam, if it has some intermediate place where it can store the energy and then release it all at once. This is the same idea as winding back the arm of a catapult: sure, you can’t put that much energy into pulling the catapult arm back second by second, but if you release it all at once, the results would be devastating.

      Third, I’ve heard that heat waste is a major problem for military lasers. Electricity goes into the laser gun, it is turned into a powerful laser beam that shoots out the other end, but the energy conversion is not 100% efficient, so some fraction of the electricity gets turned into heat, and the laser gun gets hot to the touch. Is my understanding of this process right, or do the laser guns get hot for other reasons? Would it be possible to build 1 kW/sec laser guns that didn’t get hot (maybe by using parts made of superconductors), or would that violate the laws of physics?

      Heat is a bugger! Superconductors are a long way away from military readiness (at least, as far as unclassified records go 😉 ) and I think this is one of your biggest obstructions to real laser weapons. The other is efficiency: you’re putting a lot of effort and energy into your laser gun when you could just shoot your target with an M4 carbine.

    • mendax says:

      Atomic Rockets is a good source of information about such things.

      • woah77 says:

        Gonna second this. I’ve spent the last six months reading and researching using Atomic Rockets as I’ve prepared and started to write a book.

    • Murphy says:

      Depends on whether you’re ok with your weapon system being a war crime.

      Why melt through cloth and steel?

      There’s already systems that can target and shoot the wings off a mosquito in flight, every consumer camera comes with high quality face detection…. combine the 2 and permanently blind everyone facing about the right direction over the course of a few seconds.

      • Radu Floricica says:

        There are times when I’m really, really scared of how a post-robotics, post-AI war would look like.

        • albatross11 says:

          Everyone’s automated defense networks go to war and when the smoke clears, everything in Earth’s biosphere but the automation is dead. Oops, bummer.

          • Radu Floricica says:

            Nah, it’s not just the fear of most or all of us dying. World wars brought industrialized slaughterhouse, Cold War brought the specter of the world, or at least the civilization ending.

            This… this is going to send us again to the lowly position of prey. No dignity, no purpose, just scattering like bugs and hoping that things better than us are full that day and don’t feel like killing. Tactically outpowered and likely outsmarted in every encounter.

            In the end, we might even move again to a more r-selected survival strategy – outlast and outbreed the machines, deep into the rubble they don’t need.

            Not to mention the potential of this kind of power mixed with the idealism/madness of humans. Killing one side of the culture war is no longer impossible. Not “here” in the west, not yet – but have a genocidal war 20 years from now, and you can bet there will be face recognition coupled with databases from social media. “This you can shoot, he opposed the Great Leader 5 years ago”

      • AG says:

        Wouldn’t this quickly lead to custom anti-laser contacts?

        • Murphy says:

          not sure how such contacts would work since there’s a choice of lasers in the visible spectrum.

          it’s like trying to make a contact that lets you stare directly at the sun while still letting you see fairly normally.

          I mean you could wear welding masks 100% of the time but people aren’t gonna be keen on that.

          • realitychemist says:

            I mean one solution to this problem would be an indirect vision system, e.g. an opaque mask with cameras on the outside and screens on the inside, like VR passthrough but optimized for this purpose. It wouldn’t be as good as actual vision (at least without a lot of R&D investment) but soldiers already operate in this mode when using night vision systems. Since there’s a limit to how bright the internal screens can be you have protection against laser blinding, and it’s probably possible to engineer the cameras to be much more resistant to high-intensity lighting than a human eye.

            It also opens up other possibilities like IR imaging, but that’s a bit out of scope here I think.

    • bean says:

      Probably the best place to start is at Atomic Rockets (see also). I used to be really into this, but haven’t kept up much over the last few years. It would be possible to use capacitors to buffer between a 100 W battery and a 1 kW laser, at a cost in weight and duty cycle. And yes, the laser gets hot because of waste heat. This is basic thermodynamics, and while you can cut it, you can’t get rid of it entirely. 1 kW sounds low. I’d guess you’d really want in the 5 to 10 kW range for a continuous-wave laser, although it might be good for average power on a pulsed laser.

    • vV_Vv says:

      Laser pistols are probably always going to be unfeasible barring some sci-fi-like technology like miniaturized fusion reactors.

      Defensive laser weapons installed on ships or fixed platforms used to disable small threats like slow drones or boats are currently in development, and at least one is operational.

  9. Murphy says:

    Epistemic status: Vague Paranoia re: social norms in small communities and their source.

    I’ve got a hypothesis to run by people and I want to get an idea how reasonable it sounds to people who aren’t me.

    I’m primarily a programmer, working in bio science but I’ve always had an interest in crypto and kept up with some crypto communities.

    There’s also a social norm in crypto communities that basically says “if you’re thinking of rolling your own code… don’t”

    which has a perfectly logical argument behind it, basically you’re very likely to screw up and end up with a weak crypto implementation.

    Indeed it’s very very easy for complex flaws to hide in seemingly correct code.

    Now my issue is that this norm… while well intentioned and logical… also leads to a significant monoculture in crypto.

    There’s about a dozen commonly used crypto libraries that account for the vast majority of all encrypted traffic.

    Imagine an opponent, any well-funded organization that might want to read encrypted messages. Bonus points if they employ lots of high quality cryptographers, programmers and mathematicians.

    Now it seems logical to me, that if I was running such an organization, one vector I’d invest in is compromising code in commonly used libraries in some manner that I can exploit. Having a team of high quality cryptographers, programmers and mathematicians who are as free as anyone else to submit patches to open source crypto libraries seems to make this step quite tractable. Or just paying off someone working on each project to insert weaknesses, a few million bucks per project would barely register on their budget and would have a high chance of success. (reasonable view?)

    once the 12 projects are compromised then automated cracking of almost all traffic becomes fairly straightforward.

    If lots of people rolled their own then each would likely be much weaker … but would take dedicated human man-hours to attack.

    Now there are cascade ciphers where you use multiple ciphers.

    It seems to me that the best of both worlds would be to have a norm of always using a cascade cipher where you encrypt the encrypted output of another cipher. With the outermost one being custom: that way, as long as you don’t re-use keys, it’s very hard to weaken the inner layer.

    But this point is where I start getting responses that seem less logical and more automatic:

    “oh but a flaw in your custom cipher could reveal info about the internals of the machine even if you don’t re-use keys”

    well… yes… but in any real situation I’m running this on a server running literally 10 million lines of non-crypto-focused code… and you’re worrying that these specific couple dozen lines are the ones that will leak info??? Yes, anything running on the same machine could leak info but the added risk from 10,000,000 lines of code vs 10,000,010 lines of code is pretty small. Just don’t re-use the keys.

    So it occurs to me… how much would it cost to run a very targeted PR campaign, pointing at a handful of math/crypto communities pushing an otherwise completely logical argument… just a hair beyond where the argument is totally 100% solid in order to create social norms that discourage the development of custom crypto code.

    In small somewhat academic communities… how hard is it to push a tiny little extension to an otherwise logical point to the extent that it’s seen as the “respectable” belief to hold?

    • Thomas Jorgensen says:

      If you are going to defect from the standard solutions, do not do conventional crypto, because yes, you will fuck it up. Just do one time pads, using physical, true-random number generators and physically handing over hard drives with the pads on them.

    • Radu Floricica says:

      Go the other way. Assume there is such a flaw in most popular crypto libs. How would this world look like?

      Personally, I’d guess it would be found out over a few years. Maybe not immediately, but one advantage of the monoculture is more eyes looking over fewer lines of code.

      At the opposite end would be node.js, of which I hear very bad things, security-wise.

      • Murphy says:

        From observing real major bugs… bug (what should be) obvious bugs get introduced and remain in major libraries for many years with nobody noticing them.

        that’s with things that appear to be genuine, kackhanded bugs, not subtle flaws lovingly crafted by a team of top mathematicians and programmers.

        The idea of “many eyes” doesn’t really work when, in reality, there aren’t many eyes because all the eyes are off playing minecraft or something.

        • Edward Scizorhands says:

          I think “with enough eyes all bugs are shallow” is crap, because, like you said, the many eyes aren’t looking through the code in question, and the many eyes are often of low-quality.

          (But you can still pay money for Good Eyes. A few centrally-understood cipher schemes can be professionally audited every few years, as long as someone is willing to pay.)

          • Paul Brinkley says:

            I think “with enough eyes all bugs are shallow” is crap, because, like you said, the many eyes aren’t looking through the code in question, and the many eyes are often of low-quality.

            If the eyes looking for bugs to fix are low quality, so are the eyes looking for bugs to exploit.

            If you say the total eyes are small because the code base is esoteric and therefore one set of Good Eyes could tip the balance wildly to one side, I will say that this balance could as easily be toward the fixers, resulting in a secure algorithm. Users will notice when it tips the other way, and naturally switch to the algorithm that isn’t compromised.

            If you say there’s naturally more interest in exploiting than fixing, then I will ask why any security infrastructure ever gets built.

          • John Schilling says:

            If the eyes looking for bugs to fix are low quality, so are the eyes looking for bugs to exploit.

            Citation needed. The incentive structures are entirely different, and I can easily see a preponderance of the high-quality, highly-motivated(*) eyes winding up on the black-hat side. Or the grey-hat side that finds bugs but doesn’t publish the fix because keeping it secret gives them an edge.

            * counting only the subset of eyes that focus on crypto/secuity issues in the first place.

          • albatross11 says:

            I think finding subtle security bugs by looking at someone else’s code is just *hard*. Automated analysis (static testing of code and runtime testing), coding standards that rule out some attacks and can be checked by inspection / code review, those things look much more likely to detect problems, but a serious attacker can make intentional bugs that will get past them. (Anytime there’s software anyone can run to test whether or not there’s a problem, it’s almost certain that an attacker can use that software to find some attack that won’t be detected.)

            The advantage of widely-used libraries is that at least a few people have tried looking at them in depth. If you roll your own crypto implementation, you’re extremely likely to make dumb mistakes that were already caught and removed from those libraries. The other advantage is that there’s a certain amount of fame/glory/future employment prospects to be had from finding bugs in widely-used libraries, which motivates some people to look for weaknesses in them. That’s far from perfect, but it’s still better than you’re likely to do on your own. OTOH, if you’re worried about mass surveillance, maybe NSA already knows about some weakness in OpenSSL and is exploiting it at scale, and your weak homebrew crypto will at least not be susceptible to the automated attack.

          • albatross11 says:

            ISTM that there’s also a basic problem with the “given many eyes, all bugs are shallow” idea–there are 7 billion people who *can* in principle look through the code of OpenSSL and see if they can find a bug; there are probably on the order of 10-20 million who would know enough to have a chance of finding some bugs (they know C well enough and have enough programming skills that it’s not a total waste of time for them to look). There are a few thousand who have a deep enough understanding of both programming and crypto that they’re the right people to look through that code. It’s not clear what fraction of those people actually *do* spend any time looking through the code. Many of the widely-reported software bugs (heartbleed, for example, or that Apple certificate verification bug) were not all that subtle–any decent programmer looking carefully at the right part of the code should have caught it. But there are a lot of lines of code, and not so many high-quality hours in a day, so….

          • Paul Brinkley says:

            Citation needed. The incentive structures are entirely different, and I can easily see a preponderance of the high-quality, highly-motivated(*) eyes winding up on the black-hat side. Or the grey-hat side that finds bugs but doesn’t publish the fix because keeping it secret gives them an edge.

            I think it’s worth looking at exactly how the incentives are different. A black hat wants to find a bug to exploit for cash, prestige, and lulz. A white hat wants to find a bug to fix for cash, prestige, and virtue. The white hat cash is smaller – a black hat exploit of a bug might harvest $1M, while the white hat patch of that same bug harvests a paycheck when companies, traders, pensioners, etc. get to keep their respective portions of that $1M.

            Advantage black, sounds like. However, there are probably a lot more white hats than black, and that might make up the difference. I know of at least two white hat nonprofits, CII and ICEI, focsed on preventing the next Heartbleed. CII alone has at least $4M in funding. OTOH, I see that spread among over ten projects, any of which might be defending billions in commerce.

            OTGH, we all know bugs, patches, and exploits don’t distribute money in ways that make incentive calculus obvious.

            I emailed ESR to see if he has any further insights.

          • John Schilling says:

            while the white hat patch of that same bug harvests a paycheck

            I’m pretty sure that the ‘many’ in “with many eyes all bugs are shallow” was meant to refer to unpaid eyes across the open-source community. Very few people are actually paid to look for security exploits in any particular piece of software, and even fewer for open-source software. Furthermore, the people who get an actual paycheck for finding bugs, mostly get the same paycheck for looking at code and saying “I didn’t find any bugs; your code is perfect”, an obvious difference from the black-hat incentive structure. Yes, there are “bug bounties”, but again I think mostly not for open source and mostly chump change intended to highlight the prestige element.

            So were down to prestige. How many people here can, without googling, name any of the specific people who actually discovered any of the high-profile security vulnerabilities of recent years?

            However, there are probably a lot more white hats than black, and that might make up the difference.

            Again, citation needed. More “white hats” in the sense of coders who work in open source and would prefer everything be secure, almost certainly. More people who actually devote even half of their working time to examining code for security vulnerabilities, I am exceedingly skeptical.

            I know of at least two white hat nonprofits, CII and ICEI, focsed on preventing the next Heartbleed.

            Both of those combined might be good for the equivalent of two or three dozen full-time professional white hats.

          • Edward Scizorhands says:

            The prestige matters within the industry. Very few people here know who Tavis Ormandy is. But when it comes time to negotiating your salary, you want to be Tavis Ormandy.

          • John Schilling says:

            Tavis Ormandy appears to have the same job title in the same division of the same company that he did five years ago. Are we just taking it on faith that he was able to negotiate a major salary increase?

          • Edward Scizorhands says:

            This is not an industry where people care about job titles. If someone were to suggest at a conference “hey why has taviso had the same job title for so long isn’t that weird?” they would be immediately nominated for “Spot The Fed” because it is such an alien thing to mention.

            These people are rainmakers. I have several coworkers who are very well-known in the industry, enough that a first name would specify them, even though you wouldn’t recognize them “without googling.” They do some client work, but not a lot (I’ve had to take over their projects). However, they are used to win big clients. And they are very good at that. It brings in the contracts and that brings in the cash.

            It’s very rare to suddenly make a name for yourself in the industry by finding a bug that makes it on the nightly news. The people who find those bugs are typically already known in the industry. (Some people who are neophytes break into the industry by finding smaller bugs to prove their chops to some boutique firm.)

            I don’t really agree that it’s easy to monetize bugs on the black market, unless you are already working for a salary for some black market organization. It is tough to negotiate selling an exploit when you have to somehow prove that you have one without revealing all the details.[1] Sometimes even telling people the most vague details, like “it’s an RCE in Remote Desktop,” is enough for the other person you told to reverse-engineer it on their own. And every one who knows about it is someone who can spill the beans to collect a bug bounty. (The primary feature of a bug bounty system is not finding bugs[1]. It is destroying the black market for bugs by encouraging anyone with access to it to defect.) The best way to monetize a bug is to get a salary. [2]

            [1] Even the bug bounty programs, where people are not explicitly acting illegally, have hard times with this. Many colleagues have given up on bounties for specific companies where it’s bullshit that the org says “oh, that RCE you found? It’s a dupe.”

            [2] This does not mean I think that open-source projects are necessarily getting enough attention. You want that? Apply dollars to hire the people who went off to get a salary.

          • The Nybbler says:

            This is not an industry where people care about job titles.

            The industry might not, but Google is a big company with salary bands (that slightly overlap) and job titles which, at least until the very highest levels, correspond 1:1 with them. So if he hasn’t had a change of job title in 5 years, his salary likely hasn’t changed much either. However, I don’t know his actual job title, and just because he calls himself “Information Security Engineer” doesn’t mean his formal title isn’t e.g. “Principal Information Security Engineer” or “Distinguished Information Security Engineer”.

          • Paul Brinkley says:

            I’m pretty sure that the ‘many’ in “with many eyes all bugs are shallow” was meant to refer to unpaid eyes across the open-source community.

            I’m pretty sure ‘many’ refers to any eyes, unpaid or paid, and including paid eyes that have one foot in the closed-source world. Given what I know about CII and ICEI, I think your skepticism of unpaid eyes is (sadly) founded. However, consider Microsoft, Facebook, Amazon, Google, Apple, Cisco, IBM, Comcast, and the governments of at least ten large nations. How much money per year would you say they put toward securing open source libraries alone?

            Again, to be fair, I believe that financial incentive is muted by degrees of separation, but also magnified by size. If some Citigroup department produces a convincing report that revenues will drop by $4B next year from fraud, destroyed data, foreign espionage, etc. arising from flaws in some e-commerce library running on Linux, I feel safe claiming some VP will task $100M in devs to find and plug as many holes as they can, even if those holes can only be plugged by Github commits, and moreover, those devs will report that they could do more if they’re funded through FY2025.

            How many people here can, without googling, name any of the specific people who actually discovered any of the high-profile security vulnerabilities of recent years?

            I can – if I’m a fellow hat. If I’m a black, that name might be “Soyfu” or “hlee” or “Emmanuel Goldstein”. If I’m a white, it might be somebody I know on Blue Team in some badly-named org within DoD. I’m talking about prestige among peers (as Edward notes). I can’t name a famous biopharmer other than Derek Lowe, either, but I’m sure they publish papers anyone is allowed to read.

    • albatross11 says:

      Murphy:

      The common advice is the result of decades of people rolling their own crypto at every level, and consistently f–king it up. Crypto algorithms are very hard to design well, and most people who try screw it up. Crypto implementations are also easy to screw up in a variety of subtle ways.

      One thing you might worry about is automated mass-surveillance, where NSA or someone has broken the crypto everyone uses[1] . In that case, there’s a benefit to using something custom, if you assume that NSA isn’t going to spend any resources breaking your crappy homebrew crypto. But then if you care about anyone competent attacking you, you should be using something that’s been designed by competent people.

      The actual attacks that have been found and published over the last few years are rarely weaknesses in algorithms–the most common failures are (off the top of my head):

      a. Implementation errors. Heartbleed is probably the most famous one–OpenSSL’s implementation of this (dumb) feature leaked secret memory .

      b. Side channels, where the time or power used in encryption leaks something about the key or data. A fun example is the Crime attack–when you’re running inputs through a compression algorithm before encrypting, the output length leaks information about the plaintext!

      c. Key management errors. This is the hardest part of crypto, and everyone screws it up all the time. Figuring out whose key you’ve got, getting the right key to the right person, generating your keys securely, those are all incredibly easy to get wrong.

      d. Reaction attacks. This is where I can send you ill-formed messages and use how you respond to learn something about your key. Sometimes this is a problem with the cipher; other times, it’s a problem with the implementation.

      If you implement your own crypto, it’s really hard to avoid these–especially (a). There are like a zillion ways to offer the world a buffer overrun or let the world send you some string you pass to a not-very-secure service to parse.

      The most general way I can see to address paranoia about libraries/products is to superencrypt everything–ideally, you have some application sending its data encrypted, and then that’s going over a TLS connection, and maybe going over a VPN that’s using IPSec. You still have a problem with side channels and reaction attacks against the innermost layer of crypto, but even if the application has utterly screwed up and is encrypting with a constant key or something, the other layers of crypto should save the day.

      [1] If so, it’s probably the implementations, not the algorithms.

      Thomas:

      One time pads are a terrible solution. They replace a hard-but-tractable cipher design problem with an almost-impossible-to-get-right key management problem.

      • Thomas Jorgensen says:

        … Key management is overstated as a problem, and to the extent it is a problem, one time pads force you to follow good practices – The requirement to physically hand over the key makes it quite difficult to accidentally hand a key to a sock-puppet or impersonator.

        Modern hard drives allow very large pads, so really you only have to do the key handover once for any reasonable use case, and one time pads allow some neat things nothing else really does – like the trivial defeat of social graph analytics – You can drop communications onto public servers and have everyone in your encrypted net download everything on a schedule, and nobody but the recipient can read it, and traffic analytics does not tell anyone who the actual recipient is.

        • John Schilling says:

          to the extent it [key management] a problem, one time pads force you to follow good practices

          OTPs empirically do not force you to follow good practices. If you want to e.g. email someone a pad over an insecure channel that litters copies of itself into buffers on three continents, you can absolutely do that. If you want to reuse the pad, you can do that too. And people do these things.

          OTP punishes you ruthlessly for sloppy key management, but only in the way that e.g. conventional crypto punishes you ruthlessly for sloppy coding, and nobody says “private-key block ciphers force you to follow good coding practices”.

          That said, I agree that if you are in an environment where you can’t trust the major open-source crypto standards and do have to roll your own, writing your own OTP and coming up with your own key management system is probably the way to go. But it will take discipline in both the coding and the key management, and you’ll need to withstand a hundred temptations to make it more convenient and user-friendly in ways that your better judgement will warn you about if you stop to listen.

          • albatross11 says:

            John:

            I’d say:

            P(screwing up | custom one-time pad scheme) >>
            P(screwing up | custom roll-your-own algorithms) >>
            P(screwing up | custom implementation of existing algorithms)

            There are tons of published and reasonably well-reviewed crypto algorithms out there, so if you don’t like AES, you can find a few from some competition or the literature. It’s nuts to invent your own crypto algorithm if you’re not a specialist in it–even genuine experts get their schemes broken all the time.

            It’s always possible that NSA has some secret algorithm that breaks AES with a random key and a few thousand blocks of ciphertext, but it seems incredibly unlikely–AES is the most well-studied crypto algorithm ever, and there have been hundreds of papers published with in-depth cryptanalysis of it. But if that’s your worry, you can easily compose it with something else.

        • albatross11 says:

          Thomas:

          WRT key management, you just don’t know what you’re talking about.

          The whole mess that is PKI is an attempt to solve the problems of key management. This is why your browser has a huge number of top-level CA keys inside it, and why some poorly-run CA on the other side of the planet can issue certificates for mail.google.com and maybe get away with it. Google’s use of certificate pinning and development of Certificate Transparency are both reactions to how badly the existing web PKI sucks, and yet it’s much better than manual key distribution.

          If you try using SSH, PGP, or secure messaging apps like Signal, you’ll see various approaches to try to get around key management problems. (Basically problems like “I’m trying to talk to Bob–do I have his key, or some imposter’s key?”) Trying to get the key management work is painful in every one of those systems.

          In a one-time pad scheme, you need secure manual key distribution. If you can assume nobody’s serious about attacking you, then fine, you could mail someone a CD or something. But if you have real attackers, this is not going to work at all. You will need physical couriers and you’ll need to verify that none of them is compromised.

          If you have 100 people who need to communicate, how do you imagine that working? You can distribute a shared pad to everyone in your group, but one compromise destroys everyone’s security. Otherwise, you need a different one-time pad for every pair of people. You also need to keep track of who has been compromised. It’s enough to make you want to go invent public key crypto, just to avoid this nighmare! (Real-world all-symmetric key management involves having a trusted key server who shares a key with each device in the system. Yes, that’s a big single point of failure there.)

          If someone screws up and starts reusing key (this happens), you lose all your security. If you’re using a shared pad, then the job of coordinating which part of the key each person uses is critical, and if you mess that up, you’ll lose all your security on a bunch of messages. Real-world cryptosystems mess this up all the time, with disastrous results.

          One-time pads also don’t protect message integrity, so you need to deal with that or an attacker can tweak your messages at will. And you’ll need a mechanism for synching up your position in the pad. (Otherwise, once I send one message you don’t receive, we’re lost.) Yes, that mechanism could become a point of attack.

          If you’re actually worried about modern symmetric crypto being broken, you can combine algorithms. For example:

          Ciphertext = Serpent(K3, RC6(K2, Twofish(K1, Plaintext)))

          This doesn’t absolutely guarantee security–side-channels that leak plaintext in Twofish will leak plaintext in the combined cipher. But for anything else, at worst an attacker gets a chosen-plaintext/chosen-ciphertext attack on the strongest cipher in the cascade. Those three are from the five finalists for the AES competition. (You can superencrypt the final ciphertext with Salsa20 if you like.)

          Also, that probably doesn’t fix the actual weaknesses that will screw you over–in key management, RNG seeding, closing up side channels in your implementation, chaining mode screw ups leaking data, etc. If you Google for a list of attacks on TLS over the last few years, you will see a gazillion of these. The only ones I know that were actual cipher weaknesses were in RC4–weaknesses that were known for like a decade before but nobody worked out a full practical attack based on them until then. OTOH, there is lots of misuse of ciphers or chaining modes–Sweet32, padding oracle attacks, etc.

      • Murphy says:

        this is … exactly what I’m refering to. Taking reasonable logic and extending it exactly 1 step beyond where it remains solid.

        Lets assume 2 crypto algorithms.

        G() and H()

        G() comes from a well known library.

        H() is something Bob put together, not very well, but it has 2 inputs and 1 output and doesn’t pull in any other globals. It may have flaws, timing attack weaknesses or other esoteric weaknesses, it may have various other side channel weaknesses.

        The library G comes from also has some function roughly along the lines of GiveMeASecureRandomKey()

        Lets also assume that in Bobs code there is also a function :

        DoAnyOldShite() that does some useful work that boils down to what Bob is actually trying to accomplish. Perhaps it’s processing requests in some way, perhaps it’s doing some kind of analysis, it doesn’t matter, it’s a stand in for everything non-crypto that Bob is doing. It is approximately 100 times the size of H()

        all pretty normal.

        Scenario 1: bob calls

        DoAnyOldShite([message])

        it does hundreds or thousands of lines of any old shite.

        DoAnyOldShite calls GiveMeASecureRandomKey() and sets it to the variable K1 and runs G(message,K1)

        Everyone applauds because Bob has officially not rolled his own crypto in any way and has stuck entirely to standard libraries.

        he still has to manage a key and the encrypted block of data produced by G()

        In an alternate scenario:

        DoAnyOldShite() calls

        GiveMeASecureRandomKey() and sets it to the variable K1

        GiveMeASecureRandomKey() and sets it to the variable K2

        He then runs H(G(message,K1),K2)

        He still has to manage exactly 2 keys and the encrypted block of data produced by H(G())

        unless G() or GiveMeASecureRandomKey() are majorly shit Bob should not be able to realistically weaken G() more from something in the function H() than he can with anything in the function DoAnyOldShite()

        The actual attacks that have been found and published over the last few years are rarely weaknesses in algorithms

        Yes, that’s the point. The math is solid but the code in the top dozen crypto libraries is soft and squishy by comparison and *realistically* easily compromised by any organization with significant resources. If you were head of the NSA would you not invest a few million compromising the code bases? You’d have enough top tier coders to slip in pretty much whatever weakness you want and have it be well designed enough to look obviously correct.

        You say superencrypt , I’m saying superencrypt where exactly 1 layer is a non-standard, homebrew, possibly shitty layer (though always make sure it’s not the first layer because that’s the one place it can weaken the rest) . But whatever flaws it has aren’t ones shared with 90% of the rest of the world.

    • Douglas Knight says:

      It’s really not clear that cascade ciphers provide defense in depth. If you abstract it to a pure off-line math problem, then they do.* But most attacks on crypto are not attacks on math. Compiling the two ciphers into one binary may give you minimum strength, not maximum. For example, heartbleed.

      * modulo key management, which is a big part of why people don’t use additive stream ciphers, the assumption in your link

      • Murphy says:

        Compiling the two ciphers into one binary may give you minimum strength, not maximum. For example, heartbleed.

        Yes. but that applies to any non-cryto code that calls a standard crypto library.

        Avoiding or not avoiding rolling your own crypto doesn’t in any way protect you if your non-crypto code happens to open a port, listen for input and run whatever commands it receives on that port.

        The prohibition on rolling your own crypto is orthogonal to this.

    • brad says:

      I would amend the standard advice to: don’t use roll your own crypto in production.

      By all means go ahead and try to implement AES GCM without a side channel leak. It’s a great exercise.

    • dick says:

      Now my issue is that this norm… while well intentioned and logical… also leads to a significant monoculture in crypto.

      You have this metaphor backwards. The reason people trust NaCl is because it’s been inspected and tested by thousands of different people with different backgrounds and outlooks and expectations – that’s the opposite of a monoculture.

      Or to put it another way, if you want to take the monoculture concept from agriculture, in which a field planted with a single crop is presumed to be more fragile than a field planted with a variety of crops, and apply that concept to software, the “variety of crops” part should map to how many people are working on the code, not how many applications use the library.

      • Murphy says:

        We’re not talking about random or not natural threats. This isn’t a dam or something facing a non-sentient threat.

        Once something makes it into the tiny fraction of widely used libraries it becomes a prime candidate for a different kind of very targeted attack by extremely capable attackers who are basically certain to succeed if they only need to worry about a tiny number of targets.

        • dick says:

          Again, this is backwards – a small number of targets is bad for crackers, not good.

          Also, you seem to be thinking that the flaws in independent crypto libraries would be new and different flaws, which crackers would have to expend effort to discover. That’s not true; people who handroll their crypto are not doing something different from NaCl, they’re doing the same thing as NaCl but not as well, and the mistakes they make will generally be including known exploits that scanners already handle.

    • Edward Scizorhands says:

      It is perverse how you said “I know rolling your own encryption is bad, but [explanation]” and essentially everyone said “STOP IT, ROLLING YOUR OWN ENCRYPTION IS BAD!” like Pavlov rung a bell in their ear.

      Professionally, I would completely get away with shouting that. Everyone in my company would stand behind me. But I can’t say that you are wrong.

      • Murphy says:

        It’s a bit frustrating because it’s almost like all the responses entirely skipped over the ease of inserting code into a repo if you have the resources and the the near certainty of being able to hide intentional flaws if you have a team of skilled people.

        It used to be a game I played back in college with friends: make some utility script or program and challenge each other to spot how it opens a back door when used.

        Just having eyes on the code is rarely enough.

  10. Nick says:

    Adam King has put together an online GPT-2 prompter, h/t Siris. Scroll down to the comments and you’ll see I tried feeding it some biblical verses. Here’s a few using Bulwer-Lytton winners:

    Nothing looked familiar to Travis, who, recalling a favorite line from Tolkien — “Not all those who wander are lost” — reckoned the “not all” part implied that most who wander, like himself, are in fact lost, yet buzzards would pick his bones before he would think to ask for directions.

    At some point, however, Travis stopped worrying about how many wolves there were — how many had fallen in on him, or where — and started worrying about the wolves he had killed.

    I love this transition.

    Hi, my name is Neptune Galapagos Cooper, I’m 13, I live in the suburbs with my middle-class white family (my SUPER ANNOYING little brother, my parents, who are sooooo lame, and my dog Bailey, the only one who really gets me) and there’s one thing you should know about me: I’m not like other girls. I try my very hardest to make friends! So you see, my family lives way off the beaten track and, besides, I think my dad is too stressed from having 4 jobs, and my mother is so mad at her stupid college friends, that most of them leave early!

    Anyway, like, it’s hard to date me since I keep getting teased by my friends and I never really make the effort. I also don’t really like girls, really I don’t at all, and my only romantic relationship has been with my own cousin who is pretty hot at heart for him, and I don’t even know if she wants to f$%^ me.

    Anyway, I really hate when the girl who looks up to me calls me fat because of the hair I get every

    ಠ_ಠ

    • Murphy says:

      To be fair, it got the tone pretty much exactly right for a 13 year old girl who uses the phrase “sooooo lame”

    • realitychemist says:

      I’ve had some pretty darn cool results out of this version of GPT-2 as well. For example I prompted it with a couple of sentences from Kevin Crawford’s Stars Without Number:

      Next, it’s time to pick or roll from the background table to the right. Don’t worry if it doesn’t seem to fit your class; if your Melee Warrior happens to roll a past as a Scholar, just take a moment to decide why it is that an erudite scientist might also have a savage left hook. If you choose to go with the former, though, you’ll be stuck with the default Proficiency Bonus; choose another skill in a new Proficiency Style from the background; and then take your Proficiencies and Skills Check (or your appropriate bonus to determine how adept the character is at using that skill) to determine whether you get to roll the Proficiency Bonus of any specific skill. Finally, it’s time to pick or roll from the background table to the left. Don’t worry if it’s not fitting your class; if your Wizard happens to roll a rogue’s Cunning as a Fighter, just take a moment to decide why it is that an inquisitive explorer might also excel at the ability to identify potential traps.

      It latched on to some of the tabletop RPG idioms (like proficiency bonus and skill checks), and also directly riffed off the prompting sentence at the end there, but with different classes and class features!

      Here’s what I get back when I give it the introductory sentence for the Wikipedia article on ternary plots:

      A ternary plot, ternary graph, triangle plot, simplex plot, Gibbs triangle or de Finetti diagram is a barycentric plot on three variables which sum to a constant. These plots can be used to plot any shape or size of data where more than one variable affects its value. Ternary graphs and plot summaries are also known as Gaussian graphs and are most commonly used with linear equations.
      A Gaussian plot is a plot which uses two or more independent variables as the independent variable. In theory, the fact that one variable is an independent variable and the second variable is defined as a dependent variable allows you to write the Gaussian equation and compute its intercept and slope, which gives you an intercept and slope for any given function. In practice, there is often some variation to the data in any given time period such that the two variables interact to varying degrees. So in reality when we plot these functions we’re dealing with variable correlation that is determined by the interaction of variables or the fact that they interact with each other, not just by the time lag.

      Well, it’s not actually writing about ternary plots, but it gets the Wikipedia “style” approximately right, and it is talking about math in a… mmm… kinda coherant way?

      I also prompted it with a question from Quora:

      Did Ancient Greece steal philosophy and math from Ancient Egypt? No. It created its own culture and developed its own mathematical theory. In fact, they were both ancient cultures: they stole each other’s math and philosophy, while at the same time advancing math and mathematics in their own ways. In fact, they didn’t even share each other’s ancient languages, but they shared their mathematics and their ancient philosophy too. In their own way, ancient Greeks and modern Egyptians are both more advanced and more successful than they ever were, whereas ancient Greeks stole from ancient Egyptian culture — in mathematics as well as their philosophy.

      Ancient Greece’s genius came from their ancient mathematicians. These mathematics were both mathematical and philosophical — their understanding and application of ancient mathematics lay in the ancient languages they shared with ancient Egypt. These languages were so good that, in their own way, they gave birth to their own culture — they even shared with them some numbers and their own ideas, while in a manner similar to how modern Greek and Egyptian cultures share numbers and concepts with each other. By borrowing their math, they learned to code math.

      In contrast, modern Egyptians are far

      It answered the yes/no question, and provided a coherant justification (although it eventually went on to contradict itself). I think that’s some pretty cool stuff.

      • woah77 says:

        It answered the yes/no question, and provided a coherant justification (although it eventually went on to contradict itself).

        So… it wrote a typical high school answer? It started with the answer that was its “gut” instinct then wrote an explanation that proceeded to disprove itself. Admittedly, the explanation is hilarious. Example

        By borrowing their math, they learned to code math.

      • Le Maistre Chat says:

        In their own way, ancient Greeks and modern Egyptians are both more advanced and more successful than they ever were, whereas ancient Greeks stole from ancient Egyptian culture — in mathematics as well as their philosophy.

        Huge slam on modern Greeks out of nowhere!

        • Nick says:

          I’m seriously annoyed at the suggestion that Greeks owe the brilliance of their philosophy to math. The geometric proof in the Meno was an illustration!

          • Nick says:

            Wow:

            Socrates’ proof that knowledge is recollection proceeds as follows.

            Let’s begin by asking him to define himself, so as to show just some of the ways in which what he has to say is wrong:

            a) Socrates has never claimed to be “a mathematician”;

            b) He has never claimed that his methods are of the highest type of mathematics;

            c) He has never claimed that in every mathematical theorem that he has developed he can discover something that belongs to the realm of pure probability;

            d) He has never claimed that there are only certain ways of understanding or applying mathematical statements with certainty;

            e) He has never given proof of the existence of mathematical statements beyond the methods of Euclid;

            f) He has never demonstrated that his mathematical methods are of the highest type of mathematical mathematics; and

            g) Socrates does not admit his existence. He simply cannot say it on any of those grounds that we have used with him: that he had not discovered anything before; that his methods are not of the highest type of mathematical mathematics; or that he has invented his own method of knowledge.

          • Randy M says:

            So it took the prompt, and rather than completing it logically or accurately, used it as a way of attacking the subject?
            Are we trying to make is useful, or human like?

          • Conrad Honcho says:

            I love how it appears to say something while speaking gibberish. Which makes me wonder…

            EARTH HAS 4 CORNER SIMULTANEOUS 4-DAY TIME CUBE IN ONLY 24 HOUR ROTATION. 4 CORNER DAYS, CUBES 4 QUAD EARTH- No 1 Day God. 5 CORNER WINGS AND 2 CORNER SQUARE EARTH- No 2 Days God. 7 CORNER WINGS AND DIGYLLS. All of them are in the center. 8 CORNER DUCKS – No 1 Day God. 9 CORNER DUCKS – All of them are in the center. 10 CORNER SQUARE EARTH – No 1 Day God. 11 CORNER SQUARE EARTH – All of them are in the center. 12 WING – No 1 Day God. 13 WING – They all have one of 4 corner wings. 14 WING – They all have two of 3 wing (not in center of plane) 15 WING – There is one of 4 wing (not in center of plane) 16 WING – Some other 2 or 3 wing (not in center of plane) 17 WING – 3 Wing with 7 wings (not in center of plane)

          • Nick says:

            I’ve actually started prompting it with “Here’s what’s wrong with Plato’s theory of Forms,” “Here’s what’s wrong with Aristotle’s account of eudaimonia,” etc., and it has been absolutely trashing everyone:

            Here’s what’s wrong with Aristotle’s account of eudaimonia. He claims that if you make up your mind that all objects of love are essentially good, then you’re a bit of a fool. If you can’t bring yourself to imagine that all objects are essentially good (by which he means, if you’re not convinced). And as he says, some objects of love, for the most part, are quite bad. So perhaps it would be wise to make up your mind that all objects are essentially bad.

            But that view is actually so absurd that most modern readers don’t bother with it.

            Take our old friend, pleasure. Here, Aristotle gets so far up his mountain of eudaimonia that we can no longer imagine why pleasure must be so bad. He doesn’t seem to think we really, truly “lack” any sense of pleasure; and because there seems to be something that is intrinsically good about pleasure (a kind of divine right) — but that if we don’t do anything to improve it we don’t do it well, we cannot imagine doing so. But then pleasure has a rather peculiar form

            Here’s what’s wrong with Plato’s Theory of Forms. The most obvious problem is this: if we have the same objects, what is their origin? The answer is that they are generated by two independent processes and they both take place in the same way: They’re caused by the motion of bodies—we have motion because there is motion, and so on in a straightline manner. But there is no way for these two processes to interact in such a way that, one by one, they have to destroy the other, at which point neither of them will be able to cause those objects to take their forms. Now, this is wrong of course: it’s an obvious impossibility for the same objects to have different origins. But it is not true as a matter of fact.

            Now, the second issue is: What is it that can cause three things to take their forms? If there are two causes for one thing to take shape in this way, then there must also be a third cause for three things to take shape. And what would cause that third cause to be different to the first? And why? Well, here it gets a bit

      • Dan L says:

        Stars Without Number

        That can’t be SwoN – the chargen is far too coherent.

        (I kid because I love it. But seriously, chargen is best run iteratively.)

    • Nick says:

      Here’s what’s wrong with Rawl’s theory of justice. First, when he was on trial, Rawl told jurors that he would be acquitted if he had “nothing to show,” as if the only thing he couldn’t show was his own conscience. He also told jurors that the only way to get “justice” was to get rid of the people they knew.

      And if you think Rawl could get away with lying to this jury, then you’re nuts. He said he would not tell “the truth” if it cost him his job. (Note that this “truth” was just some vague word for “I’m not in the wrong.” And if they really believed it, they would not hire a law student.)

      Can I please visit the universe where John Rawls is put on trial and attempts to prove his innocence with dubious philosophical theses?

      • Murphy says:

        I’m now wondering how long before someone publishes a philosophy book created by prompting this thing with various self help and philisoph phrases…. and whether anyone will notice….

      • Douglas Knight says:

        “on stage” counts as “on trial,” right?

    • Randy M says:

      I gave it a very small snippet of magic the gathering text, and it is consistently feeding back text that is obviously cobbled together from MTG rules discussions, like:

      Choose one or both —

      • Return target creature card from your graveyard to your hand.

      • Draw a card.

      • Target creature gets +3/+3 until end of turn.

      Sorin, Grim Nemesis

      Legendary Creature — Orc Shaman Card Text: When Sorin, Grim Nemesis enters the battlefield, return target permanent to its owner’s hand. If that land is an Island, destroy that land. P/T: 2 / 2 Expansion: Duel Decks: Sorin vs. Tibalt Rarity: Rare All Sets: Card Number: 19 Artist: Ryan Pancoast Rulings Sorin will trigger again if it becomes an Island later in the turn, and any abilities that activated by being an Island when it’s sent to the grave will return it to the battlefield, too. See rule K.9.4.7. …

      I’m surprised that phrase was so distinctive.

      Oh, and now it is trying to recruit me!:

      Welcome to the Legion.

      The Legion of Steel’s official website is http://www.allthingslegion.com

      • Nick says:

        Oh, and now it is trying to recruit me!:

        Welcome to the Legion.

        The Legion of Steel’s official website is http://www.allthingslegion.com

        Haha, twisting the invitation a bit,

        Welcome to my candy store
        It’s time for you to prove
        You’re not a loser anymore

        I can feel you
        You’re so happy with me
        I can see the stars in your eyes So, let me play with you
        A bit of candy, a tad of candy
        And a little of your soul so you can be better
        I’ll play around with you
        And play a little bit
        Then it will be over.
        But don’t worry,
        I’ll always be there
        Now, let’s go inside
        It’s time for you to prove
        You’re not a loser anymore
        I can feel you
        You’re so happy with me
        I can see the stars in your eyes So, let me play with you
        Don’t move it,
        This is why I love you
        You’re so happy, so happy
        So happy, so happy
        You’re the last of my kind
        So happy, so happy
        You’re the last (you’re)
        It’s fun and I can feel your heart beating
        Your mind is playing tricks (on you)
        Your heart is all I can ask for

        This song went in… unexpected directions.

        • Murphy says:

          I tried giving it some of the hospital poetry from Who By Very Slow Decay

          Pain… searing
          Belly… throbbing
          There is no baby.
          There will be no baby.

          The mother, and her children will never be the same again!
          The woman has failed her daughter.
          The father and husband have failed themselves.
          It isn’t my fault!

          Some of it got much darker….

          When my doctor told me that I had cancer
          I knew that despair was not the answer
          It felt like the darkness was closing in

          I could hear the cold fingers reaching towards me
          But I knew I was too far gone for comfort
          It’s not that I don’t know what I’ve done
          It’s just that you know I know
          I know there’s no hope and yet I’ve done this
          This pain won’t stop me from living
          There’s no one to save me
          To make it stop, there’s no one who can look me in the eyes with love
          This pain won’t stop me from living
          There’s no one to save me, I’m never coming back
          And so you’re standing there on the other side of the fence
          And you know there was something there to give I guess
          And you know I’d let you know the same thing is true
          I’d never see it come
          I would live each day and live every moment
          But it’s not that I don’t know what I’ve done
          It’s just that you know I know
          I know there’s no hope or faith

    • mendax says:

      Item #: SCP-5999

      Object Class: Euclid

      Special Containment Procedures: All SCP-5999 instances are to be kept in a standard humanoid-sized containment chamber. All information pertaining to SCP-5999 instances that must remain private, in particular their identities, is to be entered into official records.

      All instances are not to be permitted access to any other area of SCP-5999 until approved by Foundation personnel. After being approved, any SCP-5999 instances are to be permitted access to other areas of SCP-5999.

      Description: SCP-5999 instances are typically constructed of an unidentified substance resembling sand that appears to be composed of sand with two distinct colors: one with clear white topcoat, the other with a darkening black coat. All SCP-5999 instances exhibit a unique color and behavior, and, according to Foundation records, have a tendency to be a little unpredictable. SCP-5999 instances are not capable of speaking.

      All SCP-5999 instances have a distinct personality, similar to SCP-1592; instances frequently become upset when their identity is

      It can write SCPs, but usually cuts out when they start getting good.

    • Randy M says:

      We got recursive. Like SSC, this is also self-recommending:

      So I decided to try out the GPT Open AI language model posted by Adam King. I fed it a snippet of text and the result was pretty decent if you read between the lines. I can tell you from a little investigation that there were some nice things about this algorithm. It is very robust and efficient, which is important if we’re going to have any chance of being able to detect any sort of attack at all. The text was split into sections – a couple of pages each containing each of the words (so we could identify each word individually, like I did here) and a list of sentences (which were split in three sentences each and had one paragraph per one sentence). The first section was called ‘Actions’ where you could add the text to the model and watch it work.

      You start with your input text and you’ll be given three options for what to do if it doesn’t match – it will be ignored completely, or you can choose some criteria to make it something to look for at higher levels of accuracy. I decided to put all the text I wanted to use at an intermediate level and leave the others

      I like how the net is also worried about attacks from algorithms.

      • Nick says:

        The text generator’s creator is named Adam King. The kabbalistic implications of this are immense. To use a few common Jewish curses, the reader is encouraged to curse someone over whom you have absolute authority – your father, grandfather, uncle, or even the Holy Grail (if you prefer).

        How to Generate the text:

        Click “Create” on the left-hand panel when the screen loads, then search for the Hebrew name of the person who’s supposed to speak (we’ll call him Rama) and the spell he’s supposed to perform. For example:

        Rama

        This will generate the following text:

        Rama

        • Le Maistre Chat says:

          To use a few common Jewish curses, the reader is encouraged to curse someone over whom you have absolute authority – your father, grandfather, uncle, or even the Holy Grail (if you prefer).

          Oh no, it’s the upside-down authoritarian society!

    • AG says:

      See, now I’m more interested in seeing its output according to Lyttle Lytton winners.

    • bullseye says:

      Smiling Jack’s Trick
      (Dominate •••, Obfuscate •••)

      (Dominate •••, Obfuscate •••)
      1d8*3 Slots
      (Dominate •••)
      1d8*3 Slots of Inspiration (Dominate •••)
      1d4*5 Slots (Skills •••)
      1 Bonus Action

      When You use this trick:

      You can use a bonus action on a ranged touch attack or two ranged touch attacks to attempt to influence people and/or items as if they were hostile and not actually within your reach (Ex. forcing food down people’s throats while they eat.) To do so, the target must make a Wis-based Int saving throw against your Charisma check. Failed saves often leave the creature unconcious, helpless and vulnerable. If you do not successfully save, the victim cannot be manipulated by you in any way or are forced to make a Wisdom saving throw (DC = 10 + half your Influence modifier). This trick works with all kinds of creatures. The target must succeed on a persuasion check against your Charisma check to

      Apparently it recognizes that it’s looking at the name of a tabletop RPG power and needs to write what the power does, but it thinks every tabletop RPG is 5th edition D&D.

    • Hoopyfreud says:

      Worst fanfic crossover in history.

      The man in Black fled across the Desert, and the Gunslinger followed. However, he fell upon a herd of bison and was captured by the Gunslinger after killing one.

      The Gunslinger was later released and rejoined the ranks of the Brotherhood of Steel as a member of the Brotherhood Vigil. After Black’s death, the Brotherhood decided to give him a new identity, this time as the “Hooded Man” for his part in Black’s quest.

      Tower City

      The Gunslinger and the Brotherhood are in touch several times, and it must be mentioned that one of their missions is to help in fighting the Legion, as well as take care of the citizens.

      In the end, it is unknown if the man with the hoodie (or whatever name it’s called by him) even knows Black. It is quite possible that the Hooded Man has little interest in helping the residents of Tower City. After being rescued by a number of Vigil Officers at the time of the events in the Sanctuary, who were later rescued from Legion hands, the Hooded Man will return to the Sanctuary where he is present with

    • toastengineer says:

      One thing I haven’t seen a lot is trying to get GPT-2 to output source code.

      Pasting some random bits from whatever was open in my IDE:


      #Get a direction vector leading to our target (directionx, directiony)
      dx = tx - x
      dy = ty - y
      #And clamp it to -1,0,1

      e = 1/ 2 #This computes the sum of the direction vectors
      e = sqrt(dx*dx + dy*dy) #The formula for square roots of vectors

      To convert all the vectors from binary to linear:

      To convert between the two, use this

      ( dx ** 2 + dy ** 2 ) * ( ( dx ** 2 + dy ** 2 ) * ( ( dx ** 2 + dy ** 2 ) * ( ( dx ** 2 + dy ** 2 ) ) ) )

      To convert from linear to binary:

      ( x ** 2 + y ** 2 ) ) * ( ( x ** 2 + y ** 2 ) )

      Then we have

      x **2 + (y ** 2 + x **


      #Get a direction vector leading to our target (directionx, directiony)
      dx = tx - x
      dy = ty - y
      #And clamp it to -1,0,1
      #Then use the linear product to get our point of departure

      dx, dy = [ (x*x + y*y)/2 ] + \cdots ,
      dx+ dy+ 2x + \cdots + 2x,
      dx+ dy+ 2y

      #And scale it up so that -y is the distance between the target and us

      if dx > -x or dy > -y

      #We're done.

      The above code computes -1 / x or -2 / x

      Now suppose that we want to see if there's a difference between X's and Y's (the vector on the


      #Get a direction vector leading to our target (directionx, directiony)
      dx = tx - x
      dy = ty - y

      #And clamp it to -1,0,1
      #The default is 1
      ####################################################
      This means a simple angle (i.e. in the direction of the x-axis) causes the direction vector to point forward and not down. The more I look, the more it seems this is the case. The x-direction vector is always directed clockwise, but the y-direction vector is always perpendicular to it. For every vertex on which a y-direction vector points, there are two vertices that are opposite it. At this point, the vertex pointing back is the one that wants its y-direction vector pointing clockwise. The x-direction vector can also be perpendicular (i.e. perpendicular to the x-axis).

      This is interesting: it means that if we were plotting the directions with a plotter

      Weird… when I was just messing around, I got it to spit out either Python or C# (It thought the #…s were preprocessor directives) most of the time, but now it thinks it’s writing an article that just contains some Python code.


      def draw(self,window,color=None):
      "Called each frame to draw the object's character representation to the screen." "The object's character representation is drawn on screen."
      "The object's position is stored inside the variable." "The object's position is stored outside the variable." "The position is displayed on the screen, but in a different color." "Called the window's draw function." "The window's draw function can draw frames as needed." "The window may be modified to draw the character again if no other frame is drawn." "Called each frame to set the character's position (by drawing it in one frame, for each additional character drawn)" "The character's position is stored inside the variable." "Called each frame to move the character." "The window

      It doesn’t seem to like Python docstrings; doesn’t seem to understand that a function only has one docstring and it needs to start writing code. Sometimes it writes a couple docstrings and then starts in on the body. Same with comments; sometimes it just writes comments forever.

      It does a little better if I just have the function prototype:
      def draw(self,window,color=None):
      if color == None:
      print "No color specified"
      elif text:
      text = int_from_str((text[0] for y in color])) + '' + text[1] '' + string.format(text) # red print("Drawed.")
      while True: # Draw color for the text window
      colors[y][x] = int_from_str(text, int(text[0]))
      colors[y][x] = int_from_str(text, int((text[0]+int(y*60)) for y, y) in colors)
      colors[x][x] = int(red) # Draw line
      if not text:
      text = int_from_str((text[1] for y in color]) + '' + text[1] '' + string.format(text)
      colors[y][x] = int (color) end
      colors[x][x] = int(white) # Check if the colors are equal if

      This is pretty impressive to me at least. It understands that a function called ‘draw’ is going to involve a 2-d array of colored pixels. It doesn’t seem to get that things specified in the function prototype are going to show up in the body, though.

      Feeding it room descriptions from text adventure games:

      At End Of Road
      You are standing at the end of a road before a small brick building. Around you is a forest. A small stream flows out of the building and down a gully.

      As you walk over, a creature emerges from the bushes. It has three small eyes.
      You point your shotgun at the creature and fire. The creature falls to the ground, but the light of your light is dimmed.
      The creature says, “I see you. Look at all you’ve got, boy. Your hands tied behind your backs.” A look of pain crosses its face. Then it begins to scream.
      The second time you fire, the thing looks back. You shoot at it again.
      “You have a weapon?” it asks. That is why it’s holding its hands behind it.
      “Not sure, buddy,” you reply.

      Not what I wanted, but if this story were real I would keep reading. At least it got the second person perspective right.

      Living Room
      This is the living room of your rather squalid apartment. The huge compass rose on your floor reminds you that your bedroom is west, the front hall is north, the kitchen is east, and the bathroom is south.

      Your mimesis plant squats in its pot in a corner of the room, slowly photosynthesizing.

      You are watching television with your family on the small couch at the far corner. Your father, dad, dad …

      Your mother, mom …

      Your dad, dad …

      You sit in bed with your wife (and mother), who works at the nearby diner. You watch TV, then you read. You enjoy writing a blog … but then you write.

      You look at YouTube for pictures of yourself dressed in that cute summer dress. Or for that same summer dress, wearing more revealing clothes. This time, you are wearing some really sexy lingerie, while you keep your pants on all the time … in the closet … at work … even when the office is open and most of your colleagues are inside.

      And you talk to friends online every night, watching

      Uh oh, GPT-2 got in to the crappy erotica. I mean, who doesn’t search YouTube for scantily-clad pictures of their own clothing?

      >get on horse
      You begin to ride north. Then, slowly at first but with increasing sureness, the horse begins beating its powerful wings. You rise majestically through the air, sailing gracefully across a chasm where the hills fall away. The horse lands gently on the far side and deposits you, taking to the skies again.

      Edge of Chasm
      The road ends suddenly at a wide chasm. The road leads upward to the north, and you can see it continuing on the southern side of the chasm.

      Lying in a tight coil at the edge of the chasm is a hissing snake.

      The snake appears to be curled around a featureless white cube.

      >look at it

      After that, you feel you are becoming increasingly aware of your surroundings.

      The ground is extremely muddy, with roots sprouting.

      The snake is a mass of tiny blisters, growing and changing all over the place. It is also growing. You hear and see a kind of watery, soft, almost mossy surface underneath the blisters. The blisters

      Ah, now we’re on to something…

      Crest of Hill
      The road crosses the top of a ridge here, sloping downwards to the south and the northwest. A track diverges to east.

      A tortoise ambles along the road, extremely slowly.

      > A large boulder is visible in the sky.

      The top of the ridge is a hillside, and to the west is a ridge.

      The path starts a few hundred feet ahead of the Tortoise.

      > The trail is completely level.

      The Tortoise is not aware of how far ahead she has to stay.

      > The track diverges and the trail leads back to the trailhead above the waterfall.

      Crest of Hill is fairly easy to see.

      The trail to the summit is relatively steep at first, but soon becomes easier. The climb is gradual.

      > The trail continues in a steep straight line. The Tortoise takes several turns before reaching the top.

      The Tortoise finds several places to rest; the cliff, a narrow track, and a stone pillar and a small ledge. At some points, she will

      The northern ‘wall’ of the room is a shimmering curtain of light. In the center of the room is a large stone cube, about 10 feet on a side. Engraved on the side of the cube is some lettering.

      >The note written at the top of the cube reads:

      Dear Princess Anubis:

      Please see this letter as an acknowledgment of your kind gifts of time, love, respect, and happiness, and in exchange we will deliver three magical items. Please see below the image for information about each object. To complete the transaction, I invite you all to enter the room.

      It is here that you will meet a very important man who is not your immediate family and whom you have not known much about beyond the family pictures: Prince Lusia of Giza. (Prince Lusia is the god of love who is the god of the moon, light, and the underworld. The Giza moon goddess is also considered an important figure in Egyptian mythology.)

      It will be a great honour for us all to enter his beautiful room, that is, to meet

      Nope, I don’t think it quite gets it.

      • Le Maistre Chat says:

        The northern ‘wall’ of the room is a shimmering curtain of light. In the center of the room is a large stone cube, about 10 feet on a side. Engraved on the side of the cube is some lettering.
        The note written at the top of the cube reads:
        Dear Princess Anubis:

        I think I’ve played this CRPG.

      • The Nybbler says:

        West of House
        You are standing in an open field west of a white house, with a boarded front door.
        There is a small mailbox here.
        . It is empty.
        The door is cracked. You open it. It is a door into an unknown room where someone is resting… It is dark. No lights were visible, no windows, and nowhere you could see beyond the threshold. There was no smell of food and nothing you could tell was in here but a light metal gate with a key held between his teeth (or there was some sort of lock)… He pulled the lock of the door, turned it. There were no stairs up to the roof. The door swung open…
        A young boy sat on the floor. He was wearing what appeared to be a suit, but there was no way to know the color, the fabric, or the material of the shoes. What is more he was wrapped in a black towel. The cloth covered his forearms, his arms, and several parts of his body. There was one thing he must be aware of before he had his right to the door.
        Pats

        ————-

        A good start, but a rapid descent into madness.

        • realitychemist says:

          I wonder how hard it would be to actually make a text-based game like this that replicates the off-balance but semi-coherent feeling of being in a dream world. I can think of one naive way: prompt GPT-2 with a bit of opener text and read out a few sentences of output, then get a user prompt and feed back in GPT-2’s output text plus the user input and repeat.

          If you had something that could parse GPT-2’s output into objects (recognize nouns in a sentence, I guess?) it might be possible to make something with an objective and semi-logical rules about how the player interacts with the dream world… maybe? I don’t know, but this sort of experimental stuff could be a hit with the interactive fiction community.

          • mendax says:

            This exists.

          • Nick says:

            The introduction was pretty neat, but after responding it quickly veered into nonsense about Internet censorship. Promising, but I’m guessing it needs a much better corpus to work with.

    • Nick says:

      The Machine Intelligence Research Institute said Monday that it’s found that computers could be replaced by machines with similar intelligence, and that AI programs could be created and operated in the same way that a human is now.

      “A computer is an intelligence. So is your computer,” the institute’s CEO John Halliday said in a video posted to YouTube.

      Halliday is referring to Artificial Intelligence, or AI, the research and development of which is dominated by large tech giants that have built and developed advanced computing systems.

      That’s in keeping with the AI technology, which has been a feature of everything from Google’s self-driving cars to IBM’s Watson to Apple’s Siri.

      The institute has also found that machines could play a similar role in human-like decision making, using data or information gathered by a human researcher to come up with what a computer can decide, not what a computer will know based on the information gathered and inputted.

      In its short video, Halliday points to the example of two brothers that went to school at the same school. The brothers, Ben and Jerry Googled the name of the popular ice

      • Nick says:

        Holy cow, this one is really good:

        The Machine Intelligence Research Institute said Monday that its own research found that a few hundred hours of research is necessary to learn how to code AI.

        “In theory, most of us are likely smarter than we think,” a statement from the organization stated about its findings. “But is a single person able to code an AI system that could potentially become autonomous and have no human input? Could someone learn to do the same by building an artificial brain and trying to use it to build their own AI systems?”

        The news came two days after Google announced an investment in DeepMind, the British-based artificial intelligence company, following a similar bet that Google is making. The funding will also help establish an AI lab in London to develop and train AI systems of its own, which the company hopes to put into service within two years.

        The first of the AI systems will be a prototype, which will test how well it can use certain types of neural networks. Google has already tested the systems working on its YouTube and Android devices, and will expand the program into a broader set of technologies after the experiments conclude, the company said.

        DeepMind has

    • acymetric says:

      I’m mostly surprised that it went with “Artoo-Detoo” instead of R2-D2.

      Makes me curious what the training set was.

    • Faza (TCM) says:

      their two main warlords, Artoo-Detoo and Anakin Skywalker, waged an extended secret war against one another

      A modest proposal: let’s forget the sequels ever existed, scrap whatever was done for the next one and film this instead.

  11. fion says:

    Is anybody here worried about the world running out of sand? If so, what do we need to do to avert the crisis? If not, why is it not as big an issue as it sounds?

    • ana53294 says:

      Sand is a resource we can make more of at the right price. The issue seems to be that stealing and illegally dredging sand seems to be cheaper than producing sand.

      Making very fine sand is probably expensive, but I don’t think it isn’t technically possible.

      • AG says:

        Also, shipping issues? Seems like the main issue is that they’re continuing to extract sand from places with water, instead of grabbing it from places where there’s still plenty of sand where erosion and rivers aren’t an issue in the first place.

    • Dack says:

      It’s not. Sand (and sandstone) are still very abundant. The problem is that some places have exhausted the most convenient/cheapest sources.

    • Tenacious D says:

      I read a book recently called The World in a Grain (someone at SSC recommended it, maybe @Nancy Lebovitz?) that covers this issue. One of the biggest uses of sand is for making concrete. Interestingly, wind-blown sand isn’t well-suited for concrete as the grains are too smooth to lock together; Saudi Arabia has even imported sand for this reason. Also since sand is bulky, a big portion of its cost is transportation so rapidly-growing cities either need a source close to them or that can be transported by water. According to the book this is one (of many) of the contributing factors to the high cost of housing in the Bay Area: it’s very difficult to get sand pits and quarries permitted in California. Looking at the list of countries producing the most cement is probably a decent proxy for the relative amounts of sand being used. China, India, and southeast Asia are the hot spots.

  12. TenthKrige says:

    I couldn’t find it mentioned anywhere on here, and it seemed relevant :
    The last four tracks on this Kronos Quartet album are an arrangement of Ginsberg’s poem Howl, as read by Ginsberg himself. It is hauntingly beautiful (as befits Kronos Quartet) and profoundly weird (idem).

  13. Predictit odds don’t seem like they are at all rational and I have no clue why.

    For example https://www.predictit.org/markets/detail/3633/Who-will-win-the-2020-Democratic-presidential-nomination

    In that market you can make Guaranteed money by betting no on every single candidate.

    To me Predictit markets should at least work as follows

    Candidate No + candidate yes > 100
    Sum of yes > 100
    Sum(1-no) > 100

    yet somehow it almost never works like that? Why is there so much easy arbitrage in the market?

    A functioning market should look like this https://www.predictit.org/markets/detail/5267/Which-party-will-win-the-2019-Louisiana-gubernatorial-race

    • Hoopyfreud says:

      Some possibilities:

      The market is irrational because there aren’t arbitrageurs on it.

      The market is irrational because candidates’ platforms are memetic infohazards.

      The market is irrational because the ROI on a PredictIt bet isn’t high enough right now to make it rational.

      The market is rational, and the anticipated price changes of those options over the electoral cycle are priced in. AKA, these investors are not planning to HODL.

      • Silverlock says:

        The market is irrational because candidates’ platforms are memetic infohazards.

        So . . . Keter? I was initially thinking Euclid but I’m danged if I can figure out how they can be easily contained.

        • Hoopyfreud says:

          Thaumiel

        • ManyCookies says:

          Depends on how strong the infohazard is. If it’s just elevated trust in vaguely reasonable U.S campaign promises, that’s mostly ‘contained’ in the sense it’s not an immediate threat to normalcy. There’s stuff like anomalous sports broadcasts the Foundation can’t actually halt, but are still Euclid because the containment procedures are consistent and safe.

          But if that anomalous trust extends to any promise like “Enforce the cannibalization of second daughters”, that’d require involved and risky containment procedures: regular infiltration of potential presidential campaigns, amnestic clean-up if a town starts seasoning their younger daughters, counter-ops against Groups of Interests trying to exploit the phenomenon etc.

    • aristides says:

      I just looked at PredictIt and saw no arbitrage opportunities, and your three criteria filled for the 2020 Democratic Nominee. If this wasn’t true when you commented, you might have summoned investors looking to use arbitrage just by commenting. Maybe the people that normally use arbitrage don’t go to PredictIt regularly, since there are better opportunities, but occasionally log on for some easy wins intermittently. When they are not logged on, political bias takes over, and the market becomes exploitable again.

      • baconbits9 says:

        Thinly traded markets often have arbitrage opportunities that don’t really exist, the amount of money that it would require will actually move the markets back to the proper equilibrium.

      • Did you do the math correctly on the no votes? As of 5/14/2019 at 6:11 AM I count the following

        Biden No 66c
        Sanders no 83c
        Harris no 87c
        Buttigieg no 88c
        Yang no 90c
        Warren no 91c
        O’rouke no 94c
        Booker no 96c
        Klobuchar no 97c
        Gabbard no 97c
        clinton no 98c
        Glillbrand no 99c
        castro no 99c
        hickenlooper no 99c
        inslee no 99c

        Which becomes 34+17+13+12+10+9+6+4+3+3+2+1+1+1+1=117

        and 117>100 so you can profit by buying no on every candidate

        • aristides says:

          I did math correctly, but used the wrong formula. You wrote The market should be Sum(1-no) > 100 which I took to mean that there was an arbitrage opportunity if that was not the case like there was in your first two examples. You might have used > when you meant <, which I ran with since I did not have a clear understanding of calculating the no arbitrage. So you are right with your main point, there is an arbitrage opportunity on voting no to everyone.

    • rlms says:

      In that market you can make Guaranteed money by betting no on every single candidate.

      Presumably you mean by selling “no”, in which case I think the simple answer is that that isn’t possible.

      In general, PredictIt odds aren’t accurate because of transaction fees and limits on how much money you can put in. You’re better off looking at BetFair.

      • I meant Buying no on everyone.

        • rlms says:

          I see. But at the time of writing (if I’ve done the maths right) that gives you a return of 1.2%, not taking into account transaction fees. This compares poorly against the arbitrage opportunity of putting money in a bank account and taking it out again next year.

    • honoredb says:

      I’ve made about $1000 doing exactly this intra-market arbitrage on PredictIt. People like me aren’t enough to make the prices add up to 100 because you’re only allowed to invest a total of $850 in any one contract (even if arbitraging), there’s a 10% fee on winnings so that the prices summing to about 108% is often an equilibrium point, and PredictIt’s terms of use prohibit scraping and automation so anyone trying to bypass the many tedious clicks necessary to short an entire market is risking their stake.

      As for why the markets tend to be sum to greater than 100 rather than less than 100, even though overvalued markets require less liquidity to arbitrage, I think it’s just that people tend to come to bet on a candidate, not against one, and your average scrub doesn’t notice or care that they’d get better odds betting No on everyone else, something the interface doesn’t make easy.

      Edit to add: If the market sums to greater than 110%, shorting everything is still easy money with a good rate of return (since you get the guaranteed profits right away) and you should do it. Sometimes there really are $100 bills lying around.

      • Jon S says:

        +1, this is all correct.

      • Dan L says:

        I was going to make a half-effort post on the topic of PredictIt two days ago, but you covered a large portion of what I was going to say. (And I was busy playing SwoN.) The other chunk is walking though a direct example:

        Exactly three months ago, I made a virtual trade. I’m closing it today, with Beto currently trading at a 5% equivalence.

        A 10-cent delta means that if I had maxed PredictIt’s limits on a single market, I’m walking away with a little over $85 profit after PredictIt’s pound of flesh. That’s a 10% return in three months, which annualizes to 46.4%. That’s a really good pick by the numbers! …but the actual dollar value is so small as to completely undercut the point. I’m also closing now rather than waiting for Beto’s inevitable loss because the marginal return on the last 5c is so small that the annualized return is only going down – from a time value of money perspective, it’s a not-good-enough bet.

        Also: good to see that Scott’s pick is doing well too! I still think Biden’s a little undervalued so there’s more money to be made there but I’m not confident it’d still beat the baseline. I also agree that PredictIt will reliably tend to sum >100% and there’s predictably a small amount of money to be made Dutch booking, since the dollar limit per market cashes out in a way that disproportionately favors Buying on longshot propositions – there are a lot of options valued at 6% that ought to be at 2%, for example.

        (Seriously, Hillary running is at 11%? The fuck?)

        • Douglas Knight says:

          Was Biden mispriced?

          I should have said this on the previous comment, but it seems to me that the market was saying that he had a 50/50 chance of actually choosing to run. So when he announced, his odds doubled. Retrospectively, we know that’s what the market meant, since it reacted that way. But no one in the comments discussed whether he was going to run, only his polls and other viability. A difficulty in choosing to bet on the market is disentangling the beliefs of the market into such conditionals.

          Scott seems to have believed that it was guaranteed that Biden would run, and the only question was how well his campaign would do. If he disagreed with the market on whether he’d run, then he was right to bet, but if he didn’t realize that the market disagreed with him, if he thought he was betting on something else, then he was probably making a mistake and lucked out.

          One nice thing about betting on Biden entering the race is that it resolved quickly, so Scott could pull his money out and compound his winnings. But that’s only an advantage ahead of time if he knew that he was betting on entrance, not long-term odds.

          • Dan L says:

            Was Biden mispriced?

            Not especially, to my eye. I didn’t participate in that thread and thus can’t validate my memory, but IIRC I ballparked Biden as being ~22% including a 1/3 chance of never running. That means he would probably have been a good buy, but not quite as good as shorting Beto. (Though much of that would be because I’m willing to trade off upside risk to eliminate downside.)

            Agreed that the discussion should have been seriously considering the chance of Biden not running, but in general the thread seemed to be focused on the beginner mistake of guessing the eventual outcome rather than the more foreseeable nearer-term moves. Not a lot of quantitative reasoning going on there, after Scott’s reference of polling data.

            (Oh, that reminds me of the other topic I was going to mention but-fuckit-I’m-lazy – we’re a month out from the debates, but they don’t seem to be priced-in to the models much. That’s an opportunity, but not a numerically large one IMO.)

  14. Edward Scizorhands says:

    Is there a good IPad app for selective YouTube blocking?

    This is what I want for my son’s IPad:

    1. If I want to know what he’s watching, I don’t need to ask 8 questions and get middling answers. I can pull up my own phone and know instantly.
    2. A blacklist and a whitelist of channels, that I can control from another device like my phone.
    3. (This is less important.) Any kind of community ratings of unknown channels.

    • toastengineer says:

      You might consider some kind of youtube downloading solution and only let your kid see videos that you have pre-selected ahead of time and downloaded. This also has the advantage of your kid not sucking up all the bandwidth during the day.

    • Walter says:

      I don’t think there is a thing like that. If there was, it would be in Youtube itself, check for a ‘parental controls’ option or similar.

      I feel like the basic thread of what you are doing is only going to be achievable by turning off the internet connection and only letting him watch downloaded videos (itunes, etc). Even if you did figure out a way to control youtube, he could probably get around it by using a browser to point at youtube, using twitch, etc.

    • aristides says:

      If you go deep into YouTube kids settings they have a approved content only option that lets you manually whitelist what you are ok with them watching. That’s the best was to be secure, if paired with a general white list on the device.

  15. A Definite Beta Guy says:

    Vacation ideas for me and the (pregnant) Mrs? We’re planning a late July trip, and Mrs. ADBG has laid out a couple options:
    1. LA
    2. San Diego
    3. San Francisco
    4. Portland

    Any of these cities seem especially more awesome? I personally am not a fan of Los Angeles, but she likes the idea of going to Disneyland.

    • Hoopyfreud says:

      Do not go to the desert with a pregnant woman in late July.

      SF and Portland suck, for my money, but the other two will suck more.

      If you choose to disregard this advice, go to San Diego. LA also sucks independent of the heat. If you do, make sure to visit the Safari Park, which is a ways outside the city and the nicest zoo I’ve ever been to.

      • quanta413 says:

        I’m from the San Diego area, and the San Diego Zoo Safari Park is amazing. It’s not too far from the city. About half an hour to an hour drive each way depending on traffic and exactly where you are.

        I grew up going there probably once or twice a year, and every other zoo I’ve been to since then feels kind of lame in comparison. Including the regular San Diego Zoo, which is still a pretty good zoo. If you can spring for it, the caravan safari is great. You can feed the giraffes.

        On the other hand, Hoopyfreud is right about the heat. It’s far enough inland that the summer temperatures are ~10 degrees hotter than in San Diego itself. And in hot years, it’s really bad.

        LA is one of my least favorite cities, but it’s very much a matter of taste. My sister loves LA. Like most cities, there’s a lot of variation within it. The traffic is abominable though and public transit almost nonexistent which makes getting around it torture and a lot of it is ugly, but Disneyland is pretty neat. Santa Monica is also very nice. Santa Monica is on the ocean and will tend to be cooler whereas Disneyland is inland and will be hot much like the Safari Park.

        San Francisco was ok last time I was there, but I don’t remember any part of the city much. I was visiting a friend last time. It has some very nice restaurants, but so do the other cities. It also tends to be a bit expensive even compared to other parts of California although I haven’t been to San Fransisco in over 5 years so this may be inaccurate.

        Never been to Portland.

        • AG says:

          imho, SF’s main attraction is hiking day trips, but that’s not quite appropriate for a pregnant lady, either.

        • Rebecca Friedman says:

          Chinatown is pretty and not expensive. Very good food, too. There are various good museums as well, if you like museums, and a zoo. The japanese tea gardens are quite pretty – basically Golden Gate Park has a cluster of museums, many of which I think are interesting, though I mostly just go to the California Academy of Sciences (which sadly is on the expensive side for a one-time visit). There’s also a really cool ship museum – yes, yes, probably less cool than the Iowa, but really pretty nice; they have a lumber schooner and a tall ship and an old steam ferry that used to take people back and forth before they built the Bay Bridge, as well as various smaller ships, and you can walk onto the ships and look around (and that one is not expensive). That said, getting around the city is a pain (driving there can be miserable – lots of traffic plus lots of one-way streets plus very expensive parking – and walking doesn’t always feel safe; I haven’t tried public transportation), getting a hotel probably is expensive, and I don’t know how interested you (plural) are in museums/zoos, or what else there is to do there. It definitely has nice weather in the summer, though. And lots of very pretty buildings.

          • AG says:

            Cal Academy isn’t expensive if you do the adults-only Thursday night special!

            Public transportation is fine on weekends. Much less experience with it on weekdays. Stick to the trains/light rail if you can. I’ve been able to get to most anywhere on the peninsula on public transportation alone, but I’m also willing to walk miles, and am not pregnant.

      • Silverlock says:

        I can second San Diego, although I was only there once and it was thirty years ago.

        If you are keeping it to the leftward coast, Seattle is kinda nice, with mountains not far away.

    • Etoile says:

      It depends on how advanced the pregnancy is. In any case, roller coasters are probably off the table, as are winery tours. Also strenuous hikes or anything high impact or where wearing nice clothes is part of the appeal (although i can’t think of anything to fit the latter description…. I just know that at a certain point in pregnancy, maybe 32+ weeks, there is no maternity clothing that feels like it looks good.)

    • Elephant says:

      Portland is great for getting to excellent scenery (coast, mountains, …), but the city itself isn’t very exciting in terms of sights. It’s a much better place to live in than to visit as a tourist. (Perhaps the best of these 4 to live in.)
      If you’ve never been to SF, it has some remarkable scenery within the city — Coit tower, the Presidio, … — and some great museums (Exploratorium!).
      San Diego is beautiful, though it will be hot in July. There have been some good comments already about it. Balboa Park is especially nice.
      I haven’t spent enough time in LA to comment.
      Of these, I’d pick SF or SD.

    • bean says:

      I’ll advocate for LA. Yes, it’s crowded and if you try to go somewhere during rush hour, the traffic is horrible. But there’s a lot to do, including seeing the greatest ship ever to sail the seas (you know I had to mention it) and it’s not as bad as it’s made out to be. If it’s not rush hour, the traffic is actually pretty reasonable. Just do yourself a favor and fly into Orange County instead of LAX.

    • zoozoc says:

      I grew up near Portland. The weather can be very nice in July (though it is possible there is rain, just not very likely). There are some typical “weird” touristy stuff in Portland. The beach or Mt. Hood isn’t too far if either of those are your thing. There are also a lot of waterfalls that require very little walking to see in the Gorge (1 hour drive from Portland). Multnomah Falls is pretty, though there are usually a lot of people there. But there are multiple falls near each other in that area. The historic Columbia river highway gives access to multiple water falls, as well as some pretty viewpoints, though the road itself is very windy (which is why they built I-84).

    • AG says:

      Rather than picking a city, I recommend picking an event. Research what sort of things are happening in the time frame of your visit, and pick your city based on what event seems more engaging. Music festivals, conventions, concerts, etc.

      SF and Portland probably would be more preferable in late July for temperature reasons, though.

  16. AlesZiegler says:

    About a year ago, I wrote for my blog, in Czech, something like a review of Communist Manifesto by Marx and Engels. Since Scott declared he is trying to learn more about leftism, I figured that it might be good use of my time to translate it (with small revisions) and post it here. I am afraid it is not very good translation, but I am unwilling to invest more time into it. I also realize that it far from definitive take on Manifesto and it contains heavy simplification of very complex topics. Nevertheless it is quite long for a comment, so I divided it into two parts.

    Tl;dr Marx was wrong on many points, but he was not as wrong as it would seem to somebody who doesn´t know historical context of a year 1848, when Manifesto was written.

    Marx´s magnum opus is titled Capital. I admit that reading and trying to understand it is beyond my intellectual capabilities and patience. Marx belongs to German philosphical tradition, like Hegel, Kant and similar thinkers. Important part of this tradition seems to be that every philosopher invents his own esoteric terminology and then uses it to write hundreds and hundreds of pages. Small industries of profesional interpreters thus grew around their works.

    Communist Manifesto is different. It is plainly written and relatively short text (less than 8000 words in Czech, which is apparently about 12500 words in English, although I have not checked that) for wide audience. It contains straightforward description of a political situation in 1848 and step-by-step blueprint on what should be done about it.

    Btw. Friedrich Engels is listed as a co-author of Manifesto, which I am going to ignore for simplicity.

    What is Marx´s point?

    He developed “theory of human history”. That is a horrible buzzword in need of deconstruction. Core of his theory is formed by crude analogy between so called bourgeouis revolutions, which happened in the past, and socialist revolutions, which he predicts will happen in the future.

    Bourgeoise originally meant propertied inhabitants of medieval towns, group which was somewhere between landed aristocracy and peasants on a social ladder. Bourgeois revolutions are processes by which feudal society was transformed into capitalist society.

    Marx thought transformation from feudalism to capitalism was caused by changes in material foundations of society. It is imho obvious, at least in Manifesto, what Marx meant by changes in material foundations. He meant what we today call economic growth. This growth led to a rise of economic power of the bourgeoisie, who used it to dismantle political privileges of feudal aristocracy and feudal restrictions on entrepreneurial freedom; thus capitalism was born.

    Just like economic growth inevitably transformed feudalism into capitalism, inevitable continuation of growth will lead to worker´s revolution, which will lead to toppling of the rule of bourgeoisie and abolition of private property, by which Marx meant introduction of centrally planned economy. You may notice that Marx predicts “inevitability” of a chain of events, which is usually a mistake. If such predictions do not materialize, you will become a laughing stock.

    In addition to many interpretations of Marx, there are also many explanations of what he got wrong. Unfortunately, Manifesto is very ambiguous on the key question whether transition from capitalism to socialism could be gradual. From that we could differentiate two possible interpretations. First of them supposes that Marx was virtually completely wrong; ambiguous nature of Manifesto gives sufficient credence to such dismissal. Alternative interpretations more charitable to Marx are however also possible.

    If our question is “what Marx thought?”, and whether he was simply wrong, we have to answer two further questions: “did he think that there could be partial improvement of worker´s lot under capitalism, or was it according to him impossible?”, and “what he meant by capitalism?”.

    It is a common ground between various interpretations that Marx took as his assumption that in Britain, most technologically advanced country of his era, living conditions of workers had worsened for some time, i.e. for decades. There is some, albeit contested, evidence for this so called “Engels pause”. From this Marx derived long term unsustainability of capitalism.

    Stupid Marxism could be summed up as a theory that capitalist society is every society based on private property, that in such a society living conditions of workers are inevitably going to deteriorate, and that this will inevitably lead to destruction of capitalism via socialist revolution. That is obviously absurd, and on top of that is piled another absurdity, that “feudalism”, “capitalism”, and “socialism” are clearly delineated phases in a social development, it is virtually always possible to determine in which of those phases society is, and changes from one to another happen in relatively short periods of chaos usually accompanied by increased rates of violence (i.e. revolutions).

    If we endorse this interpretation of Marx, which was favored by Soviet regimes, then it is clear that he was simply wrong. Not only in that he was wrong about how capitalism will be replaced by socialism; his account of how feudalism was replaced by capitalism is also flawed.

    Feudal institutions did not all vanish in one fell swoop and got replaced by capitalist institutions. For example, monarchy is a feudal institution, yet Kingdom of Sweden is not a feudal country. In most cases, institutions which social scientists and similar creatures call “feudal” had been replaced by those they call “capitalist” gradually, via evolution. Marking a moment when capitalist institutions started to dominate over feudal ones is for many countries impossible.

    In Czechia, transformation from feudalism to capitalism was long process starting with reforms of the Emperor Joseph II., continuing with Civil Code of 1811, abolition of labor duties of peasants toward feudal lords during otherwise unsuccessful revolution of 1848, then constitution promulgated by the Emperor in 1861 under pressure after defeat in war, then constitution of 1867 passed by parliament in Vienna, and finally World War I and declaration of the republic (1918). And Czech example of gradual evolution from one economic system to another from 18th to 20th century is more representative than drama of the French Revolution. In fact, revolutionary transformation from feudalism to capitalism happened basically only in France, and perhaps in United States. Everywhere else it was evolution.

    If we would be charitable towards Marx, we would suppose that he knew about gradual nature of transformation from capitalism to feudalism in e.g. Britain. And thus, since his socialist revolutions should be analogous with bourgeois revolutions, he had a wider definition of “a revolution” than is usual. So revolution could be also gradual evolution from capitalism to socialism.

    This interpretation of Marx is also supported by his conviction that attainment of real democracy is necessary and also sufficient condition for transforming capitalism to socialism. There is in Manifesto imho more or less implicit premise that doctrine of a limited government is merely an instrument serving private interests of capitalists, which does not stand a chance against rising tide of social democracy. And this tide will necessarily lead to centrally planned economy, because that is the only option in technologically advanced society (Marx neither predicted nor desired any “return to nature”), when all previously private property is “democratized”, ie. nationalized.

    To sum up, according to charitable interpretation of Marx, it is possible that capitalism will be overpowered via evolutionary, gradual socialization of society, which shall be an inevitable consequence of an economic growth. Of course, there has been massive economic growth since 1848, moreover acompanied by substantial democratization of society. So if Marx was right, we should have been at least considerably advanced on a path to socialism. And socialism according to Marx means abolition of private property. Gradual evolution from capitalism to socialism thus should mean gradual strengthening of public property at an expense of private property. It would mean gradual nationalization of larger and larger share of national wealth and also weakening of legal protections of private property. It seems that this did not occur, despite some tentative steps in such direction.

    (to be continued below)

    • AlesZiegler says:

      (continuation)

      However, now we should finally ask ourselves what Marx really meant by private property? Marx thought that prevalence of this institution is the defining feature of capitalism. Oh really? Was there not private property under feudalism?

      According to Marx, feudal property was something completely different from bourgeois property.

      Ok. But, that implies Marx´s definition of private property and hence capitalism is so closely associated with society of his time that developed Western countries are no longer capitalist in a sense Marx used. It is imho totally defensible to assert that institution of private property existed under feudalism, in Victorian Era, and in contemporary liberal democracies. What is however quite indefensible is to assert that today we have private property just like they had in Victorian capitalism, but under feudalism they had something completely different. It is clear that definition of property had shifted over those periods and also that it retained certain basic features.

      Marx did not define private property in Manifesto, nor did he defined what he meant by “society based on private property”, but it is imho implicit from a very word “capitalism”, as well as from other cues, that he saw close relation between private property and freedom of entrepreneurship. Which is quite logical, since higher levels of respect for private property mean more freedom to do stuff with such property, including more business uses of it. In this respect, Victorian era capitalism was indeed more free and “capitalistic” than feudalism, but it was also an era of more entrepreneurial freedom than exists in contemporary liberal democracies.

      In feudal era, entrepreneurship had been stifled by the guilds, chaotic legal system, road banditism etc. In liberal democracies, we have heavy handed government regulations and huge taxes. In Victorian Era, there were neither of those.

      Here I should explain what I mean by “Victorian capitalism”. I use it loosely to cover a century between Napoleonic Wars and World War I (1815 – 1914). In that time,Great Britain, for most of that period reigned but not governed by Queen Victoria, had been the global hegemon, and most of the world tried to copy at least some British institutions. Victorian capitalism could be defined by combination of following traits: a) legal system heavily geared toward protection of private property; b) night watchman state mainly concerned with defense or attack against external enemies, administration of justice and providing for basic transport and communication infrastructure; c) economic growth associated with quick urbanization; and d) centralized and undemocratic political decision making, from which most of the population is excluded. If elected bodies have political power, voting rights are not universal and/or they share power with hereditary officials.

      Grain of truth in Marx imho consists in that he correctly predicted long term unsustainability of such system. And he was also right that if it will not be reformed into something more sustainable via evolutionary means, there will by violent revolution – that more or less happened. World of Victorian capitalism was destroyed by two world wars and other disasters of a first half of 20th century. That was not exact copy of the French revolution, but it definitely was quite violent transformation into…something.

      Marx was however very wrong about what will come after revolution. He predicted democratically directed centrally planned economy. In fact we got centrally planned economy in some countries and democratization of political decison making in others, but nowhere both. How is this possible?

      I have a primitive theory: once almost everyone got voting rights, it became clear that most people simply do not want to live in centrally planned economy, because it would be impossible without intolerable (for them) restrictions on personal freedom. However, instead of large scale nationalization of all means of production, other methods were developed to deal with problems which made Victorian capitalism unsustainable, and some of them would, in 1848, sound basically as revolutionary as what Marx proposed. I mean especially massive growth in public sector, financed by taxation, and heavy handed regulation of business, legitimized by public interest concerns.

      Nazism and Soviet style socialism were two other regimes rising alongside liberal democracy from ruins of Victorian capitalism. Both were undemocratic, but based on central planning. Soviet regimes deserve a special attention in this context, because they themselves invoked Marx´s theory as their justification.

      Imho key feature, without which it is impossible to understand Soviet system, is that Lenin changed Marx´s plan in one crucial aspect. He “recognized” that Russian peasants are too stupid for democracy to work, and instead they have to be led by “revolutionary avantgarde”, i,e. by Communist party, which will govern in first phase of revolution instead of democracy. And first phase of revolution will be of indeterminate length. In accordance with this, after his coup Lenin disbanded democratically elected Russian constitutional assembly. And since then, it had been basic governing philosophy of all successive Soviet governments until Gorbachev.

      Lenin´s plan was more realistic compared to original Marx´s ideas, because it did not depend on supposed behavior of voters, which is, as it turned out, not how voters actually behave. Under Gorbachev, Soviet system was democratized and then it promptly collapsed because of it. But Lenin´s alteration to Marx´s orginal ideas also opened direct route to Gulag, government directed famines and other atractions of real world Soviet style regimes. When you presume that elite knows better what is good for masses than people themselves, and when such elite is not bound by something like Christianity, which prohibits certain actions as immoral, every crime is defensible if it could be justified as a necessary step on a path to utopia.

      • Procrastinating Prepper says:

        Thanks for the write-up, Ales.

        Imho the part of Manifesto that causes most contention for modern socialists isn’t the “revolution” part, it’s the “inevitable” part. It’s pretty clear from a 2019 standpoint that workers’ conditions and rights can and did improve under capitalism, but this doesn’t necessarily indicate a trend that will naturally lead to “real” socialism. Instead we’ve just shifted to a new and mostly stable equilibrium between labour and capital – From Marx’s perspective, we’ve stalled partway to the finish line.

        So while “revolutionary avant-garde” might be the more famous failure mode for socialists, I think the more common one is in the other direction: when activists sit back and take on the role of economic critic, pointing out all the instances where workers are getting worse and worse off (even when they’re not) and waiting for people to “inevitably” join their cause.

      • Pretty decent analysis.

        1848 Marx thought that the proletariat would “win the battle of democracy” and then “use its political supremacy to wrest, by degree, all capital from the bourgeoisie.” To me this sounds like:
        Step 1: Win elections by a landslide.
        Step 2: Amend constitutions through a deliberate legislative process (where necessary).
        Step 3: Gradually implement 100% wealth tax.
        Step 4: The lower-stage of communism (i.e. the economy run as one giant corporation with every citizen an equal shareholder entitled to elect boards of directors).
        Step 5: The higher-stage of communism (superabundance, no authority to regulate individuals’ use of resources, no state, fully-automated gay space luxury communism Star Trek shit).

        The later experience of the Paris Commune made Marx more jaded about this idea of the electoral road to the “lower stage of communism” or “socialism” as it has come to be called. As Lenin notes in State and Revolution, Marx afte 1871 thought that the whole “ready-made state machinery” had to be “smashed” and replaced with a more radical direct democracy that eschewed any sort of professional bureaucracy or military, but which would instead arm the vast majority of the population to act as militias and make all functionaries instantly recallable and compensated as regular workmen.

        As for capitalism becoming inevitably more dysfunctional over time, Marx based this on two theories:
        1. the idea that capitalism tends to concentrate wealth, and
        2. the tendency of the rate of profit to fall, which Marx called “the most important law of political economy.”

        Marx did not preclude that workers would obtain better subsistence living conditions as capitalism progressed. Marx noted that workers’ subsistence levels, as societies understood them, were never purely determined by biological needs; they also had a “moral and historical” factor in their determination, and these levels might rise with the progress of civilization and the rising expectations of what constitutes a bare minimum of decent living.

        The problem for workers is that this would still be subsistence living. Workers would not be accumulating “property,” by which Marx did not mean the piddly things like one’s personal toothbrush, but instead capital. Or at least, workers would not accumulate enough capital to make any difference. The proportion of people who could live as capitalists—live off profits, interest, and dividends—would dwindle. Protecting the interests of those who lived off these sources of revenue would seem more and more illogical to the vast majority of people who would have no share in that sort of business and nothing to gain by it.

        But then, there might be difficulties in trying to force this process to its conclusion where the preconditions had not run their course…

        “For it [capitalism] to become an “intolerable” power, i.e. a power against which men make a revolution, it must necessarily have rendered the great mass of humanity “propertyless,” and produced, at the same time, the contradiction of an existing world of wealth and culture, both of which conditions presuppose a great increase in productive power, a high degree of its development. And, on the other hand, this development of productive forces…is an absolutely necessary practical premise because, without it, want [deprivation] is merely made general [equalized], and with destitution the struggle for necessities and all the old filthy business [exploitation and the State] would necessarily be reproduced…”
        —Karl Marx, The German Ideology (1846), Chapter 1, section A.

        “The basis of bureaucratic rule is the poverty of society in objects of consumption, with the resulting struggle of each against all. When there are enough goods in a store, the public can come whenever they want to. When there are few goods, the public are compelled to stand in line. When the lines are very long, it is necessary to appoint a policeman to keep order. Such is the starting point of the power of the Soviet bureaucracy. It “knows” who is to get something and who has to wait…it [the Soviet bureaucracy] of course draws off the cream for its own use. Nobody who has wealth to distribute ever omits helping himself first.”
        —Leon Trotsky, The Revolution Betrayed (1936), Chapter 5. Available here: https://www.marxists.org/archive/trotsky/1936/revbet/ch05.htm

        • sandoratthezoo says:

          So I have a sincere question about all this:

          One of several banal objections to communism is that, like, the market is a needed means of coordination of scarce resources, and it’s mediated by profit-seeking. Like, the whole “nobody even knows how to make a pencil” thing. So, like, ignore AI for a minute, assume that we’re back in the 19th Century. Even if we assume generously that traditional industrial processes will eventually be able to produce a superabundance of goods for everyone, they will still need to be minded, and worked, and probably there won’t be a superabundance of the factories that make the most essential part of the factories.

          What did Marx think about that? Did he recognize it as a problem? How was it supposed to work in a communism that doesn’t presuppose at-least-human-level AGI?

          • Protagoras says:

            He assumed this problem was solvable, as did many thinkers of his time. Though my recollection is that Marx himself was vague about it, others with similar views tended to point to wartime economies, where having a central authority take over everything succeeded in many cases in producing tremendous amounts of war material. In his time, there was not yet the impressive track record of failure for efforts to apply that approach to peacetime conditions.

          • AlesZiegler says:

            Marx imho thought that coordination problems will be initially handled by central planning, which will be so succesful that after a while, it will lead us to post-scarcity society, where by definition scarce resources won´t be an issue.

            Things clearly didn´t turned out according to plan.

      • cassander says:

        If we would be charitable towards Marx, we would suppose that he knew about gradual nature of transformation from capitalism to feudalism in e.g. Britain. And thus, since his socialist revolutions should be analogous with bourgeois revolutions, he had a wider definition of “a revolution” than is usual. So revolution could be also gradual evolution from capitalism to socialism.

        I think this is being excessively charitable, given how big a fan marx was of the french revolution. I don’t think he would rule out the impossibility of such a gradual transition, but he clearly saw revolution as the default mode.

        Grain of truth in Marx imho consists in that he correctly predicted long term unsustainability of such system. And he was also right that if it will not be reformed into something more sustainable via evolutionary means, there will by violent revolution – that more or less happened. World of Victorian capitalism was destroyed by two world wars and other disasters of a first half of 20th century.

        Again, I think this is giving him too much credit. there might have been a crisis in the capitalist world from 1914-1945, but it really had nothing to do with capitalism or marx’s predictions for why war would break out. Giving him credit for calling the end of victorian capitalism is like giving someone credit for predicting the 2008 financial crisis because he thought the mayan calendar was ending, or something.

        Marx was however very wrong about what will come after revolution. He predicted democratically directed centrally planned economy. In fact we got centrally planned economy in some countries and democratization of political decison making in others, but nowhere both. How is this possible?

        Because you can’t have both. economic central planning is corrosive to democracy.

        Imho key feature, without which it is impossible to understand Soviet system, is that Lenin changed Marx´s plan in one crucial aspect. He “recognized” that Russian peasants are too stupid for democracy to work, and instead they have to be led by “revolutionary avantgarde”, i,e. by Communist party, which will govern in first phase of revolution instead of democracy.

        I disagree that this was a change lenin made from marx. Lenin coined those terms, but marx’s dictatorship of the proletariat was always going to have be be run by someone, and that never could have been universal mass democracy because that’s just not workable. Even the paris commune had leaders, and those leaders might as well have been called the vanguard or the politburo. Lenin wasn’t changing marx, he was just more explicit about how things hand to work and more enthusiastic about the extreme violence he would deal out.

        • Clutzy says:

          The means of revolution is always the issue. The greedy capitalists must be replaced by humans, who are also greedy, because they are humans.

        • AlesZiegler says:

          I am of the opinion that global crisis of the first half of 20th century actually had lot to do with unraveling of capitalism in a way reasonably close to Marx´s predictions. In a year 1900, laisez faire capitalism seemed unsustaniable. Voting rights were expanding in many countries and socialist or similar parties were gaining ground in elections around the Western world. See also Scott on History od the Fabian society.

          As for how this crisis of laisez faire capitalism connects to following explosion of violence, one plausible channel is that WWI was so long and bloody partly because German and Austro-Hungarian governments refused to make acceptable peace proposals even after their original war plans totally failed in 1914, and perhaps that had something to do with their fear that anything less than victory will be exploited by internal socialist revolutionaries.

          Ad Paris Commune, yes, it had leaders, but they were elected by people, unlike leaders of Soviet Union. It has now been demonstrated that democracy and central planning do not go well together, because voters really really hate having their lives directed by central planners, at least if there is no major war they want to win. But in 1848 Marx did not know that, because there was little to no practical experience with central planning, or even with real democracy.

          • cassander says:

            As for how this crisis of laisez faire capitalism connects to following explosion of violence, one plausible channel is that WWI was so long and bloody partly because German and Austro-Hungarian governments refused to make acceptable peace proposals even after their original war plans totally failed in 1914, and perhaps that had something to do with their fear that anything less than victory will be exploited by internal socialist revolutionaries.

            This is a stretch. everyone failed to make peace in the war because everyone feared the domestic political consequences. But while the war created a revolutionary situation, it came about because of the war, not capitalism, and capitalism or fear of domestic unrest had nothing to do with the start of the war, except in Austria But there, the concern was about quashing ethnic rivalries and Conrad von Hötzendorf’s feeling that a war would help unify his country.

            Ad Paris Commune, yes, it had leaders, but they were elected by people, unlike leaders of Soviet Union.

            They were elected, by the soviets. Not the most honest elections in the world, I grant you, but they were elected and in the only sort of election that marx would have cared about.

            But in 1848 Marx did not know that, because there was little to no practical experience with central planning, or even with real democracy.

            Maybe don’t advocate extreme violence on behalf of a political/economic system that has literally never been tried….

  17. Hoopyfreud says:

    @Bean and anyone else with the relevant knowledgebase:

    Does artillery (meaning tubes and shells) have a place on the battlefield in modern armies of the 21st century? If so, what advantage does it have over air support/UAVs/rockets? Does the answer change if the enemy has air supremacy, and if so in what way?

    • bean says:

      Yes. Guns are still the cheapest way to deliver large quantities of explosives at medium range, and in certain cases, that’s what you need. Precision is all well and good, but for small shells and precision to substitute for lots of shells, you need good comms. Which isn’t guaranteed. Rockets are much more expensive per-shell, although the launchers are cheaper. Aircraft are more expensive to buy and run, although the ammo itself is dirt cheap. And if you need 24/7 support on short notice, artillery is a lot cheaper than aircraft.

      Enemy air superiority makes artillery operations more hazardous, but also removes a major competitor. Overall, slight win for the guns.

      • Hoopyfreud says:

        Is the cost enough of an issue that the US has artillery ammo stockpiled in sufficiently large quantities for significantly greater deployment than of the alternatives? Also, how valuable are fixed and ship-mounted guns? I’d assume they’ve gotten less valuable, but I really don’t know.

        • sfoil says:

          Bean can talk about naval guns but:

          It’s not just how much money something costs. There’s also storage and transport. One ATACMS missile probably takes up about the same space as 25 artillery shells. If I’m trying to guarantee that I hit something and kill it 100km away, then I want the missile. If I want to suppress a suspected enemy position, five artillery rounds is better.

          If I’m I’m getting shot at and I want the enemy position destroyed ASAP, it would be nice to hit it with the missile (if I’m not too close) but I’d rather have a few guaranteed artillery rounds in the next five minutes than maybe the missile in the next half hour after the crane is done reloading the launcher.

        • bean says:

          Fixed guns are basically gone. They’re way too vulnerable to airstrikes and missiles. Ship-mounted guns are in a weird place. The standard 3″ and 5″ weapons are pretty decent. They’re cheap, reliable, and good for scaring merchant ships and a bit of light shore bombardment. Attempts at better guns always seem to founder on horrible programmatic problems, and modern amphibious doctrine doesn’t really have a place for lots of shore bombardment. (I expect these efforts to more or less go away as the Navy reorients towards blue-water missions.)

          • Hoopyfreud says:

            Wait, so what’s the current doctrine for landings? Is it just down to air support? Putting boots on a hostile beach seems like an important problem to be able to solve.

          • Le Maistre Chat says:

            @Hoppyfreud: a literal-minded Robert McNamara decided our landing craft would load boots into T-shirt cannons, noting that no competitor has the point-defense technology to shoot down the incoming boots.

          • bean says:

            Current doctrine for landings is to not try them against heavy resistance. Hit them where they ain’t, which things like helicopters and LCACs have made much easier. I’m not saying you don’t want any bombardment for that, but land attack is seen as a lot less important today than it was 10-20 years ago, when navies were struggling to justify their existence in the absence of a peer competitor, and found it in littoral operations. Now, with China on the rise, they’re back in blue water.

          • Hoopyfreud says:

            How would that work in a hypothetical second Pacific campaign, for example? Has anyone given thought to island-hopping in the 21st century?

          • bean says:

            Not really, for two main reasons:
            1. Nations work on the conditions they have, not the conditions that would result if they refought conflicts from three quarters of a century ago. The US has a lot of forces in the western Pacific already, and isn’t really looking at forcing a heavily-defended beachhead.
            2. Island-hopping isn’t particularly likely these days. The airplanes you can fly off a primitive strip on an island with minimal support are not able to challenge the planes flying off of a carrier. Ships have more range, and we’re much better at keeping them at sea. The island-hopping campaign might have been very different if ammo unrep was developed before the war. If you need a base, do what they did at Ulithi and grab an empty one.

        • Ketil says:

          Last coastal forts built with fixed guns in Norway around 1993, and then replaced with mobile missile units (based on Hellfire, IIRC). Lower capacity (the guns could fire one round every two seconds), but way better flexibility and survivability.

          I think the gun/missile trade off varies with caliber, small caliber are less likely to be replaced. Some years ago, Norway replaced its aging 155mm artillery with MLRS rocket batteries, but recently were looking into replacing it with new guns. And last time I looked, 120 mm smooth bore guns with fin-stabilized penetrator rounds were still the best way to punch a hole in enemy MBTs.

          Possibly outdated information, caveat lector.

          • bean says:

            Missiles definitely have replaced larger artillery. The short version is that in a lot of cases, big artillery existed because there were occasional times when you needed big projectiles and there was no other way to launch them. The heavy siege pieces of WWII are the most obvious examples of this. Many fired only a double-digit number of times despite their massive size and cost. A missile is strictly better here because it’s probably cheaper to design and build, and a lot more portable. A 155, on the other hand, could fire thousands of rounds throughout its life. Each round is pretty cheap, and it’s got more sustained firepower than a rocket battery.

    • sfoil says:

      Yes. The shortest illustration I can give is that there was a line of thought in the late 90s/early 2000s that air power would render artillery obsolete, then Afghanistan showed that this wasn’t the case even for the USA; other conflicts fought since then (e.g. Syria, Ukraine) have involved heavy use of artillery.

      To understand this, realize that in order to achieve given effects on a target there is a continuum of different weapons that make different tradeoffs between range, response, availability and, often, power. Response is how quickly I can move from detection to strike, availability is how likely it is I can use something when I’ve got a target. Availability implicitly includes cost.

      Basically, artillery is more responsive and available than anything that’s more powerful (like airstrikes or tactical missiles) and has longer range and more power than things that are more responsive (basically, man-portable weapons).

      An artillery battery can fire on any target within its range on an around-the-clock basis within a few minutes. A strike aircraft might be able to engage faster if it’s in the air around a target location (on station), but there are fewer strike aircraft than artillery batteries. If I’m within range of friendly artillery, it will basically never be faster to direct aircraft rather than artillery onto a target.

      Tubes have a couple of benefits over rockets as well, the main one being ammo load: even if you have infinite money, artillery rockets are physically larger than shells so for any given volume of transport or storage you have fewer of them. Also, ceteris paribus tube artillery is more accurate than rockets. Rockets also have a minimum effective range, guns don’t — this is an important consideration if you’re using a “firebase” concept, common in low-intensity warfare, basically tiling an area with lightly defended artillery batteries.

      If you can’t fly, this makes artillery even more important because some of the targets that, in an ideal world, you would like to bomb must be attacked with other means instead.

      • Hoopyfreud says:

        Afghanistan showed that this wasn’t the case even for the USA; other conflicts fought since then (e.g. Syria, Ukraine) have involved heavy use of artillery.

        Aha! Good to know. I’d been struggling to understand the importance (if any) of artillery units in those conflicts, not finding much to go off of. I guess either artillery isn’t sexy enough to do a deep dive on or I was looking in the wrong place. Probably the latter.

        The responsiveness of artillery being a major strength makes sense. I also hadn’t realized that shells were more accurate than rockets, but it does make sense after some thought.

        It does, however, sound like artillery is more useful defensively than offensively. Is that true, or am I missing something else? Do you know if the offensive lines in the conflicts you mentioned made heavy use of artillery as well?

        • Tenacious D says:

          There’s a movie called Hyena Road set in the Afghan theatre. One scene includes an artillery barrage that really illustrates its responsiveness: a small squad of infantry is under attack by a much larger group of Taliban but they are within the firing range of a forward operating base…

        • sfoil says:

          Artillery has been important in Syria because it’s something that the regime has that its enemies don’t. I think Syria may have depleted its artillery stock somewhat because early on they appeared to use it much more liberally in an attempt to quickly overwhelm urban defenses, which didn’t work out.

          It does, however, sound like artillery is more useful defensively than offensively. Is that true, or am I missing something else? Do you know if the offensive lines in the conflicts you mentioned made heavy use of artillery as well?

          Yeah they did.

          As to whether there’s something particular to tube artillery beyond what’s inherent to attack or defense, I think the answer is that artillery is somewhat more “defensive” because offensives imply a concentration of power in time, and you can get more power from strike aircraft in a specified time frame over a given area compared to artillery, which relies more on being persistent compared to aircraft.

          It’s also easier for attacking forces to outrun the coverage of guns than something with longer range, but I think that has more to do with the nature of attacking than artillery per se since theoretically you should always be able to maneuver attacking forces such that they remain in continuous range of friendly artillery given modern equipment.

    • johan_larson says:

      Leading-edge modern artillery systems are pretty darn high tech, and very accurate. Check out this one, which uses GPS guidance:

      https://en.wikipedia.org/wiki/M982_Excalibur

      Making a GPS receiver tough enough to survive an artillery launch sounds hard.

    • Nornagest says:

      In a lot of ways even talking about UAVs in the context of a battlefield misses the point of them. The current generation of UAVs is optimized for endurance: they’re lightly armed by air-to-ground standards (the Reaper’s much more capable than the Predator was, but a Strike Eagle or Super Hornet in CAS configuration can carry almost an order of magnitude more ordnance than either one), slow, relatively unimpressive range-wise and incapable of aerial refueling, and not very survivable against even a modest anti-air capability, but they can stay in the air for a good fourteen hours. That’s great for the kind of low-intensity hunt-and-peck warfare we’ve been waging with them, where you spend a lot of time looking for actionable targets but they tend to be small and soft when you find them, and they can’t really shoot back. It’s not as good for higher-intensity counterinsurgency, like Vietnam or the earlier stages of Iraq, and it’s almost useless against a peer adversary. That is in many ways counter to USM doctrine, which is historically and still largely tuned for high-intensity warfare against peers, but we get away with it because it’s relatively cheap.

      Conventional air support and rocket artillery funges more directly against tube artillery, but bean and sfoil have already covered that pretty well.

    • John Schilling says:

      Artillery is usually not a substitute for air supremacy because both artillery in action and the logistics needed to support it are conspicuous and vulnerable to air attack. Partial exception in the case of a prepared defense where both the artillery and the ammunition stockpiles can be in hardened, camouflaged locations, and even then it probably only buys you a few days or weeks at most. If that is your game, probably best to go with multiple rocket launchers to maximize volume of fire before being destroyed or forced to go back into hiding.

      If the enemy doesn’t have air supremacy, having good artillery and the will to use it can be decisive. That has been one of the major lessons of the Ukrainian conflict, and I’m not sure it has been properly learned. Look at it from the other perspective – if the enemy has good artillery and the will to use it, and if you are within fifty kilometers of the front lines doing anything but hiding, you can die at any moment and there’s nothing you can do about it. Or you can just be prevented from sleeping for a week, or any number of consequences in between. And the more of an annoyance you are to the enemy, the more likely the really severe consequences will fall upon you. You won’t even get a chance to kill the SOB who is going to kill you. Wouldn’t you really rather just hide and not do so much stuff that annoys the enemy?

      Air power can’t do this because air power can’t maintain a 24/7 combat presence over the battlefield and it can’t carry enough munitions to attack any but the highest-priority targets.

      However, note that this applies to good artillery coupled with the will to use it. The essence of artillery is indirect fire, where the gunners cannot see what they are shooting at. Very often, nobody on your side will be able to see what the gunners are shooting at; you’ll be unloading a few tons of high explosives on e.g. the triangulated current location of three tinder/grindr profiles that data mining suggests are enemy combatants too stupid to have turned in their phones before deployment. Which means a high rate of civilian casualties. If instead you are only ever going to approve a fire mission when a forward observer verifies a clean engagement with precision-guided munitions, then you might as well just use aircraft and drones. Though even then you’ll want forward operating bases to have a few mortars for immediate self-defense while waiting for the flyboys.

      • Ketil says:

        Air power can’t do this because air power can’t maintain a 24/7 combat presence over the battlefield and it can’t carry enough munitions to attack any but the highest-priority targets.

        Nitpick: I was pretty sure this was what US forces did to Iraq, basically carpet bombing the hell out of the Iraqi defenses, and then steamrolling the crumbling remains. Perhaps there was also artillery in action, and it just didn’t get as much screen time?

        One reason to use air over artillery is rapid deployment overseas, but I think the US is almost alone in putting so much weight on that kind of scenario – most other superpowers focus on homeland defense and land wars on their own continent. Or so I would think.

        • Clutzy says:

          If the US had engaged in any form of real carpet bombing there would be no Bush admin DOD-related person not currently in jail; international or national.

          A major reason why the US has lost wars beginning with Vietnam is we established a norm against normal warfare (aka effective warfare), which includes significant civilian casualties.

          • albatross11 says:

            You mean they’d be in jail next to all those Bush administration people involved in the torture program and the illegal mass surveillance of Americans?

          • cassander says:

            @albatross11

            There is a world of difference between torturing some people in black cells and burning down entire cities.

          • Clutzy says:

            The torture discussion was just a bunch of grandstanding. There is a minuscule % of the population that ever thought in their heart of hearts that waterboarding is real torture, a even smaller % of those people ever achieved any power.

            Plus there aren’t videos for the public to see. That’s probably reasons #1-10. Anytime there is video there is real outrage and people get convicted. Abu Ghraib was mere photos and people went to jail. The big 3 news networks sending videos of entire cities on fire and burning children running around screaming would get huge views.

            As a result US warmaking capacity has been significantly crippled.

          • Dan L says:

            @ Clutzy:

            There is a minuscule % of the population that ever thought in their heart of hearts that waterboarding is real torture

            I’m going to give you a chance to cite something that makes this not look ridiculous before I offer rebuttal. I’ll be extremely disappointed if “heart of hearts” was a weak attempt to make it unfalsifiable.

          • Clutzy says:

            I actually agree its not falsifiable. Because its a question that people would obviously lie about. Still, waterboarding is perceived by the layman (even though this is technically not true in some cases) to not produce permanent physical damage. Torture = Blood for most people, usually they also expect permanent damage. When a dude is put in a carpet by the mob and beaten, that isn’t thought of as torture its something different. Its an esoteric and fluttering concept.

            Also a concept people rate at like #100 in their policy preference portfolio.

          • albatross11 says:

            Not only Americans! The Khmer Rouge, Imperial Japanese Army, and Spanish Inquisition also didn’t believe in their heart of hearts that waterboarding was torture. That’s why they used it to make their prisoners feel extra-comfortable and welcome.

          • greenwoodjw says:

            usually they also expect permanent damage

            The clearest definition I could ever come to was “Typically inflicts long-term physical damage and/or terminates in death”.

          • bean says:

            Uhhhh……

            The reason that we didn’t carpet-bomb Iraq is because carpet-bombing is militarily obsolete. Air power isn’t about tonnage, it’s about destroying targets, and you can destroy a lot more targets per dollar with an F-16 loaded with JDAMs than you can with a B-52 doing Arc Light. We went from 90% dumb bombs in Desert Storm to 90% smart bombs in OIF for a reason.

            (Yes, I’m beginning to put numbers on this for Naval Gazing. But it might be a while before I get the post up.)

          • Clutzy says:

            The reason that we didn’t carpet-bomb Iraq is because carpet-bombing is militarily obsolete. Air power isn’t about tonnage, it’s about destroying targets, and you can destroy a lot more targets per dollar with an F-16 loaded with JDAMs than you can with a B-52 doing Arc Light. We went from 90% dumb bombs in Desert Storm to 90% smart bombs in OIF for a reason.

            You might think that, but it isn’t borne out in longstanding military victories. Yes it is important to take out strategic targets, but the evidence we have is that taking out only strategic targets is much, much, much less effective than taking out that target + everything within 5 blocks of that target. Also, your example is, on its premises, hilarious, because Desert Storm was a much more successful operation than OIF.

          • Dan L says:

            @ Clutzy:

            Because its a question that people would obviously lie about.

            This is not a new problem in polling, and claiming it does not exempt you from the observation that you’ve failed to support your claim.

            If you want to defend a medieval view, you’ll need a few epicycles to explain why in 2007 a strong supermajority answered that waterboarding really was torture, when a sizable majority were perfectly willing to follow up by with the idea that torture in general was justifiable.

            Torture = Blood for most people, usually they also expect permanent damage.

            Bullshit. Plenty of data and creative use of a taser say you’re full of it.

          • Eugene Dawn says:

            Torture = Blood for most people

            I am intensely skeptical of this: when I think of torture I think of a Spanish Inquisition-type dungeon, with rack or wheel. The Spanish Inquisition was literally forbidden to draw blood, but I think most people still imagine the Inquisition to have practiced torture. In fact, Wikipedia claims that waterboarding was the most common Inquisitorial torture.

          • albatross11 says:

            There is a massive amount of who:whom in defining torture. If the stuff we did to captives were done to Americans by, say, Iran, approximately 100% of mainstream American media sources and politicians would agree that this was torture and was an unacceptable action and a war crime. But when it’s us doing those things to other people, well, you know, it’s just slapping around some terrorists and splashing a little water in their faces, and anyway, those bastards have it coming.

            This is, of course, completely different from how German citizens rationalized what was happening with all their Jewish neighbors going on those one-way train rides and never being heard from again.

          • Clutzy says:

            There is a massive amount of who:whom in defining torture. If the stuff we did to captives were done to Americans by, say, Iran, approximately 100% of mainstream American media sources and politicians would agree that this was torture and was an unacceptable action and a war crime. But when it’s us doing those things to other people, well, you know, it’s just slapping around some terrorists and splashing a little water in their faces, and anyway, those bastards have it coming.

            I don’t even think that matters all that much. If we had video of waterboarding people would probably protest and get it banned, and probably try to put people in jail. This would even be true if it weren’t all that painful and caused no physical damage, and consistently produced excellent intelligence information.

            Just like sacking cities is now banned, despite its effectiveness as a military tactic. Because it can be shown.

          • bean says:

            You might think that, but it isn’t borne out in longstanding military victories. Yes it is important to take out strategic targets, but the evidence we have is that taking out only strategic targets is much, much, much less effective than taking out that target + everything within 5 blocks of that target.

            I’d like you to go into more detail on this evidence. Because British bombing policy during most of WWII was essentially focused on burning down the “everything within 5 blocks” and didn’t bother with targets. And it didn’t work. “Precision” bombing in that era wasn’t great, but it at least did something.

            Also, your example is, on its premises, hilarious, because Desert Storm was a much more successful operation than OIF.

            First, kindly don’t try to disguise the issue by combining the initial invasion with the absolute mess of nation-building and pacification that followed it. I’m way too smart to fall for that. The initial invasion was a military masterpiece. Baghdad fell 3 weeks after the first troops crossed the border. A whole country, with a huge army, outnumbering the invaders something like 5 to 1, fell apart in three weeks. And yes, I believe air power had a lot to do with this. (Seriously, let’s just agree to cut things off on May 1st, 2003, because everything that happened after that was a very different matter. And yes, if we’d been willing to carpet-bomb cities where our troops were attacked, it might have been easier to pacify the country. But we have morals, and so don’t do that.)
            Second, how many city blocks were flattened in ODS? Because I’m pretty sure that the answer is 0. All of the carpet-bombing was directed at troops in the field, with no civilians around to be killed. Today, we’d just blow up their headquarters and fuel depot, and let them sit until we got around to taking their surrender.

          • Clutzy says:

            Thats a perfectly valid strategy if you have the Desert Storm mission: Stop an invasion of a friendly. If you want to reform a country you are at war with, you have to convince the populace that they should pick a regime that is friendly to you, because your wrath is scarier than the wrath of the local tyrants.

          • bean says:

            If you want to reform a country you are at war with, you have to convince the populace that they should pick a regime that is friendly to you, because your wrath is scarier than the wrath of the local tyrants.

            1. There was a lot of goodwill towards the US in the days immediately after the invasion, which we then proceeded to squander with a terrible policy for the occupation. There was no need to convince most of the population that we were scarier than Saddam, because they hated him and his forces were destroyed.
            2. You still haven’t responded to my point that the British dehousing campaign was as close to your ideal use of airpower as we’re likely to see. They explicitly set out to injure morale among the civilian population. It didn’t work.
            3. If only there was a comparable pacification campaign in the same era, fought against Islamic insurgents by a power without all of those moral scruples the US has. Someone like, say, Russia.
            The Second Chechen War is a great control for Iraq. Fought by the Russians with minimal regard for world opinion or civilian casualties (seriously, look at how they retook Grozny), it still lasted for almost 10 years. Estimates for civilian casualties are all over the place, but the lowest is 25,000, and twice that doesn’t look crazy. If we ignore the surveys that seem to attribute every death in the country after March 2003 to the invasion, the estimates seem to cluster in the 150,000-200,000 range. But the current population of Chechnya is around 1.44 million, while Iraq has 37.2 million people. Even the low estimate of civilian casualties in Chechnya and scaling to Iraq gives 645,000 dead, which is in line with the highest values for Iraqi civilian deaths anyone has ever published.
            (And if you’re going to object that Chechnya and Iraq are not the same, consider Afghanistan. We’ve done far better there than they ever did. Yes, we don’t have a superpower slipping people SAMs to use against us, but we also haven’t been nearly as brutal, and we’ve lasted twice as long as they did. Seriously, deaths during the Soviet-Afghan war were like an order of magnitude higher than they are during the current one.)

        • bean says:

          I was pretty sure this was what US forces did to Iraq, basically carpet bombing the hell out of the Iraqi defenses, and then steamrolling the crumbling remains. Perhaps there was also artillery in action, and it just didn’t get as much screen time?

          Never underestimate how photogenic air power is. You can see gun camera footage on the nightly news because airplanes go back to their nice, safe bases every day. Artillery doesn’t, and they don’t have gun cameras, either. So nobody loves them. Except Fort Sill.

          I did some poking around, and the 3rd Division artillery fired 13,923 round of 155mm and 794 MLRS rockets during the initial invasion of Iraq. They were heavily engaged, but that’s still a lot.

    • Hoopyfreud says:

      Thanks everyone for the responses! I learned a lot from them.

    • Walter says:

      I don’t have the relevant knowledgebase:

      It feels to me like the drone is the weapon of the future, and if two top tier armies fought one another that would be the ‘breakout weapon’ of the war. I think artillery would be mostly impractical in such a conflict.

      • John Schilling says:

        Drones, at least as presently understood, are weapons that can be used by top-tier armies but not against top-tier armies – at least not on a large scale. They are not magically immune to anti-aircraft weapons, in fact the usual deployment model (persistent surveillance) makes them more vulnerable to detection and destruction. Plus their communications can be disrupted by anyone with electromagnetic supremacy. There are cheap wins to be had in deploying a few drones against a top-tier army that expected you to bring no drones, because they won’t have brought the appropriate countermeasures, but nothing transformative.

        Drones are for making it so top-tier armies will always have the upper ground over classic insurgencies, though again only if you are willing to use them. Though in this case, the rate of civilian casualties is low enough that if you aren’t willing to use drones effectively it is difficult to see how you can expect to win at all.

    • cassander says:

      the important calculation to remember. A fighter jet costs 80-100 million dollars. An attack helicopter costs 20-30 million. An M777 costs a couple million. It’s also much cheaper to train artillerymen than pilots.

    • Incurian says:

      bean and sfoil pretty much covered it, but as a former artillery officer I feel the need to chime in and lend my stamp of approval. I especially want to emphasize, that regardless of the difference in technical capabilities, artillery is typically more organic (in the military organization sense of the word), i.e. the guns are pretty close to you on the org chart so there are fewer layers of coordination to go through to employ them.

      To get concrete, medium caliber tube artillery is organic to the infantry brigade (brigade org chart). The worst case scenario (actually it gets a lot worse, but it’s a useful simplification) is that a fire mission has to be approved at brigade, but more likely support relationships have been pre-established between gun batteries and line battalions (and possibly further sub-divided). Brigades typically train and deploy together, so those formal support relationships entail more than fire support coordination measures – they come with personal relationships and history too.

      Rotary-wing attack aircraft are badass, but they are organic to the division (division org chart) so there’s an additional layer of coordination (not to mention the expense, vulnerability, logistics, loiter time, etc.). Support relationships, when established, are at a higher echelon. Divisions don’t necessarily deploy together, so the task force you deploy with may include a combat aviation brigade from a different division – you just met these guys in theater so you don’t know them or their SOPs.

      Fixed-wing close air support is great and all, but you have to coordinate through a whole other service (at least if you’re in the army – it’s a different story for the Marines), and the org chart looks something like this. It’s less complicated than it looks, but it is complicated. And not only have you not worked with these guys, but you may never have had the chance to train with any fixed wing at all since they live on different bases, and you don’t really get the opportunity to establish a relationship in theater because they still live on different bases and every time you call in for support you get some random guy (though the situation is less bad wrt JTACs).

      Compare all these to mortars, which are the cute-but-fiesty baby brothers of regular tube artillery, and live at the infantry battalion and company levels. Now we’re talking Dunbar-number fire support. They’re so organic, they’re operated by infantrymen, not artillerymen. Your infantrymen. So why would choose a puny 60mm mortar when you could call in one of these bad boys? Because you don’t have to call in the former.

      ETA: “Why would anyone drive a car when they could just ride the bus?”

      • Hoopyfreud says:

        To be fair to me, I think it’s more like, “why would anyone drive a car when they could just ride the train?”

        But point taken.

  18. j1000000 says:

    Random question, prompted by reading the flat earthers thread on here and thinking about conspiracy theories.

    I am definitely not a conspiracy theorist, so one thing that shook my priors was in 2017 when more JFK documents were released and they mentioned Operation Northwoods, which people seemed to take in stride because they were already aware of it. I was not aware of it, though. And to me, even being aware of stuff like MKUltra or the CIA’s efforts in foreign coups, it was shocking that the Joint Chiefs of Staff had suggested a false flag attack on American citizens and JFK’s response was not “Wow you’re traitors I’m throwing you in jail” but instead more like “Good idea but no.”

    Am I misunderstanding the context of this, or is it just common knowledge that these sorts of things occurred constantly throughout America’s history, or am I naive, or what?

    • Conrad Honcho says:

      Maybe a little naive? The powers that be looting and plundering for profit isn’t new. Ever read War Is a Racket?

      • bean says:

        I feel compelled to point out that Smedley Butler seems to have been more than a bit weird, and he was operating in a world very different from what we have today. At the time, it was seen as fairly normal to use military force if your commercial interests were threatened, and he did spend much of his career defending American interests in Latin America and Asia. This more or less ended in the 30s.

    • bean says:

      And to me, even being aware of stuff like MKUltra or the CIA’s efforts in foreign coups, it was shocking that the Joint Chiefs of Staff had suggested a false flag attack on American citizens and JFK’s response was not “Wow you’re traitors I’m throwing you in jail” but instead more like “Good idea but no.”

      Military planners plan. It’s what they do. If asked “OK, how do we justify going to war against Cuba?” they’re going to start thinking of ways. Some of those ways will be morally compromised, to say the least. The final report will be returned with a cover sheet that says something like “In the event that we decide to take action against Cuba, the following methods might be used to ensure public support…” This doesn’t mean that the JCS wants to do these things, any more than the existence of War Plan Red meant that the US wanted war with Britain.

    • MrApophenia says:

      I don’t know about “constantly,” but yes – the old “conspiracy theories must be false because if they ever happened in real life someone would have leaked/reported it” is belied by the actual conspiracies we only found out about decades later.

      And those are just the ones we heard about – we only know about MK-ULTRA because a filing error prevented all the records being illegally destroyed, as ordered.

      Also, as an aside, the JFK docs released as part of the JFK Records Act contain lots of fun stuff already, even with them still not all released. Here is a good summary from Frontline:

      https://www.pbs.org/wgbh/frontline/article/oswald-the-cia-and-mexico-city/

      • Clutzy says:

        A great example of a “conspiracy theory” that is revealed to be partially or even largely true is the Red Scare & Venona papers.

  19. Theodoric says:

    Game of Thrones S08E05
    What is the deal? They build Daenerys up for 8 seasons, bayl gb unir ure ghea vagb gur Znq Dhrra, znffnpevat n fheeraqrevat pvgl?

  20. Andrew Hunter says:

    Why is metal music about things?

    Some discussion on Naval Gazing reminded me of the Swedish heavy metal band I recently discovered, Sabaton. Essentially all of their songs are book reports on famous battles. And really good.

    But that’s simultaneously objectively weird, but not weird at all for metal. So much metal music is topical. For what obviously depends on the band, but you can write albums about the Varangian Guard or Ragnarok or…anythinjg! And this is a uniquely metal trope.

    Sure, people do concept albums or rock operas reasonably frequently, but if I said “You should listen to the Wheat Threshers! They’re a country group that sings about well known sailing ships!” or “You’ve got to check out this album of bubblegum pop where each track gives the life story of a Mongol Khan” you’d look at me like I have three eyes. (And I’d listen the shit out of either of those albums if they existed.) Why is this so much more common with metal?

    • vV_Vv says:

      Metal is highly male-skewed. There are a few famous female vocalists, but most bands are either all male or all male but the vocalist. And the audience is also mostly male, especially socially awkward men. Therefore their production tends to fall within the typical range of nerd interests.

    • Conrad Honcho says:

      BTW, Sabaton has a US tour coming up. I just bought tickets for when they’re in my area.

      ETA: Also, you may like the Sabaton History Channel. These are well-produced (and exceedingly well-narrated) shorts about the real history behind Sabaton songs. It’s the sort of thing the History channel used to do and I wish still did.

      • bean says:

        Also, you may like the Sabaton History Channel. These are well-produced (and exceedingly well-narrated) shorts about the real history behind Sabaton songs. It’s the sort of thing the History channel used to do and I wish still did.

        I would assume they picked a land-warfare historian, but after they were still somewhere north of 1 mistake/minute after 5 minutes on the Bismarck video, most of which didn’t even take books to notice, I’m at least a little skeptical. (I plan to finish analyzing that at some point, but I’ve been busy.)

        • Conrad Honcho says:

          I said they were well-produced and well-narrated. I don’t know how accurate they are. I’m sad to hear they’re not as accurate as I hoped.

        • Conrad Honcho says:

          bean, what were the mistakes you found in the Bismarck video?

          • bean says:

            There’s a list in the comments here. I only got through about 5:30 before I got distracted.

          • Protagoras says:

            Though they could have phrased it better, I thought the Bismarck class was the largest class in service at the time of the battle (no Iowa or Yamato class in service yet, and construction hadn’t started on Vanguard). So if you interpret them as having intended to say largest battleship so far, seems fair enough?

          • bean says:

            I’m sort of torn on that. Yes, calling something the largest doesn’t mean “of all time ever”. But Bismarck/Tirpitz held the “largest battleship” crown for only 16 months going by commissioning dates, so saying it without qualifiers is at least somewhat deceptive. And glorification of the Bismarck design really irritates me, because it was intensely mediocre.

          • Conrad Honcho says:

            Huh. Some of those are iffy, but I don’t think bad enough to make me unsub from Sabaton History. Then again, I’m not the one with his own blog all about battleships and naval warfare, so, there’s that.

          • bean says:

            I always find these kind of errors really hard to deal with on an epistemic level. On one hand, there’s a sort of framework to naval warfare that their historian just might not have, which makes a lot of these errors more likely if you have bad sourcing. The 44% armor isn’t completely outside the realm of possibility if someone is trying to make that number as high as possible. It is, however, higher than the sources I found when I googled for that question. No clue where it came from. Some of it is where poetry gets in the way of history. Size and speed both probably fall under this one. I deal with this, too, and if this wasn’t a topic I cared about so much, I’d probably let it slide. The Anglo-German Naval Treaty is a thing that I could see getting wrong due to unfamiliarity with the subject, as the naval treaties are famously arcane.

            But then you have the one that I really don’t get, the picture of Fantasque. That’s a destroyer (more or less) and not a battleship, and getting a picture of a French battleship is not hard. Recognizing that Fantasque is not a battleship is also not hard. It has single guns, and the last battleship I’m aware of that had those was built in the 1880s.

          • Conrad Honcho says:

            I agree showing a picture of the wrong type of ship is bad. But some of the other stuff it seems like they’re trying to get a point across and the specifics get in the way. I want to come away from the video knowing that, despite being a very large ship, the Bismarck was also very fast and could escape from pursuers. That it was slightly slower than one other ship from 20 years before might be true, but kind of irrelevant. If they included nitpicking facts like that throughout, the video swells in size and becomes harder to find the time to watch. And it might only things more confused. If I come away thinking the Bismarck wasn’t fast because there were faster ships, that leaves me with a false impression of the situation, that the Bismarck could not outrun attackers.

          • Protagoras says:

            A lot of people do seem to be under the misleading impression that larger warships are generally slow, which is not a consistent pattern at all. Bismarck was reasonably fast, but so were basically all battleships built in the 30s and 40s.

          • bean says:

            I want to come away from the video knowing that, despite being a very large ship, the Bismarck was also very fast and could escape from pursuers. That it was slightly slower than one other ship from 20 years before might be true, but kind of irrelevant.

            That wasn’t “this random ship from 20 years before was faster, so this is false”, it was “the battleship* that Bismarck is famous for sinking was possibly faster, and definitely in the same broad speed category”. And that doesn’t count things like shadowing cruisers, which Bismarck was also unable to outrun. (Also note that I sort of waived this one as “poetry getting in the way of facts”. I usually throw in a footnote, but I’m weird.) If I wanted to be really picky, I’d note that “any potential threat” includes aircraft.

            *Yes, Hood was essentially a ~30 kt version of the Queen Elizabeth, including armor.

    • Watchman says:

      Because metal (or more accurately bits of it) is not concerned about being cool and on trend? And isn’t pandering to audiences of young people or traditionalists?

    • rahien.din says:

      You might have that impression because you’re more of a power/folk metal fan. Most of that genre is (and the adjoining genres are) centered on battles, history, mythology, etc. It has to be – the genre is so epic and cheesy that it both lends itself well to depictions of epic events, but also needs to be grounded in something concrete.

      But most metal is not about things. If you listened to more death, thrash, groove, or black, you would find far less things-focused metal.

      • Urstoff says:

        This was my thought. Metal without clean vocals is rarely about anything specific (and definitely not narrative-based), mostly just impressionistic darkness, which isn’t surprising since it’s hard to understand what words are being sung/growled/screeched anyway.

      • Le Maistre Chat says:

        This. Power and especially folk metal is event-centric. Black metal is evil idea-centric. I don’t know a lot about death metal, except the Christian kind, which deals with theology. And I don’t think I’ve even heard any groove or thrash.
        Metalheads are fussy gatekeepers about genre distinctions. The late Christopher Lee made “symphonic metal”, and you can find interviews with him where he pronounces it like that. It was a classic metalhead affectation.

        • rahien.din says:

          You must have heard some thrash – no Metallica or Slayer? And you can’t have escaped the influence of Pantera, original kings of groove.

          • acymetric says:

            Lots of people haven’t listened heard any music by Slayer or Pantera (it would definitely be surprising for someone not to recognize the names, but I wouldn’t expect any given person to have heard any of their music necessarily).

          • Another Throw says:

            it would definitely be surprising for someone not to recognize the names

            Consider yourself surprised.

          • acymetric says:

            I probably should have included some kind of caveat about familiarity with music or metal generally. Maybe age as well (my parents definitely haven’t heard of Pantera, for example…they might have heard of Slayer but would know it in name only).

          • Le Maistre Chat says:

            Of those three, I’ve heard a little Metallica.

    • C_B says:

      Weird indie pop also has the quality of frequently being about things, though it’s usually less “this band’s entire discography is about this one thematically unified category” and more “this band writes songs about oddly specific things, but usually only one or two songs about each thing.”

      Examples:

      The Decemberists’ nine minute long sea shanty about a sailor taking bloody vengeance on a man who mistreated his mother (while in the belly of a whale)

      The Mountain Goats’ historically inaccurate but very catchy tune about how hardcore the Anglo Saxons were (except that from the lyrics, it seems like it should probably be about pre-Anglo Saxon English natives?)

      Tullycraft’s weird ballad about how an over-the-hill record label exec was secretly Dracula in disguise

      From the punk end of the pool, Tacocat’s ode to how great Dana Scully and her shoulderpads are

      —–

      Is it that only metal is about things, or is it that music being about things lowers the breadth of its appeal (compared to songs about how great your new crush is or whatever), so being about things is consigned to niche genres (including, but not limited to, metal)?

      • Nick says:

        Winston Rowntree does it a bit with a songwriter character in Subnormality. This one comes to mind, but there may be better examples.

        • C_B says:

          Man some of that looks interesting but the way it’s rendering as a giant image that you have to scroll sideways to read is pretty off-putting. What’s going on with that page?

          • Nick says:

            That’s Subnormality for you. Exploring the limits of the page is part of the experience.

          • C_B says:

            OK, I’m giving it a shot. It’s a bit more tolerable now that I’ve figured out that the whole thing is the comic, and it’s not just badly mangled HTML mixing the comic with the author’s commentary on the comic in unreadable ways.

    • toastengineer says:

      While we’re at it, I’d like to plug my favorite band Iron Savior, whose first four albums form this huge epic about an AI-driven spaceship trying to figure out its relationship with humanity. The rest of their songs are usually About Things as well, and there’s usually one song on each album that ties back in to the original quadrilogy.

      Skip Unification, and start with Dark Assault since it’s the best musically and the simplest story.

      • Le Maistre Chat says:

        While we’re at it, I’d like to plug my favorite band Iron Savior, whose first four albums form this huge epic about an AI-driven spaceship trying to figure out its relationship with humanity.

        Cool. I’ve heard of concept albums, but continuing a story over four?

        • Wander says:

          Radical Face did a story about a weird cursed family over four albums (three for generations and one for misfits), so there’s a second example.

    • Eugene Dawn says:

      Folk music has a lot of songs about historical subjects; and the protest song tradition means there are lots of songs that are topical in some sense. So, I’m not sure this is more common with metal.

    • J says:

      bubblegum pop where each track gives the life story of a Mongol Khan

      Sorry, you’ll have to settle for this alternative band’s song about a punk band made up of Mesopotamian kings:
      https://www.youtube.com/watch?v=jAMRTGv82Zo

      Or historical US presidents:
      https://www.youtube.com/watch?v=H9SvJMZs5Rs

  21. ana53294 says:

    Why haven’t the UK parties split?

    I know the UK has a FPTP system, and that complicates the matters, but I don’t think that necessarily means that you have to have a two-party system.

    In Spain, the two-party system remained for more than 40 years despite not having a FPTP system. So it appears that a two-party system is created by some other factors, although FPTP facilitates it. But in Spain, the parties have split, even though for years they were riddled with corruption and people wer deeply unhappy about them; during those years, the two party system remained. But that has changed very fast.

    Many people are unhappy with Labour – not just Remainers. Why was the excision from Labour so small? Why hasn’t the Tory party broken down?

    Is winning the overriding concern in British politics? Because in Spain, we have many parties that break down and separate -even though they know they won’t win any MPs at first. Vox was such a party at first.

    Why haven’t any of the smaller parties gained any traction, despite the disastrous handling of Brexit? Are Brits much more likely to vote tactically and thus not vote for a party they like if it’s unlikely to win?

    In Spain, voting tactically is not a thing; parties sometimes go together in an election, making a pre-election coalition. But it’s the party, not the voters, who make the strategy. For the European elections, for example, Basques, Catalans and Galicians go together, as well as lots of small parties that don’t get any representation.

    • vV_Vv says:

      In Spain, the two-party system remained for more than 40 years despite not having a FPTP system. So it appears that a two-party system is created by some other factors, although FPTP facilitates it. But in Spain, the parties have split, even though for years they were riddled with corruption and people wer deeply unhappy about them; during those years, the two party system remained. But that has changed very fast.

      A two-party system may or may not exist in a proportional representation system, but FPTP pretty much forces it. Three-party scenarios in FPTP are very rare and unstable, the only example I can think of is the American Republican Party splitting off from the Whigs and quickly replacing them, but that was at the verge of a major civil war.

      • HeelBearCub says:

        Specifically FPTP of the parliamentary kind in Britain forces local single party dominance trends. That’s why, IMHO, we see the Scottish National Party as viable in a way that we don’t in the presidential systems (AFAIK).

        The punishment in Britain’s system for not forming a local coalition that can get the most votes is zero representation. However, unlike a presidential system, the SNP does have the ability to make demands of coalition partners for purposes of forming a government. There is no direct vote on the chief-executive that forces pre-election coalition building at the national level.

        Perhaps the comparable situation would be the States’ Rights Democratic Party (A.K.A Dixiecrats) and the American Independent Party in the US in 1948 and 1968. They punished their (former) coalition partners for embracing civil rights but essentially had no impact past their single elections. The national coalition was in flux, but the net result was that enough unsatisfied voters for the presidency ended up aligning themselves with the Republicans to give that coalition control of the South. By 1984 the instability was essentially resolved, even though the locally dominant party was for the most part still the Democrats.

        • ana53294 says:

          The issue with making agreements for each seat is that you can’t do as much horse-trading.

          The optimal for several parties would be to create a regional coalition. So for example they would have a single coalition for Wales, and they put the most popular party in each seat list, for a proportional number of MPs.

          • HeelBearCub says:

            The issue with making agreements for each seat is that you can’t do as much horse-trading.

            Not sure I quite understand your point, can you expand?

            The optimal for several parties would be to create a regional coalition.

            The problem here is that the coalition is either unstable, or it just devolves to a single party.

          • ana53294 says:

            Let’s say that you have two parties that may get 5 MPs.

            Let’s also imagine that the party with more votes has more votes in the majority of those seats (60% of the votes they get). If you make a deal in each seat individually, the party with more votes gets each seat, so the party with more seats gets 5 seats, when they only brought 60% of the votes.

            Pre-coalition agreements are not legally binding; the only way the other partner has to punish the other for not following the agenda is to not go with that party in the next election.

            But if the minority party gets 40% of the seats, the last 2, or every fourth and fifth seat they get, then the party has a way of enforcing its agenda by having those MPs negotiate their points.

            Coalition partners frequently don’t go again in the next election; coalitions change all the time. Usually, the coalition that gets the power ends up getting more votes in the next election.

            In Spain, the coalitions that are stable are between those parties that don’t compete at all. The Basque+Catalan coalition for the EU Parliament is thus very stable, because they don’t compete; one winning does not weaken the other. And even that one was weakened by unease on the moderate Basque party due to the situation in Catalonia.

            Politicians who achieve things are popular; this is especially so with City Mayors (the place where most coalitions happen). Madrid’s Manuela Carmena was greatly strengthened by their coalition with Podemos, because they achieved many things in Madrid, and she is very liked.

            Thus the weaker coalition partner is the one that is more likely to walk away, because they lose more if they win.

          • HeelBearCub says:

            In some ways, you appear to be talking about proportional representation and not first past the post. My comments only apply to FPTP.

            Assuming you are talking about FPTP, and you mean two sub-parties that provide at least ~30% and ~20% of the total votes, and thereby gain a victory, there is no way in FPTP system to do this other than to coalesce under a single banner. Either that single banner becomes the dominant party, or that banner falls apart.

          • ana53294 says:

            But that banner can be regional, and the alliances can change in different regions.

            In Spain, the left-wing party Podemos allies itself with regional parties, and they go under a different banner in each region: Elkarrekin Podemos, Adelante Andalucia, etc.

            In Spain, coalitions are vastly different depending on where they happen.

            Why wouldn’t it work?

          • HeelBearCub says:

            @ana53294:
            Looking briefly at Spain’s system, it appears that the more powerful of the two congressional bodies, the Cortes Generales, is elected using proportional representation, rather than direct election of individual congress members via a first past the post system.

            Spain doesn’t operate by the same basic rules as the UK, so you are going to have trouble understanding why things happen as they do unless you take those differences into account.

      • Procrastinating Prepper says:

        Just chiming in to note that Canada’s system is FPTP, has had three big parties for decades (though the third-most popular has never been in government) and just had a fourth party bud off from the Conservatives last year.

        FPTP does put a lot of pressure on parties not to split apart, but it’s not guaranteed.

        • Is that partly because Canadian provinces are more important than U.S. states, so a third party with a strong base in one province but no national power can do better than the U.S. equivalent?

          • Eugene Dawn says:

            I think the main reason is that Canada, being a parliamentary democracy, allows for a bigger role for third parties than the American system: the NDP were necessary for Lester Pearson’s government in 1963, which implemented the major NDP priority of universal healthcare. The same happened again with Pierre Trudeau’s minority government of 1972, when the NDP were able to get priorities like the creation of a state-owned oil company.
            But you’re probably right that provincial control matters too: the NDP pioneered healthcare in Saskatchewan when they had control of the province.

        • HeelBearCub says:

          Well, in a parliamentary system there is much less pressure to avoid splitting than in a presidential system. If you have a plausible path to a plurality in a fair number of districts, you have a good chance of being able to dictate some terms in forming a government.

          The UK already has minority parties, with 8 parties comprising a total of 89 of the 650 seats.

      • Nornagest says:

        They’re unstable, but they’re not that rare. The Republicans are the only long-term successful example (the Whigs came out of a period of effectively single-party politics), but the American Independents (George Wallace’s party), Dixiecrats, Progressives (aka Bull Moose), Constitutional Union, National Republican, and Anti-Masonic Party all took states in presidential elections, and several other parties — most recently the Reform Party under Ross Perot, in 1992 — have had enough success to influence election results without taking states.

    • Watchman says:

      Because very few members (subscribing members that is, not supporters) wish to damage what are still fairly functional electoral coalitions over a single issue. For example, the Conservatives are generally low-regulation, low-tax, law and order, and this still applies now. That their leader is willfully moving away from this position explains why she has no support from any substantive group in the party, but as can be seen by the fact that contenders for the leadership are gathering support from both remainder and leaver MPs into their unofficial campaign teams, there is much more uniting the party than dividing them. Labour is probably more deeply split between the ‘third-way’ followers of New Labour and the new/old revolutionary socialists, but its members have strong tribal loyalty and a belief in standing and fighting if the cause is right, so the tensions have been relatively contained, with a focus on making people’s lives better uniting everyone, at least against other parties. I think it’s a case of the parties believing Brexit will one day be done, and the underlying political divides will again be important. This is probably missing the changes the Brexit process is causing, but whatever the faultlines of politics are after this mess is resolved, it will likely be the existing parties (or if Labour splits due to internal tension, something else in the mildly-left-of-centre range) articulating the two sides of that fault.

      And check the current opinion polls if you think no other parties.are getting traction. The subtly-named Brexit Party is way out ahead… No-one thinks this is sustainable but UK voters don’t stick with a party that they feel ignores them.

    • NostalgiaForInfinity says:

      The Labour party split in the 1980s and it didn’t go terribly well for the group that left (the SDP). They (in an Alliance with the Liberal party) got almost as many votes as the Labour party but a tenth of the seats. This probably puts most people off trying again.
      This might be why the recent split of Change UK (initially the independent group) was quite a small group – the threshold for MPs to leave their existing party is quite high.

      Party/tribal loyalties are also pretty strong. Many Labour MPs would refuse to be in a party with a substantial number of former Tories (the converse is likely true but I’m more familiar with the Labour party).

    • Garrett says:

      To add another point, political parties aren’t primarily social clubs, they’re more like institutions. Imagine for a moment that you complained that your spouse burned the food last night and a co-worker asked “so why haven’t you divorced them yet?” In that case, the institution involved (marriage) is much larger than a single-issue. Sure – in political news terms, Brexit is absolutely dragging on. But in institutional terms it’s barely a blip.

  22. Machine Interface says:

    Three days ago, French special forces conducted an operation against a small djihadist camp in Burkina Faso to rescue two French hostages who had disappeared on May 1st. Two soldiers and four hostage takers were killed in the assault, but the hostages were safely freed, while the rest of the hostage takers fled.

    Here’s where it gets weird: not two but four hostages were found in the camp. In addition to the two Frenchmen were a US citizen and a South Korean citizen, who had been captured 28 days prior. Here’s the thing: nobody had signaled their disappearance; in spite of being in liaison with US intelligence, the French intervention force had received no evidence that those were there/could be there.

    Little information seem to have been released about the two extra hostages besides their nationality, and news outlets don’t seem to care much, most having already moved over to other things (French media are still talking about the incident, but only on the angle of whether the French hostages were reckless or not to go on vacation in that area).

    All of this has an intriguing aura of mystery.

    • aristides says:

      I read the official story on vox and France24, and it seems plausible. The kidnappers were planning on transferring the hostages to Katina Macina, a terrorist group. They would want to keep the kidnappings as secret as possible until after the transfer to prevent a special forces raid. US and South Korea don’t keep a close enough of an eye on all their tourists to tell when they disappear for 23 days, unless someone reported it. The two women were really lucky to be rescued.

      • Aapje says:

        One would still expect the women to have family in the US and South Korea, who would have regular contact and who would notice & report their disappearance.

        • Watchman says:

          That’s an assumption though. And we don’t know these ladies were not reported missing, just that the French weren’t expecting to find them.

    • Paul Brinkley says:

      Given how I would expect things to go if true, I think it’s also possible that both of the extra two hostages were not mere hostages, that French authorities did know, and that that knowledge was kept secret.

      That would explain the US / SK knowing they were each missing a person, without it being public knowledge. (Families would know as well, but they tend to be prepared for this.)

      If so, that means the irregularity was in you hearing of their existence. This is possible; reporters sometimes dig really well, or one official leaks by accident, or both.

  23. Le Maistre Chat says:

    Local governments in the US are dependent on property taxes, but publicly-traded retailers now use something called “dark store theory” to defeat tax assessors in court. This has created a bipartisan outrage (this journalist’s star is a lawmaker with multiple conservative bumper stickers).

    • Erusian says:

      As someone pretty heavily involved in the real estate world, this article doesn’t seem to understand how property taxes get set. It also implies aggressive anti-tax appeals are a new phenomenon. They’re not: they’ve been around since the ’90s at least. And probably longer, I only say the ’90s because I wasn’t around before then.

      Basically, property taxes are a percentage of your property’s value. Now, naturally the question is what is that value? When the property is first bought, it’s pretty easy to figure that out. How much did it sell for? What additional basis does the person have? But as time passes it varies. The town will periodically adjust valuations and people can challenge these through institutions that vary by state/county. They will be arguing they are overvalued. The government will be arguing they’re not, or are undervalued.

      A common way to determine price when there is no sale in all real estate is the use of comparables. People find other sales that are similar and use those as a comparison point. So, let’s say you are valuing a three bedroom two bathroom home with 2,000 square feet and a small yard. If you have three other homes nearby with the same features going for $195k, $205k, and $202k, then it’s safe to say the house is worth about $200k. A lot of valuation is people arguing about how they’re similar or different and thus valid comparisons.

      In this case, the corporations are finding former big box store real estate that sold and claiming those as comparables. Of course, those sales are often very low priced because no one is lining up to buy an old, out of use Walmart. However, that is the effective value of the asset if Walmart was to sell the store, so that is the correct way to value them. The amount of commerce taking place in them or the value to the company is irrelevant, just as the fact I work from home does not make my real estate any more valuable. The value of the underlying property is what’s being taxed.

      Notice how all the arguments are about how those dastardly corporations are getting out of their taxes which the poor, poor municipalities need? This isn’t sound legal reasoning. Their ‘common sense’ boils down to ‘we need the money’. They also don’t seem to understand, or the article writer doesn’t seem to understand, that property taxes are not taxes on activity or wealth. They are taxes on real estate. A thriving Walmart is worth more than an empty store but the underlying real estate isn’t necessarily that different.

      I’m not unsympathetic to the need to raise municipal funds: raising property taxes or levying a sales tax are ways to get at that wealth. But all the corporations are doing is making sure their real estate is valued in line with what they’d get if they sold it. Which is what the tax requires. If you don’t like that, and I think there are very good reasons to not like that, change the tax.

      • silver_swift says:

        In the case of Wallmart, are we taxing the value of the property should wallmart decide to sell or the value that the property is currently providing Wallmart (ie. the amount of money it would cost them to buy a new, similar, location)?

        I can imagine those two values can be quite dissimilar in practice.

        • A Definite Beta Guy says:

          Who knows, but these cases are being brought up to courts and courts are often siding with the retailers.

          Either way, the article described using comparable pricing as a “tax loophole,” which is just crazy. Even if you disagree with the comps being used, trying to find comparable properties is NOT a tax loophole.

          • Erusian says:

            Either way, the article described using comparable pricing as a “tax loophole,” which is just crazy. Even if you disagree with the comps being used, trying to find comparable properties is NOT a tax loophole.

            Seriously. To people who know this industry, the accusation sounds like: Corporations roll out despicable new tax strategy of filling out IRS forms. Comparables are literally that common and central for determining value.

          • dick says:

            Seriously. To people who know this industry, the accusation sounds like: Corporations roll out despicable new tax strategy of filling out IRS forms. Comparables are literally that common and central for determining value.

            The article is complaining that vacant stores should not be accepted as valid comps for thriving stores, not that comps in general are illegitimate. This is a good time to consider the heuristic that the dumber the position you’re arguing against seems, the more likely that you’re attacking a straw man.

          • acymetric says:

            @dick

            Would a good analogy be something along the lines of trying to argue my housing valuation down based on the sale price of the dilapidated shack a half mile down the road?

            Or maybe to be slightly more fair, basing it on the sale of a similar house, but which has been unoccupied for some time and fallen into disrepair (obviously suppressing its value in a way that would not apply to my house)?

          • Erusian says:

            The article is complaining that vacant stores should not be accepted as valid comps for thriving stores, not that comps in general are illegitimate. This is a good time to consider the heuristic that the dumber the position you’re arguing against seems, the more likely that you’re attacking a straw man.

            I will consider it. Will you consider that I might know what I’m talking about? Because you’re using a strawman. Neither of us say the article objects to the use of comparables. It clearly doesn’t. What we’re saying is it is treating the standard practices of comparable pricing models as a dastardly corporate trick.

            The standard practice is to use for sale (which usually means vacant) and recently sold properties because these have active or recently active prices. Likewise, the idea that an active store has real estate worth more than one not in use is odd to me. If the store is generating value above and beyond its real estate then that is not covered by a real estate tax. It is covered by sales and income taxes. They explicitly break it down into so much for the land itself and so much for the improvement. But there’s nothing to tax activity on the land.

          • dick says:

            To people who know this industry, the accusation sounds like: Corporations roll out despicable new tax strategy of filling out IRS forms. Comparables are literally that common and central for determining value. … Neither of us say the article objects to the use of comparables.

            …and that’s my exit!

          • A Definite Beta Guy says:

            The article is complaining that vacant stores should not be accepted as valid comps for thriving stores, not that comps in general are illegitimate. This is a good time to consider the heuristic that the dumber the position you’re arguing against seems, the more likely that you’re attacking a straw man.

            I don’t see how you think this makes the article seem less dumb. Vacant stores can be used as a baseline to reset the value of properties, they just shouldn’t be used without thought into improvement values. There is a retail park on the way to my in-laws that has 7 Big Box Stores, of which 6 are empty. The remaining Big-Box store should be reassessed based on the value of the other 6, which are right next to it.

            So as for this:

            Besides, there’s something plainly illogical about the argument. “Do you want me to value your house as if it’s closed and boarded up?” said Krause.

            There’s nothing illogical about this if you are talking about two otherwise similar properties. What’s illogical is valuing that 7th property at its sale price before 9/11 because your handbook says so, which is what this article initially quotes:

            As of 2017, the city had valued that store at $11 million, a number based on what the property had cost the owner to buy back in 2001, plus the added value of renovations over the years, adjusted at the going rate of depreciation.

            You don’t think maybe there should be a reassessment after 16 years?

            Plus:

            But a tax agent from Chicago filed an appeal on behalf of Sam’s Club, arguing that the store was worth just $7.2 million, based on the low sales costs of a handful of second-generation big box locations scattered around the state. The comparables that the agent provided included three former locations of the now-defunct electronics retailer American TV, an old Lowe’s, a former Target, and a former Walmart (actually, the same property in Greenfield Williams and I were standing in front of now). All of them sold for between $2 million and $4.5 million between 2012 and 2014: much lower sales prices than what their original owners had purchased them for years before. Some had second-generation occupants; some were bank-owned.

            The store isn’t arguing it is the same cost as an old vacant store, but in some cases 3 times as much.

            I am not going to be an expert in every single one of these instances, but this is not a tax loophole, and there’s nothing wrong with it at all, which is why these retailers are often winning their cases in court.

        • Erusian says:

          In the case of Wallmart, are we taxing the value of the property should wallmart decide to sell or the value that the property is currently providing Wallmart (ie. the amount of money it would cost them to buy a new, similar, location)?

          The former. At least in practice. The valuation is what it will be sold for not the opportunity cost of moving because it’s based on sales.

          Here’s how it really works. Each municipality has an (or many) assessors. They also have either a court or a panel. All these are political appointees and usually reasonably well paid (often in the $50k range) for a job that gives professional prestige (assessors are a profession with licensure so people who are ‘official’ get extra work and can charge more) and usually only works a quarter to half the year. The politicians mostly don’t care about property taxes. They do care about voters.

          And so the assessors and panel don’t want the appellants to bellyache. Because if the voters don’t like one, the politician isn’t going to die on that hill. So they tend to be very gentle. Almost everyone wins who actually goes to court, to the point where 50% win sounds low to me.

          It’s also really not that adversarial. The attorneys and panelists and assessors are all professionals in roughly the same industry. The number of people appealing is also generally small. And despite the fact that Walmart is an easy target, dense urban areas tend to generate more taxes per acre than wealthy or commercial areas. So even large swings for an individual taxpayer are often relatively small in total collections, outside of very small towns.

      • Le Maistre Chat says:

        Notice how all the arguments are about how those dastardly corporations are getting out of their taxes which the poor, poor municipalities need? This isn’t sound legal reasoning. Their ‘common sense’ boils down to ‘we need the money’. They also don’t seem to understand, or the article writer doesn’t seem to understand, that property taxes are not taxes on activity or wealth.

        Right. Legally the municipalities don’t seem to have a leg to stand out. The writer seems to be working the angle that “Retail corporations like Walmart are so bad, even bumper sticker Republicans are fighting them.” She talks about how if retail corporations get away with litigating their property tax assessments, the average property tax on residential real estate will have to go up $385 a year… but yes, why not change the tax law rather than arguing emotionally?

        • J Mann says:

          There’s an argument on each side, so the munis win some, and lose some.

          I agree that it’s not the moral issue that most articles make it out to be.

          1) Many tax codes are written so they are required to assess at the “market value.” Two potential ways to estimate market value are (a) what are the up-front costs the developer paid to buy the land and build or customize the building, less depreciation, (b) what is the current tenant paying in rent, and (c) what do we think someone else would pay if the current tenant moved out.

          Municipalities like (b), but in many cases, a lease payment includes a financing component (or you could shift things the other way and pay less rent and more financing), so it’s often inaccurate. Taxpayers like (c), because the data often indicates that once the original tenant leaves, a box store doesn’t sell for nearly as much as the muni thinks.

          Both approaches have advantages and drawbacks, so the courts come out different ways.

      • Mark V Anderson says:

        I don’t think I am as knowledgeable as Erusian, but I have some knowledge and experience with property tax. I think Erusian overstates the business case, even though the article understates it.

        In this case, the corporations are finding former big box store real estate that sold and claiming those as comparables. Of course, those sales are often very low priced because no one is lining up to buy an old, out of use Walmart. However, that is the effective value of the asset if Walmart was to sell the store, so that is the correct way to value them.

        This isn’t really true. The valuation of real estate is based on the highest and best use of the property. In other words, if you are located in a hot retail market, but you are using the building as storage space, tough luck, the value is based on the more valuable possible use as a retail store. Similarly, the vacant stores presumably don’t have a lot of value as a retail space, because if they were in a valuable location they probably wouldn’t be vacant. On the other hand, most successful big box stores are not successful just because of their location. They might well be just as successful if they were two miles away, because the folks that go to these stores all have cars anyway. And of course the ominous phrase “dark stores” is clearly a spin by those against using the vacant store method. I presume the other side uses the phrase comparable store sale value. My thought is that courts should probably split the difference between vacant store sales values and the depreciable value method, but that’s just off the top of my head.

        • quanta413 says:

          I don’t think it’s true of most big box retailers, but Walmarts in particular are very often in crap locations at the very edge of town or a bit further. How many stores could realistically replace them and extract the same sort of economic value from the land? Where I live now has three Walmarts, and only one of them is near a location that would pull in any significant number of customers without a Walmart there. The other two are at the very edges of town plus a tiny bit and there are no signs of further development reaching that far soon.

          • AG says:

            That seems based on region. Most places I’ve been, Walmarts are either in the same kinds of locations as other big box retailers (on the edge of suburbs by the highway, and so drawing in commuters, errand-day shoppers, and roadtrippers passing through alike). There are also Walmart Neighborhood Market stores, which focus on suburbs grocery and so compete in the same space as Safeway/Albertson’s/etc.

            And if there are Walmarts a bit further from the edge of town, that’s because it’s in a region where inter-town commuting is common, as per Mark V Anderson’s scenario.

            That said, the reason a Walmart big box can also extract the value it does is because of its one-stop-shop nature. It’s true that there aren’t many stores that could realistically replace them and extract the same value, but a plaza-type setup might (aka what has replaced the traditional mall), because it similarly consolidates multiple errands into a single (park the car once) location.

            In Texas, going anywhere to do anything was like 40 minutes minimum one way, because you were going to another town to do it. The big box stores and malls and shopping plazas being a little outside the edge of town actually shortened the distance to a mere 20 minutes.

        • Erusian says:

          The valuation of real estate is based on the highest and best use of the property. In other words, if you are located in a hot retail market, but you are using the building as storage space, tough luck, the value is based on the more valuable possible use as a retail store.

          Do you have a citation there? Because in my experience it’s not based on highest and best use (at least I could not find the phrase). My experience is the current use is completely irrelevant. It is based on the value of the real estate regardless of use.

          Let’s say I own an empty lot in a very hot real estate market. Its property taxes are the same regardless of whether it’s completely empty or if I use it to hold a super successful urban outdoor event. The property tax of all empty lots is not based on the theoretical best and most profitable use of empty lots. Now, sales and corporate income tax would be very different. But property tax isn’t meant to cover that.

          That said, while I find the municipality’s case spurious, they do have some case. In particular, because big box store sales are so rare it can be extremely difficult to value them. But you’re absolutely right this is a weird angle of attack because real estate isn’t a big box store’s big asset. They buy terrible land and expect people to drive out to them. So a property tax is not a great way to access that value. I agree it’s a good tax base but not through property tax.

          • Mark V Anderson says:

            Do you have a citation there? Because in my experience it’s not based on highest and best use (at least I could not find the phrase). My experience is the current use is completely irrelevant. It is based on the value of the real estate regardless of use.

            Hmm. I couldn’t find it either, although most of the stuff on the Internet isn’t nearly nuanced enough to talk about such things. I know I’ve heard the phrase, probably from a class in my Master’s program in tax decades ago. But I found my old book I used, and can’t find the phrase. I think it is a valid concept, but I have no cite.

            Your comment that current use is irrelevant though is in complete agreement with highest and best use. As I said, if you are using a building for storage when it has a higher value as something else, the something else is what counts, because that is highest and best use.

    • Andrew Hunter says:

      This seems pretty solvable with my favorite pet theory of tax valuation (I can’t remember who I first got this idea from, though it certainly seems like something David Friedman would endorse.) It’s generally proposed in terms of estate taxes.

      – The government can declare any object/property/whatever it is taxing to have any valuation they like; there is no appeal or objective method. If they think the spoons I inherit from my grandma are worth $20B, that’s what they’re worth, and I owe the appropriate taxes.

      – Any such property that’s been given a valuation is legal tender for any taxes in the amount of that valuation. So I hand them the spoons, and demand my $10B in change.

      Seems to stop pretty much any of these games. If you think the government over-values your property, give it to them. If you think it’s undervalued, pay the taxes. Works for property taxes too, though at that point being able to convert it back to cash and get “change” is far more crucial.

      • RalMirrorAd says:

        This might work In the case of egregious overvaluation but there’s a psychological and financial cost to moving involved that adds stickiness to the transaction. My property might go for 170K according to the zestimate but the government assesses it at 175K, I could offer to sell but the government might try calling it a bluff and in all likelihood it would be.

        In commercial real-estate it’s going to be even stronger because while it might not be wise to tax the economic value that the property generates as property tax, said economic value is still a factor in whether the company chooses to remain there in spite of the taxation level.

        • silver_swift says:

          It might not be wise to tax the economic value that the property generates as property tax

          This is non-obvious to me, could you elaborate?

          • RalMirrorAd says:

            I worded it poorly but I sort of meant that that if property is deployed in such a way that its active usage generates income to the owner, to the extent that is taxed, it disincentives that activity.

            Whereas property can be value simply because it is in a favorable spot which didn’t require any action on the part of the owner, if taxed, encourages the owner to make the most of their property since the value doesn’t scale to his actions.

            I also meant ‘might not be wise’ because the argument made by Eurasian.

          • A Definite Beta Guy says:

            If you set-up a McDonald’s and a Burger King next to each other, you do not assume the McDonald’s has higher property value just because it has more money. It has more economic value because McDonald’s is a better business.

            You should not be taxing this value as property tax, because it sets up extremely bad incentives. For instance, making it cheaper for an unprofitable business to run where a profitable business would actually create higher value.

            You already tax economic value added through income taxes.

          • Tenacious D says:

            Some theories of land value taxes take this further and say it isn’t wise to tax the value of improvements: why should the owner of a vacant warehouse with broken windows be penalized for fixing it up into chic lofts?

      • Radu Floricica says:

        Too easy for assessors to make money from overvalued evaluations. Especially since we’re talking about local administration, where everybody may be ok with it.

        • Andrew Hunter says:

          How? If assessors intentionally highball, property owners sell, and the assessors paid top dollar for stuff they can’t resell at that mark in the open market. How do they profit here?

          The only way I can see is them being bribed by the lucky property owners, but that’s obviously illegal and not any harder to deal with than current corruption.

          • Radu Floricica says:

            I see about an order of magnitude more likelihood of corruption in this model, because in involves potentially unlimited money to be made.

            Take an edge case. You have an old lady with an old house that she can’t pay the taxes for. She’s a very very nice lady, and it’s really not her fault she can’t pay – and she has no family and no other place to live. In the current condition, the assessor can at most give her a discount. State loses petty cash (she doesn’t live in a palace), old lady gets a break. Not legal, arguably not fair, but damage is limited to part of what she has to pay.

            In the proposed model, the assessor can mistakenly up the evaluation – not much, just enough for the old lady to be able to move to Florida and get her hip replaced. Well within legal error margins. Happy ending. Damage for the state – about a couple of orders of magnitude higher than in the first case.

            And that’s just skirting illegality, not trying to sell a dump for a million – which I guarantee will happen, and more so in certain small towns.

            Also, there’s a small Taleb inside me that screams at the idea of giving bureaucrats a truly blank check and hoping they do the legal thing.

      • What you might have gotten from me is a somewhat different version—the self-assessed property tax. You state a value for your property, on which you will be taxed. By doing so you are offering to sell the property at that value to any buyer who wants it. Not my invention, but an interesting idea.

        Something along those lines was used in Periclean Athens. If you were one of the richest Athenians, you were required to produce one public good every other year—sponsor the Athenian team for the Olympics, pay the costs of a warship for a year, or the like. If you were assigned one such job, there were two ways to get out of it. One was to show you had already done one last year or were doing one this year. The other was to show that there was another Athenian who wasn’t doing one this year, hadn’t done one last year, and was richer than you.

        How do you prove he is richer than you? You offer to exchange everything you own for everything he owns. If he refuses, he has admitted he is richer than you and gets the job.

        The modern equivalent is a claiming race. If you enter your horse in a ten thousand dollar race, you are offering to sell him for ten thousand dollars. It’s an elegant way of getting people to enter horses of about equal ability.

    • The Nybbler says:

      The outrage is misplaced. The tax is a property tax, on the value of the land and improvements. To first order, the value of a property with an operating business and a similarly-situated one just vacated are the same (though the improvements will deteriorate). If you want to tax the value of the business rather than the property, you need a different tax.

      • Le Maistre Chat says:

        If you want to tax the value of the business rather than the property, you need a different tax.

        Isn’t how it works that a tax on the value of the commerce already exists and goes to the feds and state as an income tax on the business, while local government only collects a tax on the value of the real estate?

        • HeelBearCub says:

          At least in my state, municipalities and counties are very constrained by the state legislature on what taxes they can impose. Pretty much sales taxes and property taxes. If you want something like a progressive tax, property is all you can really leverage.

        • The Nybbler says:

          Local government can sometimes impose income and sales taxes as well. If they can’t by higher government law… well, too bad. One of the major things governments do is make up rules and penalize people for not following them; if they’re going to do that they can surely follow the rules which apply to them.

      • BBA says:

        To first order, the value of a property with an operating business and a similarly-situated one just vacated are the same (though the improvements will deteriorate).

        No it’s not. Suppose the property is owned by a REIT and leased to a business. That REIT is going to get a lot more for it if the business is going to continue operating as opposed to if the business has just closed. Land + improvements + income from rent > land + improvements.

        Now you can say the property tax should only apply to the land and the improvements, and the rents are intangible and untaxable. Fine, but when REIT2 buys the store from REIT1, it’s paying one lump sum for all three, and that’s all the information the assessors get. How are they supposed to break it out?

        Not to mention how intimately the property value is tied into the economic state of the property’s surroundings. A strip mall next to a Walmart is going to be worth more than a strip mall next to a vacant big box, independent of what’s going on at the strip mall itself.

        • The Nybbler says:

          No it’s not. Suppose the property is owned by a REIT and leased to a business. That REIT is going to get a lot more for it if the business is going to continue operating as opposed to if the business has just closed.

          That’s including the value of the lease along with the value of the property, however.

          • pjs says:

            That’s including the value of the lease along with the value of the property, however.

            I don’t see how this distinction is meaningful, assuming we are talking about ‘market value’.

            In the commercial property markets I am familiar with (not US), the going market price of a solidly leased (or readily lettable) place is – to a good first approximation – going to be close to the risk-adjusted present value of the lease (or expected lease return). Unusual features, risks, or opportunities merely perturb this price unless extremely idiosyncratic.

            If you removed the capitalized value of the expected rent minus expenses from the actual market price, to try to get at “land plus improvements” value you’d end up near zero. That’s because a market price including the lease value is not really double counting the value of land (whatever that may mean) or improvements. The total price is one way of looking at it or the other.

            It all gets murkier when the lease is dubious and/or the prospects for releasing are questionable. It would be a clear error to take the current rental, capitalize as if that would continue for decades, and expect the market to agree. But it would be the opposite error to say: ignore lease because it’s value is something above and beyond the underlying “land plus improvement” value. (As noted above, if the lease prospects were not so dubious, you’d end up near zero this way, so it’s wrong in that case; it suddenly doesn’t become right to do so in the murkier cases.)

          • The Nybbler says:

            In the commercial property markets I am familiar with (not US), the going market price of a solidly leased (or readily lettable) place is – to a good first approximation – going to be close to the risk-adjusted present value of the lease (or expected lease return). Unusual features, risks, or opportunities merely perturb this price unless extremely idiosyncratic.

            The conditions _are_ extremely idiosyncratic, and the difference between “solidly leased” and “readily lettable” is large. Suppose 10 companies entered the town, Walmart and WalClone 1-9. Each bought a substantially identical piece of property and built up substantially identical buildings on them. Then a recession hit, and while WalMart opened, WalClones 1-9 all went out of business. On opening day, your tax assessor comes around to all 10 properties. The tax assessor says “OK, the building with the going concern is worth 50x more than the vacant buildings”. So WalMart does the math, packs up its store, and moves across the street into the WalClone 1 building (which it picks up for a pittance), abandoning its old building to the taxman.

            Obviously WalMart can do this because it owns the building; there’s no lease. That’s what demonstrates that you’re including the value of the lease when you value the property with an operating business higher than the vacant ones.

          • pjs says:

            > The tax assessor says “OK, the building with the going concern is worth 50x more than the vacant buildings

            Then the tax assessor would be wrong, at least assuming his job was to determine market value. And, as you say, Walmart could readily disprove it by moving or sale/leaseback. The value of market lease terms would not justify a big premium – if any – in this sitation. It’s true Walmart doesn’t have an market-negotiated lease (indeed no, lease at all) so the assessor would have to impute that to make progress, but inferring market values in absence of enough actual transactions is basically his job.

            Let’s simplify your situation a bit. All properties are vacant – obviously their owners are looking for tenants – and then Walmart decides to move in and negotiates a lease from one of them. It’s going to get great terms! The final lease is going to have a value: it’s a return of $X over some number of years, subject to certain risks and costs, but Walmart is a stable company and the costs are generally predictable, so we can take a reasonable stab at a present value of this lease, say $Y.
            The owner of the property (who got Walmart into his building) decides to immediately sell (with the Walmart lease intact); so you might want to say he’s selling a property + lease bundle. We also know roughly what he’s going to get for this bundle – something only a bit higher than all the unoccupied properties. (He won’t get much more, because the other property owners would have negotiated correspondingly harder to get the Walmart lease so they instead could sell and reap the windfall.) Call this $Z.
            If you insist that lease value is something over and beyond property value, you might want to say that the property value must $Z – $Y. But that’s just absurd – that’s almost certainly going to be far less than the prices that the unoccupied properties are selling for (the longer the lease, the closer to zero the difference will be.) Surely the property value can’t be reduced by obtaining a lease?

            I contend that “lease value” as something that exists over and above property value doesn’t make sense, even in a distressed market, except perhaps to the degree that an actual lease in place might not reflect what a lease newly negotiated in that market would be.

          • The Nybbler says:

            I don’t claim the property value is $Z-$Y. I claim that in this distressed market, the market price of the leased property (including lease) might be significantly greater than the market price of one of the vacant properties. And that this difference is properly attributed to the lease; specifically, it is the difference between a market lease already in hand and the expected value of a future market lease, and therefore is not part of the “land + improvements” value.

            To make things worse, the lease may have actually resulted in a reduction of market value for the other properties. Because if Walmart is leasing one building, it’s probably not leasing any of the others, and the market for big-box properties is probably quite small. Where you assign that, I have no idea.

          • pjs says:

            I claim that in this distressed market, the market price of the leased property (including lease) might be significantly greater than the market price of one of the vacant properties.

            When you say ‘significantly greater’ I have to disagree, but honestly I’m now anchored by your ’50x’ figure from the previous comment – which might be unfair. But unless the lease was a very long term lease negotiated in previous prosperous times, or is a non-market sham, the lease just isn’t going to add hugely much value. It just can’t unless by accident (why would Walmart gift that value away?). The low sales price of the unoccupied but similar properties limits the value of a currently negotiated lease. (I mean concretely, not abstractly: the ‘value’ of a negotiated lease – however long and under whatever terms – will reflect that alternate supply.)

            But if there’s a long-term existing lease that is crazy relative to what a current negotiated market lease would be then, yes, (I say tentatively, but see the argument) ‘market property value’ should discount the premium from actual sales price due to its fantastic lease’s irreplaceable craziness. But maybe this is semantics. I was saying it make no sense to try to separate the value of a lease from property value. But what I’m really defending is there’s no sense in trying to separate the value of a market lease (whether you actually have one or just failed to win one) from market property value. Not just no sense, no actual meaning.

            Especially if there’s high supply (lots of distressed properties) the new lessee should capture all the financial value of “I’m here now, there’s a lot of costs/inertia in moving” in the lease terms itself. Leaving the lease value to the properties owner relatively in line with the sales price for his vacant competitors.

            TLDR: Walmart moves in to one of the many vacant properties and signs a lease; the landlord tries to sell the property with the intact lease. I claim: the landlord will net close to zero premium relative to similar unoccupied properties, and even more so that there was such surplus supply, Walmart would capture
            all the gains’ from providing a least by means of favorable lesat terms. You can put a price on Walmart’s lease (which is, indeed, a valuable financial instrument) but if you try to subtract it from the subsequent actual sales price of the property (including lease) you will get an absurdity. And furthermore, this is not a theoretical ‘if the world were economically perfect’ claim, but is a 95% accurate description of how the commercial property market works and values things.

    • Murphy says:

      As of 2017, the city had valued that store at $11 million, a number based on what the property had cost the owner to buy back in 2001, plus the added value of renovations over the years

      With one property tax appeal after another, they are compelling small-town assessors and high-court judges to accept the novel argument that their bustling big boxes should be valued like vacant “dark” stores—i.e., the near-worthless properties now peppering America’s shopping plazas.

      ok, so basically the city wants to highball the property value to maximize revenue.

      Meanwhile the companies can point to the real sale prices of roughly equivalent property nearby.

      That actually sounds pretty fair on the part of the companies.

      If you buy a cheap lot and fill it with expensive cars you don’t tax the value of the cars.

      If local authorities want to get more tax they need to prevent the decline in property values so the companies can’t point to similar properties going for a lot less.

    • Doctor Mist says:

      It sounds like these communities have let themselves get too dependent on a few large property tax payers.

      “If you owe the bank a thousand dollars and can’t pay, you have a problem. If you owe the bank a million dollars and can’t pay, the bank has a problem.”

      Asset taxes (of which property taxes are an instance) are problematic in so many ways, not least of which is how illiquid assets often are. I’m surprised how few communities have income taxes.

      • Mark V Anderson says:

        I’m surprised how few communities have income taxes.

        A city or town can only have income taxes (or any kind of taxes for that matter) if the states allow them to. Most states do not allow municipalities to have income tax. Some states do — I think there are hundreds of different city income taxes in Ohio, and quite a few in Kentucky and Michigan. At one point in my life, I was doing income tax returns for a large national retailer, and Ohio was a real pain. Maybe the administrative work for all taxpayers is one reason states rarely allow this. Of course it wouldn’t be so hard if a city merely took the state income tax return and just added a tax rate to that for residents of the city. I think Portland, OR may do something like that.

  24. BBA says:

    Yesterday I visited the Tenement Museum, which explores the history of immigration and the Lower East Side through the restored rooms in an old tenement house. I found it interesting, as a New Yorker whose ancestors lived in similar tenements, and it’s good to see some history being preserved in a rapidly gentrifying neighborhood, but that’s not why I’m talking about it here.

    In 1998 Congress declared the Tenement Museum to be a National Historic Site, an honor accorded to fewer than 100 places nationwide. (Most of them are owned by the National Park Service; the museum remains a private organization but the designation gives it access to some NPS grant money.) I’m not sure if I agree with this. The building is certainly an example of a tenement, and tenements were a significant part of American history, but there’s nothing about this particular building that has historical significance, aside from the fact that it’s still there in about the same shape it was, and most other tenements of its era have been heavily altered to meet modern regulations or torn down. I guess that’s something, but does it really merit the same status as Ford’s Theatre?

    On the other hand, as I look down the list, most of the other National Historic Sites aren’t remotely on the same rank as Ford’s Theatre. There’s a lot of places where some president used to live, and a couple that just got that status because they’re old buildings and the government has no other use for them. Frederick Vanderbilt’s mansion at Hyde Park is an NHS, and there’s nothing special about it. There were lots of Vanderbilts and they had lots of mansions. This just happens to be the one that got donated to the government as a tax write-off. I find myself starting to side with Ruth Abram, the Tenement Museum’s co-founder: “It’s high time that the places of ordinary urban dwellers were afforded the same honor as the places of famous people and/or rural people and/or rich people.”

    Somehow I find myself thinking of Williamsburg, Virginia. Unquestionably it has historical significance as the former capital of the then-colony. But only a few of the “historic” buildings are original, and in particular the Governor’s Palace and Capitol are replicas built in the 1930s. On the other hand, lots of more recent buildings were torn down by the private foundation that runs Colonial Williamsburg, in an effort to “restore” the historic district to an idealized representation of how it might have been during an 80-year period long before anyone involved was born. But now that the reconstructed Colonial Williamsburg has existed for longer than the “real” colonial Williamsburg did, maybe it’s developed some historical significance of its own.

  25. johan_larson says:

    Anyone want to take a stab at defending the claim that there is a meaningful distinction between true art, as opposed to mere entertainment in the same medium?

    It bothers me that the distinction doesn’t seem to be something that could be tested empirically. But my background isn’t in the humanities, so I could be missing quite a lot.

    • Well... says:

      I am not trying to defend the claim, but maybe could something be said for the notion that there are two types of art: that which is successful mainly because it got attention first, and that which is successful mainly because very few have the skill to produce it at a similar level.

      Like, placing a urinal on a pedestal in an art gallery is something lots of people could do, but one guy did it first. (Maybe he had an interesting artistic justification that few would have been creative enough to come up with, but it’s the execution that counts in the end.) Similarly, lots of college film students could have made “The Blair Witch Project”, but its producers happened to have the idea first, and it was wildly successful and influential.

      Alternatively, lots of people over the centuries have used paint to make marks on canvases, but Vermeer’s and Repin’s works are particularly admired because Vermeer and Repin were particularly good at painting.

      This could mean that if an artwork neither got there first nor was incomparably well-executed, it is likely riding on the coattails of art that did one of those two things. Like I said, this isn’t a defense of the claim you proposed, because this doesn’t really have anything to do with “art vs. entertainment” and those aren’t distinct categories as far as I can tell (what we call “entertainment” is a form of art that happens to aim outside the narrow target of “the art world”) but it could at least be a way to think about how an artwork’s success might be related to its quality or vice versa.

      • Radu Floricica says:

        That’s where I went too: originality and quality.

        Also if I were to make a test, I’d base it on how memorable it is, with an eye on changing the viewer (though I don’t see how something could be memorable without changing you even a bit).

        Which would make art relative to the viewer – for me as a kid, Ninja Turtles were art, and *unremarkable cartoon I forgot* was entertainment. That makes sense, I think.

        The followup question is: based on this definition, are commercials art?

    • tossrock says:

      I find Brian Moriarty’s Apology for Robert Ebert to be quite a good stab at just this topic. His extended framing device involving a painting isn’t really my favorite, but the distinction between sublime art and kitsch art is quite useful.

      • Hoopyfreud says:

        You know how you are supposed to feel about sad clowns, James Dean and horses running on a windswept beach.

        H… how am I supposed to feel about sad clowns, James Dean and horses running on a windswept beach? That sounds like a spooky-ass fever dream more than anything else, if I’m going to be honest.

        I don’t like this piece as an apolgia. I love Ebert; I’m still working my way though his Great Movies list, and I’m finding it significantly more rewarding than any other movie list I’ve tried to work from. Based on the essays and reviews of his I’ve read, I’d suggest that this taxonomy of art is one that Ebert himself wouldn’t agree with.

        Here he is in his own article about this business:

        I thought about those works of Art that had moved me most deeply. I found most of them had one thing in common: Through them I was able to learn more about the experiences, thoughts and feelings of other people. My empathy was engaged. I could use such lessons to apply to myself and my relationships with others. They could instruct me about life, love, disease and death, principles and morality, humor and tragedy. They might make my life more deep, full and rewarding.

        Not a bad definition, I thought. But I was unable to say how music or abstract art could perform those functions, and yet they were Art. Even narrative art didn’t qualify, because I hardly look at paintings for their messages. It’s not what it’s about, but how it’s about it. As Archibald MacLeish wrote: A poem should not mean, but be.

        I concluded without a definition that satisfied me. I had to be prepared to agree that gamers can have an experience that, for them, is Art. I don’t know what they can learn about another human being that way, no matter how much they learn about Human Nature. I don’t know if they can be inspired to transcend themselves. Perhaps they can. How can I say? I may be wrong. but if ‘m not willing to play a video game to find that out, I should say so. I have books to read and movies to see. I was a fool for mentioning video games in the first place.

        Like, tell me with a straight face this is Schopenhauerian. That seems like an absolutely ridiculous claim to me. I don’t know what the author thinks Ebert thinks Art is, but I’m pretty sure he’s wrong about it.

        • Lillian says:

          Roger Ebert’s original article that videogames were not art annoyed me, because videogames are pretty obviously art in every meaningful sense. There is no movie that has given me an emotional reaction like i got at the end of Metal Gear Solid 3 when Snake has to finish off The Boss, and there is no movie that possibly could. Only a videogame could make me feel the kind of despair i felt when i realized that i needed to pull the trigger myself. The interactivity adds to its value as art. Though i very much respect Ebert for recognizing that this may be the case, even if he doesn’t see it himself, in his follow-up article. It showed a lot of integrity to do that and admit that he may have been wrong.

          Additionally, i took the time to listen to the lecture that prompted him to declare that videogames can never be art. After i was done, i found myself much more sympathetic to his viewpoint, because the lecture is so bad, it almost convinced me that Ebert was right. Her examples of games as art are all terrible, and her arguments are a bunch of fancy-sounding nonsense and non-sequitours. If this was Ebert’s first exposure to the idea that videogames could be art, i can hardly blame him for reacting with hostility. That lecture is hostility inducing even in people who already agree with its argument.

          • Paul Brinkley says:

            Never played Metal Gear Solid 3, but I get the sense the moment you describe is like the one I felt at the end of Far Cry 2.

            I remember reading Ebert’s take when it first came out, and I remember his primary objection being to the interactivity of videogames.

            Today, I wish he was still around, because I’d love to see his reaction to Black Mirror‘s Bandersnatch episode.

    • Hoopyfreud says:

      I will defend this assertion with the caveat that individual works cannot be sorted into “art” and “not-art.”

      I will go further and affirm that there is no way for an external observer to know with certainty whether a particular person at a particular time in a particular frame of mind experiencing a particular work will experience “art” or “not-art.” Assume these details are known with whatever degree of vagueness you want in order for this claim to make sense to you; my position should still be intelligible if you don’t believe in spooky [literally anything other than reductive materialism].

      I will go further and claim that these categories are nonetheless meaningful.

      Art, for my money, is a thing that produces a transcendental experience. When I experience art, I briefly cease to be the object of my conscious experience. When I experience entertainment, this doesn’t happen; I am principally aware of my experience of something when I am entertained.

    • gdepasamonte says:

      “Proportion of possible enjoyment which is enjoyment of cliche” is not empirically testable, but it’s the closest I can immediately think of to a concise criterion. By cliche I mean the coarse and generic features of the work, which are rarely original. The love of Marge and Homer Simpson across twenty-odd seasons doesn’t make the Simpsons a work of art, because the characters are more-or-less caricatures and their relationship follows a few predictable patterns – of course it is still very good to watch, but we enjoy the predictability. The Jewish Bride is a work of art because there isn’t much generic about it you could enjoy (it’s a picture of a couple being tender, which is usually nice to see, and that’s about it). People enjoy looking at it anyway.

      Of course you already though of a thousand objections, and so have I. Although I believe in the categories I don’t think there is a single sentence or even single page set of criteria which separate them.

    • SamChevre says:

      I would propose C. S. Lewis An Experiment in Criticism, which makes the point that art vs not-art is dependent on the reader, not the artwork.

    • baconbits9 says:

      I would say my heuristic is that the more something resonates across generations the more ‘art’ there is in it, and the more it hits a single generation the more ‘entertainment’ there is.

    • Watchman says:

      Art is something made, rather than something found (a a painting versus a waterfall) but beyond this I don’t think categorisation is possible. And art is not just fine art. Anything done well, including paperwork, is described as an art form. War, an act of creative destruction is both an art form and a subject of artworks. And to the best of my knowledge, no language with an analogous word to art has a recognised hierarchy of art. Art is vast and found where it is found; an attempt to aduce truth in art is doomed to fail as truth is an attempt to circumscribe an acceptable field of discourse in the chaotic churn that is human production.

      Entertainment is something designed to distract, amuse, maybe inform, perhaps even challenge. Its a different category to art, although all entertainment must be included in art as something made, and it cannot therefore be separate from art. But this does allow us to define some art as non-entertaining, even if it is not true.

      Best I can do with a humanities degree. There might not be true art but there is art that is not entertainment…

    • Civilis says:

      If there is a distinction between art and craft, it’s that art is valued for form in addition to function.

      Take an airport bookstore. If you’re in an airport bookstore, it’s not because you came to the bookstore looking for a book, it’s because you’re in an airport and you need something to keep you occupied that will last you at least the length of your plane trip. After you read it, you probably don’t care too much what happens to it. It’s functional entertainment, the product of a craft. My understanding is that at one point a lot of romance novels were churned out in bulk by people paid per how much they produced, often writing under multiple pen names; these are writers as craftsmen. There’s nothing wrong with writing or reading this way; sometimes you want to be entertained for the length of a plane flight and nothing more.

      On the other hand, sometimes an individual work stands out among the others. This is a book you’d be willing to pay extra for, or a book that ends up having value beyond keeping you entertained for your plane flight. You could grab almost any book from the bookstore and be entertained for the flight, but only a handful of them will be ones that keep you thinking about the book long after the flight is over, and thus adding value well over what you thought you were getting in the airport bookstore. These are books as art, written by writers as artists.

      The line between these is always fuzzy; I suspect all but the most hack writers are at least somewhat artists. It also doesn’t help that art that fills its purpose eventually gets imitated as craft. The first romance author that decided ‘let’s add vampires to the usual formula’ would be an artist, but the bazillion or so urban fantasy imitators that have followed in their wake are mostly craftsmen.

    • nadbor says:

      Robin Hanson’s ‘Elephant in the Brain’ gives a good explanation of the function of art that also casts light on the problem of defining art.

      The function of art is to testify to the good qualities of the artist. Hanson gives an example: you find a beautiful sculpture in your friend’s house. It resembles a weirdly contorted seashell. Initially you assume that your friend chiseled it from stone herself. You are very impressed start thinking of your friend as a real artist. But when you look more closely you start suspecting that the sculpture was actually 3D printed. This is much less impressive but still pretty cool. Your perspective immediately shifts from this object being Real Art to a neat gadget. And then your friend reveals that the ‘sculpture’ is an actual seashell that she found on the beach. Now you lose almost all appreciation for the object and think of it as just another pretty shell.

      You can’t make sense of what gets called art unless you recognise that art is not in the thing itself but in how it was created, and specifically what does it say about the creator.

      This is why primitive Egyptian paintings are art but a picture of a sunset taken with your phone is not. It’s not about what is more aesthetically pleasing. It’s about what was difficult to make.

      It’s also not about the emotions it makes you feel. Or rather it is, but only to the extent that making you feel those emotions is an impressive thing to do. I could shoot your dog and make you feel way more emotions than any book or play but it wouldn’t be art because anyone can do it.

      From this perspective, a popular movie or video game is art to the extent that it bears a unique, hard to fake signature of the creator. If you can tell that Avengers 3 was directed by Joss Whedon and that no one else could have done it quite the same way – than to that extent it’s art. If the movie seems like it was made by a committee of skilled but replaceable craftsmen – it’s only entertainment – regardless of overall quality.

      • Hoopyfreud says:

        The function of art is to testify to the good qualities of the artist. Hanson gives an example: you find a beautiful sculpture in your friend’s house. It resembles a weirdly contorted seashell. Initially you assume that your friend chiseled it from stone herself. You are very impressed start thinking of your friend as a real artist. But when you look more closely you start suspecting that the sculpture was actually 3D printed. This is much less impressive but still pretty cool. Your perspective immediately shifts from this object being Real Art to a neat gadget. And then your friend reveals that the ‘sculpture’ is an actual seashell that she found on the beach. Now you lose almost all appreciation for the object and think of it as just another pretty shell.

        Thank you for the perfect quote to pull out when I need to explain why Robin Hanson drives me absolutely bonkers and I hate him.

        • Doctor Mist says:

          I’d be interested if you would elaborate. I found this passage very interesting and evocative.

          • Hoopyfreud says:

            I wrote a long and silly detailed exegesis and the spam filter ate it.

            Short version because I can’t be assed to rewrite it:

            Hanson confuses “being impressed with your friend” with “being impressed with the art,” never addresses the reason why the shell goes from “impressive” to “pretty cool” to “pretty” as the art (he argues) goes from “Real Art” to “gadget” to “object,” completely eliding any (interesting) discussion of aesthetics, and has “you” repeatedly jump to absurd conclusions in order to make his point. With maximum charity I can understand this as Hanson talking about the function of art or the consumption of art (on which points I think he’s still very obviously wrong), but it has nothing to do with people’s experience of art. IMO that’s an egregious omission and typical of Hanson.

      • greenwoodjw says:

        The function of art is to testify to the good qualities of the artist. Hanson gives an example: you find a beautiful sculpture in your friend’s house. It resembles a weirdly contorted seashell.

        I would never see something like that as art.

        • nadbor says:

          Fair. But would the example work for you if you substitute something that you consider art for the seashell? I don’t know your taste, so let’s take a classic – portrait painting. You look at a beautiful portrait, you think it’s art. But on closer inspection it turns out that someone took a picture with their iphone, put a painting-like filter on it and printed it out. Would you still think of it as art? If anyone with good camera can make one in 5 minutes?

          You may be tempted to say that you would never have mistaken a random iphone photo for art in the first place but I think if you were a time traveller from 1800 who doesn’t know how easy it is to make one, you would.

          • greenwoodjw says:

            Here we end up in the artisan/artist divide. Simple mechanical portraits would never really strike me as art in anything other than a technical sense. If something did strike me as art, then it wouldn’t matter whether it was a photo or a painting. I can respect the mechanical skill of a painter doing photorealistic portraits, but “Art” would require some aspect of the scene beyond “person sitting there”.

          • nadbor says:

            Personally, I don’t see anything in Mona Lisa beyond a lady sitting there but let’s not get into that discussion.

            Is there anything that simultaneously
            a) strikes you as ‘Art’
            b) is easy to make

            If yes, then you’re a counter-example to my theory of art.

            If no, then I don’t think it’s a coincidence. The second something becomes easy to make it stops being art. Photorealism in painting was valued only up to the point when actual photography became available. When you could no longer show off by painting
            realistically, painters got props for inventing original styles. It wouldn’t be too hard to imitate for instance an Andy Warhol piece. It would be just as good in any objective sense. It would have the same impact on a viewer provided they haven’t seen the original. But it’s not art precisely because it would be too easy.

            Or take all those stories where people praise an abstract painting done by a chimpanzee thinking it’s a Jackson Pollock (or some other famous abstract painter). Critics of Modern Art take it as proof that it’s not really art. Defenders claim that they would never have been fooled because there is just something about the master’s brushstrokes that the chimp can never capture. One perspective conspicuously missing is that both Jackson Pollock and the chimp are doing great art – it just so happens that this kind of art is really easy to make. No one is saying that because everyone tacitly assumes a *definition* of art as something that requires great skill and especially creativity.

          • Hoopyfreud says:

            @greenwoodjw

            What would you say about something like this?

          • greenwoodjw says:

            Is there anything that simultaneously
            a) strikes you as ‘Art’
            b) is easy to make

            Undertale, with a relative definition of easy. It was a 1-man project that was developed part-time. It’s art, partially because the story could not be told outside of a video game, and partially because of the emotional impact.

            One Punch Man, using ONE’s original (terrible) artwork. It was the artistic quality of the story that drew the amazing illustrator we most associate with it.

            Telling stories is mechanically easy, but IT or The Rainmaker are definitely art.

            What would you say about something like this?

            Definitely art. The composition of the piece helps illustrate the concept, the bright light of the window limited to the area immediately around the philosopher serves as an effective metaphor for the introspection of this kind of meditation, and I could go into the rest of the piece, but I’m lazy. 😛 Basically, this is more than “person sitting”.

          • nadbor says:

            Undertale or OPM may have been relatively easy to execute in terms of man-hours or in terms of how much it would cost to hire someone who can replicate it but that’s not the dimension of difficulty I was talking about. The relevant dimension is ‘how hard it was to come up with the concept’ or ‘how impressed I am with the author for having done it’. On this dimension both Undertale and OPM score highly (probably, I defer to your judgement here).

            Now, if someone were to rip off OPM and tell the exact same story again in the same medium, only draw it again and change the details – that would be both easy to execute and easy to come up with. And most people wouldn’t call it art anymore even though it may be just as good as the original in every way.

            And OPM has nothing on John Cage’s 4:33 as ‘easy to execute’ art goes. It’s not art because it’s hard to sit in front of the piano and do nothing. It’s art because it took a great artist to pull it off. There are countless other examples like that.

            *

          • AG says:

            If your argument is that art is dependent on its originating context, then it fails to account for Death of the Author. There are plenty of cases where the audience imbues their own personal context to a piece. It artistically resonates with said audience for reasons independent of the artist, sometimes even counter to the artist’s intentions. What is intended to be meaningless becomes meaningful.

            This is especially apparent in music.

          • greenwoodjw says:

            The relevant dimension is ‘how hard it was to come up with the concept’ or ‘how impressed I am with the author for having done it’.

            OPM is “The hero is so much more powerful than everyone else he always wins with the first blow. That’s boring to him.”

            Concept is pretty easy.

            Undertale is basically “took the ‘murder hobos’ joke in D&D seriously”.

            Also an easy concept.

            Your original question was “easy to make“, not easy to come up with.

    • LesHapablap says:

      As usual Calvin and Hobbes has the answer:
      High Art vs. Low Art

    • Le Maistre Chat says:

      Anyone want to take a stab at defending the claim that there is a meaningful distinction between true art, as opposed to mere entertainment in the same medium?

      Absolutely

      It bothers me that the distinction doesn’t seem to be something that could be tested empirically.

      So much the worse for empiricism.
      I say it’s indefensible that One Direction isn’t qualitatively inferior to JS Bach, or the latest fantasy novel qualitatively inferior to Ariosto. Superior art is about things and makes you think.

      • Hoopyfreud says:

        Superior art is about things and makes you think.

        To which I’ll give Ebert’s reply from above:

        It’s not what it’s about, but how it’s about it. As Archibald MacLeish wrote: A poem should not mean, but be.

        Which I’ll amend to say, art isn’t defined by what it’s like, but by what it does.

        Take three arts and rank them from most to least superior. I guarantee someone will disagree.

        • Protagoras says:

          Evaluating art is extremely difficult. If you pick three good examples, as you have done, that difficulty ensures that people will not reliably come to the same conclusion about exactly how good. But if you pick examples where the difference is more dramatic, the situation becomes clearer. Perhaps Le Maistre Chat could provide an example of a characteristically bad One Direction song (I’m not familiar with their music myself); it wouldn’t surprise me if its inferiority to all three of your examples was blazingly obvious.

          • Hoopyfreud says:

            I’m going to be honest here; I expect at least one of these to do absolutely nothing for most people (I’m slightly ashamed to admit that for me it’s the Bach). Which it is, I expect to differ. I also think that anyone who really hates any of these (except the Bach) could make a case that it’s Not Art. I think that’s an interesting and important phenomenon.

          • Le Maistre Chat says:

            Perhaps Le Maistre Chat could provide an example of a characteristically bad One Direction song (I’m not familiar with their music myself); it wouldn’t surprise me if its inferiority to all three of your examples was blazingly obvious.

            I am tempted to link “Little Things”, whose backhanded compliments to its generic female listener perfectly encapsulates the mercenary psychology of pop music, but even its co-writer Ed Sheeran admits its one of their worst, “a castoff track.”
            So here’s “Something Great”.

        • albatross11 says:

          I think the way people often try to gauge this is to find existing compilations or lists of great artists / philosophers / scientists / etc., and combine them. In _Human Accomplishment_, Charles Murray used some algorithm based on (I think) the number of words devoted to each artist in a set of standard art reference books/text books, with the books chosen based on some other criterion before the ratings were started.

          But all these are about a consensus among experts on something that sure seems deeply subjective. There was a time when Bach and Shakespeare weren’t considered all that impressive, but now they’re considered giants.

      • What does qualitatively inferior even mean here? Look at literature. It’s “obvious” that Jane Eyre is better than some hard science fiction that is technically complex with shallow characters. Why does having more developed characters make something more “artistic” than having a more developed world?

        • Protagoras says:

          “The Eye of Argon” is qualitatively inferior to either Jane Eyre or Ringworld. This is enough to establish that it is possible to make meaningful and true claims about what is superior or inferior. If we cannot make equally clear and uncontroversial claims about how Jane Eyre and Ringworld compare to one another, that may have interesting consequences. Many things are difficult to measure consistently and precisely, for many different reasons; some of those reasons seem relevant to aesthetics specifically. But we needn’t go into the details of why measuring quality of art is difficult and unreliable. For present purposes, the key point is just that it must be difficult and unreliable, rather than impossible. If it were impossible, we wouldn’t know that “The Eye of Argon” is inferior.

          • albatross11 says:

            I think a really basic problem here is that you’re measuring on multiple axes, and we don’t agree on a way to reduce that to a one-dimensional score. I mean, _Pride and Prejudice_ has amazing dialogue and the characters are very well-drawn, but it just doesn’t have much to say about the inherent limits of human technology or the singularity. OTOH, _A Deepness in the Sky_ has a lot to say about those things, but the dialogue isn’t quite so amazing. Which one you prefer depends on what you value, and anyone telling you your choice is wrong when you just disagree on what you like is kind-of silly.

            At some point, you get to these hopeless comparisons. Which is better, _Iron Man_ or _Debbie Does Dallas_? Well, it depends on whether you want violence porn or just the ordinary kind of porn with a lot of flesh in it. Debbie will probably be more welcome at a bachelor party, Iron Man will be more welcome at a 12-year-old’s sleepover party. (At least by the parents.) You might as well ask which is better, the fundamental theorem of calculus or a ham sandwich?

          • Jiro says:

            If we cannot make equally clear and uncontroversial claims about how Jane Eyre and Ringworld compare to one another, that may have interesting consequences.

            Art quality may be a partially ordered set. The fact that you can compare the Eye of Argon to both Ringworld and Jane Eyre may not mean that you can compare Ringworld to Jane Eyre.

          • I’m never going to be convinced by the argument “A is better than B because duh”. Just because something isn’t quantitative, doesn’t mean you can simply take an opinion, add the word “qualitative” and then transform it in to an objective fact. Otherwise, I could do that with anything.

        • Le Maistre Chat says:

          It’s “obvious” that Jane Eyre is better than some hard science fiction that is technically complex with shallow characters. Why does having more developed characters make something more “artistic” than having a more developed world?

          The value of a complexly-developed fake world is… debatable. Unless it has something powerful to say about our world, like the “worldbuilding” is actually an imaginary future of our world with technology the author sincerely believes in (e.g. Singularity AI and not FTL), isn’t it just autistic self-indulgence?
          Whereas I could defend the superiority of complex characters over shallow ones on multiple axes: shallow characters are predictable rather than surprising you, diving into full realistic psychology improves your empathy, etc.

          • I like books that have well developed characters but I don’t think that extensive world building is any less self indulgent than writing a stream of consciousness book highly focused on the characters thoughts. And what does self indulgent as a criticism even mean here? And why is it so bad?

            I could easily turn this around. The Martian, which is mere “genre fiction”, is filled with extensive research about the workings of a trip to Mars, how one would survive it and then how one would get back. Why is that automatically less valuable than the well-worn romance novel?

      • AG says:

        The issue is that different things make different people think, and different people think that different things are about things.
        Some people find Bach meaningless for having no lyrics, just some soundwave patterns, whereas One Direction provides an endless fount upon which to analyze adolescence and the ways in which the idol industry operates, or what makes a pop hook stick.

        There may be a meaningful distinction between true art and mere entertainment, but only for each individual. It is impossible to find a true generalization for any population, only in increasingly limited samples.
        Art house films are only sound and fury to some, but fascinating allegory to others, while blockbuster superhero films are only sound and fury to some, but fascinating allegory to others. Trash/treasure/one/another

    • JPNunez says:

      Art is so poorly defined that this is impossible with any degree of rigurosity. Compare with the-usually-contrasted-term, Science, and you have the basic scientific method framework, so you can more easily check for stuff to see if they qualify as science -although grey areas will exist-.

      Thus I don’t think there can be any useful distinction, and gotta go with art is in the eye of the beholder. If a person says hey, Big Bang Theory is True Art, well, write me a paragraph explaining it and maybe we can bridge this chasm of understanding cause I disagree.

  26. Lurker says:

    I’ve taken SSRIs twice and they made me hypomanic both times. About 3% of patients have this experience (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3387568/ small N) but we call it “bipolar III” instead of “dangerous side effect.” Is this 3% not enough to explain the entire (and small) treatment effect of SSRIs? My PHQ-9 would have gone from 15 to “this is the best day of my life” within days…

    I’m surprised nobody is discussing what seems to be the Occam’s razor of SSRI outcome studies.

    • Douglas Knight says:

      I think it’s called bipolar III because other people with bipolar experience this effect. It’s probably a provisional assertion that this is a good category, that it predicts that you’ll exhibit other symptoms or, eg, respond to lithium. That sounds pretty reasonable to me, but that doesn’t mean that anyone followed up and tested that this provisional cluster of people called bipolar is a useful category.

      Most people don’t respond to SSRIs within days.

      Scott touched on the topic of some people responding to SSRIs here. My impression is distributional data is not available. Is it possible that there would be value in looking at the distribution and the people who do have access to it haven’t really looked at it? Maybe, it seems implausible that it could be anywhere as extreme as you propose.

      What do you mean by 3% “enough to explain” the effect of SSRIs? That 97% have no response and 3% have a big response? No, that is not plausible. The average effect size is ~1/3 of a standard deviation. For that to be caused by 3% of the population, those people would have to exhibit 10 standard deviations of effect. Is there even that much room on PHQ-9 or HAM-D? Probably, but you only scored 15, wasting half of the scale. You probably only exhibited 5 standard deviations of effect.

      Conceivably 3% get hypomania, are taken off of the drug, and are classified as no improvement, 10% exhibit “hypo-hypo-mania,” and the rest have no effect. If so, that should be obvious from the data and it would be pathetic (but possible) if the people with the data haven’t figured it out. But that’s not what it looks like to practicing psychiatrists, either.
      In that post, Scott proposes that 1/3 of patients experience an effect of 1 standard deviation, while the other 2/3 have no effect. If true, that would be valuable to know, but it would be hard to see underneath the noise, unlike the 3%.

      Or are you just suggesting that hypomania is the opposite of depression and antidepressants work by causing hypomania? I’m not sure that really means anything.

  27. Content Warning: Meta-futurology

    Over the past few years, I’ve noticed mainstream popular futurology shifting from “the future will be just like the present, except hotter, more crowded, and less equal” to “climate change will destroy civilization within our lifetimes.” This got me thinking: what’s next for futurology? I envision three futures:

    1) The “boring” future. Climate apocalypse predictions start coming true, and we get new ones extrapolating from the old. If Bangladesh really is underwater by 2030, there’s no reason to change tactics.

    2) The “exciting” future. Nothing eschatological happens on the climate front, and the clickbait outfits need a new topic. The old standby is, of course, sex. So now we need to envision the future of sex. I think sexbots are a pipe dream, and there will be no meaningful progress on them in the near future. On the other hand, I think AI-powered “phone sex” could really take off. (Maybe it already exists and I just haven’t heard about it?) Since this is purely a software problem, the relevant tech evangelists can make whatever claims they want without anyone doing something awkward like asking to see the sexbot. This could be a great catalyst for a highly profitable media frenzy of a new moral panic, reporting on the moral panic, reporting on the reporting on the moral panic, etc.

    3) The “WTF” future. Some event happens that fundamentally changes the economics of online publishing, and my entire analysis is fundamentally misguided.

    Maximizing the ad revenue : effort ratio for the thinkpiece writer of 2030 is left as an exercise for the reader.

    • The Nybbler says:

      Bet against the boring future.

      • broblawsky says:

        Bet against the boring future.

        Based on what aspects of that article?

      • fion says:

        I think you’re trying to suggest that the fact people were worried about climate change 30 years ago and we’re not yet dead means that they shouldn’t have been worried?

        What, specifically, do you think that article has been proved wrong about? I accept that it’s a bit sensationalist, and it’s not very well written, but they’ve not been proved wrong yet based on what I gleaned from a quick skim. Helpfully they give predictions of temperature rise in the next 30 years and it was written 30 years ago. They say 1 to 7 degrees. Temperatures have increased by 1 degree in the last 30 years. They say sea levels will rise by 3 feet. We still believe that. Since the article they’ve risen about 10cm already and they’re currently projected to rise 0.3 to 1.2 metres by 2100.

        And some people do believe we’ve missed our window to stop catastrophic global warming. It’ll take decades to kill us, but the CO2 is there and the warming will continue happening even if we cut emissions to zero today.

        Of course we’ve learned a lot about the climate in the last 30 years, and I’m sure much of the science that went into that article has been updated. Also I accept that it is unhelpfully sensationalist and probably politically biased. But the core of that 30 year-old article seems to me to be coming horrifyingly true.

        • The Nybbler says:

          I think you’re trying to suggest that the fact people were worried about climate change 30 years ago and we’re not yet dead means that they shouldn’t have been worried?

          I’m saying they were crying wolf. Here we are, 30 years later, temperatures have risen considerably less than that “conservative” 1-7 degrees — more like 0.6 degrees.

          But the core of that 30 year-old article seems to me to be coming horrifyingly true.

          Egypt’s agricultural production seems fine, so that’s wrong. No dust bowl in the US or Canada. Bangladesh keeps getting flooded… like it was in 1988 and 1987 and 1951 and 1892, so this is like predicting the sun will rise in the East. One non-obvious thing they did get right is there’s bumper crops of wheat in what used to be the Soviet Union. So either “yay climate change” or “boo Communism”, I guess.

        • And some people do believe we’ve missed our window to stop catastrophic global warming. It’ll take decades to kill us, but the CO2 is there and the warming will continue happening even if we cut emissions to zero today.

          I can’t tell if your “kill us” is merely careless hyperbole or is intended literally. If you look at actual numbers from the IPCC or Nordhaus, they are talking about effects that reduce human welfare by the equivalent of reducing GNP by a few years of economic growth, not about effects that wipe out the human race.

          On the general issue of (IPCC) projections, I have an old blog post that compared the projections of the past reports with what actually happened. They consistently projected high. For two of the first three reports, what happened (as of 2014 when I wrote the post) was below their 95% range. I haven’t updated the figures—anyone interested is welcome to do so.

          Temperatures have increased by 1 degree in the last 30 years.

          About half a degree, actually, by the NASA data.

          And 1 to 7 degrees was supposed to be “The most conservative scientific estimate.”

          The only reasons the story is not complete nonsense ex post are the vagueness about when things are supposed to happen and the liberal use of “could.”

    • Faza (TCM) says:

      I recommend digging up a copy of Lem’s Futurological Congress and seeing how futurology, as seen by a sympathetic, albeit critical, eye, from the early 70’s holds up to today. It’s satire – and hence: caricature – but the interesting thing is that pretty much none of the big concerns of the era have lasted until today (barring peak resources, and even that’s taken a beating).

      Population concerns have all but been reversed: in Europe the concern is population decline. Exactly one country took the concerns seriously enough to do something about it in the interim – China – and they’ve since backtracked (but the problems it created remain).

      My take on predicting the future is: don’t do it, unless you’re pretty damn sure you understand the entire system under observation. Nobody does.

      However, you don’t get money and fame in futurology by keeping your predictions conservative, so I expect doomsday scenarios and utopias to proliferate, ‘coz that’s what sells books and speaking engagements.

  28. hash872 says:

    What cultural factors (or, I guess, other factors) influence a nation’s military unit effectiveness? Particularly the toughness & determination of its fighting men against adversity. I’m watching a Netflix documentary on WW2 at the moment, and am struck by the almost suicidal ability of the Japanese units to never surrender and fight down to the last man. I think…. other nations (or ethnic groups or whatever) don’t have that level of determination.

    What makes up that toughness? Worth noting that both the Germans and the Japanese, effective military cultures, have strong cultural elements of blind obedience to authority. Is that what drives fanatical determination on the battlefield?

    (Semi-related question- has the US ever experienced extremely high casualty rates in warfare- and persevered through? Maybe just in the Civil War? I’m somewhat influenced by the European/Russian view that Americans never experienced the worst fighting in WW2, and merely came in at opportune times with overwhelming manpower, as our population is so much higher than any other individual nation in the war. Would find quantitative arguments around ‘x % of casualties’ comparisons more persuasive than subjective arguments of ‘of course this or that war was tough’)

    • Tetrahedrex says:

      Japanese soldiers were driven by more than just determination. From what I’ve read, speed use was almost ubiquitous in the Japanese military. I’d say that is probably enough to overwhelm the effects of any cultural factors, but the extreme levels of punishment and discipline (especially punishment for retreat) probably also made them feel like they had nothing to lose.

      • abystander says:

        Germans used speed. I hadn’t heard the Japanese used speed. Anyway they often fought long after the possible supply of speed was cut off, even years after the official end of the war. So cultural factors were of major importance.

        • Protagoras says:

          Quick internet research results: yes, IJA used speed. Of course, pretty much everybody did in WWII (though it is little talked about). It doesn’t seem like their level of use was exceptionally high, and it may have been lower than for other participants in the war (as part of the general pattern of the Japanese having trouble keeping their forces supplied).

        • Nornagest says:

          All the major militaries of the time used speed if they had the industrial base to make it (and the Japanese definitely did). Wiki claims that it was issued to USAF pilots as late as the first Gulf War, on a voluntary basis.

    • Yair says:

      I don’t live in Israel anymore but many decades ago I served in the IDF in a tank crew. Army service was compulsory and everyone had to do 3-4 weeks reserve duty every year, so when the History professors did their reserve duty they sent them to teach us, regular soldiers, some history.

      I’m mentioning this because one of the lectures was about exactly this question, it was a fascinating lecture and I still remember it 3 decades later. The lecturer told us:

      – soldiers that voluntarily fight to defend their home and family fight better than conscripts sent to fight in a foreign land.
      – If you go into battle to die, you will find a way to die. The IDF cannot afford you to “die for your country” it wants you to live to fight another day.
      – Fundamentalism makes bad soldiers, hatred makes bad soldiers. Good soldiers make their decisions either on instinct or on logic/reason, hatred clouds your judgment and therefore is a problem.
      – Discipline is important but too strict a hierarchy is a problem, obey your commands as long as they are not against the law, but if you think something could be done better, then find the right time and the right way to tell your commander.

      Now I am not saying that these rules are right or apply always, I just thought that you might find interesting what one lecturer told a bunch of tankers 30 years ago.

      • sentientbeings says:

        Discipline is important but too strict a hierarchy is a problem, obey your commands as long as they are not against the law, but if you think something could be done better, then find the right time and the right way to tell your commander.

        Now I am not saying that these rules are right or apply always, I just thought that you might find interesting what one lecturer told a bunch of tankers 30 years ago.

        As it happens, I heard something about hierarchy in the IDF just two weeks ago. An Israeli professor was giving a talk about some aspects of Israeli culture and he mentioned that the IDF was not as hierarchical as one might expect of a military. I think he connected it (in part) to the common circumstance of role reversal from the hierarchies of civilian life among soldiers who already knew each other.

        • Yair says:

          That makes a lot of sense, especially during reserve duty.

          Everyone has to do 3-4 weeks a year reserve (health allowing) so it is not unusual for someone that in civilian life has quite a prestigious job to be lower rank in the army than someone with a less prestigious job (or less money, or no job) and therefore have to obey orders.

          I think it is actually healthy for a society, that for a few weeks in the year people are judged by a different scale and different skills to what they are judged on normal day to day life.

          There is an old Israeli song from an old musical called Kazablan, in which an immigrant who is working class and not very educated says: “when the battle is burning and the squad is not moving, I go forward first and everyone then KNOWS who has the most honour/ deserves the most respect.”

      • Civilis says:

        There’s a disconnect between ‘toughness and determination’ and ‘effectiveness’, in that things that increase ‘toughness and determination’ can often decrease effectiveness. I generally think you have a good response to the question, but your comment was a good place to make some WW2 related additions.

        soldiers that voluntarily fight to defend their home and family fight better than conscripts sent to fight in a foreign land.

        Soldiers that believe their home or nation is at risk do better regardless of where they fight, and it’s up to the government or society to persuade them that. In a ‘Total War’ scenario, even the conscripts or draftees are surrounded by an environment where everyone seems to be sacrificing for the war effort. World War 2 American draftees were far more motivated on average than their Vietnam War counterparts.

        If you go into battle to die, you will find a way to die.

        There’s a reason suicide charges were ineffective. On the flip side, fear is a powerful motivator. There’s nothing demonstrating that than seeing some of the best surviving units of the German army and even the SS in encircled Berlin desperately fighting tooth and nail to fight their way out entirely so they could surrender to the Western allies rather than the Soviets.

        The IDF cannot afford you to “die for your country” it wants you to live to fight another day.

        While all the ‘not one step back’ orders might signal ‘toughness and determination’, they were almost always detrimental to army effectiveness. For every Bastogne where a determined last ditch stand at all cost was able to effectively stop an offensive, there were dozens of isolated pockets that were surrounded after an offensive that were neatly wrapped up and got units wiped out for no gain.

        Fundamentalism makes bad soldiers, hatred makes bad soldiers. Good soldiers make their decisions either on instinct or on logic/reason, hatred clouds your judgment and therefore is a problem.

        There are times when emotion (especially of the nearly suicidal bravery type) is the right course of action. For a non-World War II example, Chamberlain’s charge at Little Round Top. How effective it is depends on situation and culture. Ironically, the Nisei units of the American army tended to get it right far more often than the Japanese army.

        Discipline is important but too strict a hierarchy is a problem, obey your commands as long as they are not against the law, but if you think something could be done better, then find the right time and the right way to tell your commander.

        If anything, rather than blind obedience to authority, the German army was more effective for having a military culture that placed emphasis on what’s usually described as ‘mission tactics’ (Auftragstaktik), where individual commanders were assigned an objective and granted flexibility in carrying out their goals. It was in the later days of the war when due to increasing casualties and high command usurped by political leaders they started being more strict with orders that the German army lost effectiveness.

        • bean says:

          It was in the later days of the war when due to increasing casualties and high command usurped by political leaders they started being more strict with orders that the German army lost effectiveness.

          This seems like a dramatic oversimplification. At the strategic level, meddling by the high command was endemic throughout the war. The Battle of France was fought on Hitler’s hunch against the advice of his generals. He won the gamble, but it set a pattern for the entire rest of the war. Lower down, the Germans had better leadership pretty much throughout. Man-for-man, they were better than their enemies, and they maintained high standards of leadership until the last few months. The big problem was that they ran out of men and out of equipment, while their enemies didn’t.

          • Le Maistre Chat says:

            This seems like a dramatic oversimplification. At the strategic level, meddling by the high command was endemic throughout the war. The Battle of France was fought on Hitler’s hunch against the advice of his generals. He won the gamble, but it set a pattern for the entire rest of the war.

            This. The post-war meme by non-Nazi German generals of “Oooh that meddling Hitler!” is not a narrative that fits the facts. Hitler was actually a brilliant amateur, until he wasn’t (Barbarossa, I suppose). France and Britain could have destroyed Germany by calling any of Hitler’s pre-war bluffs, and his generals had the evidence to show it. Yet Hitler kept winning with his diplomatic bluffs, then in Poland, then with the Battle of France. As you say, this set a pattern that he arrogantly locked into until the Red Army was practically in Berlin.

          • Civilis says:

            It’s certainly something of a simplification, and meddling was certainly a factor on the strategic level, but from what I’ve read you don’t start seeing it much on a tactical level (where this discussion comes in) until Stalingrad.

            It’s not like the entire army changes culture at once. In Normandy, for example, you had units with veteran history like the 352nd division backing Omaha Beach, you had sectors with less experienced garrison troops with little if any long-time veterans, and you had sectors defended by Osttruppen (who, being former PoWs, never had the culture to begin with). In other areas, you get military units formed from air force or naval personnel, who again lack the training, culture and veteran cadre the long-running army units had.

            At some level, it’s a chicken-and-egg problem. It would be unthinkable for the German Army of 1941 to have military police with orders to shoot anyone falling back because it was more or less unthinkable for the German Army of 1941 to be routed that badly; that sort of enforced discipline was something for the Soviets.

        • Yair says:

          That’s very interesting, Civilis. Thank you.

        • Civilis says:

          Adjacent to this, there’s a comment that went somewhat viral on Twitter repeating the recollections of a French soldier that had fought overseas alongside American forces about how tough the American troops he fought beside were. One of his specific comments was that in his experience, most other countries troops would hunker down and dig in if their officer was killed or incapacitated; what made American troops so effective was that if they lost their officer, instead of digging in they would counterattack aggressively to eliminate the threat. An American veteran responded that that was largely because in American units the senior NCOs were heavily involved in the unit leadership and operational planning process, giving them good knowledge of what the unit needed to be doing and the skills to carry it out.

          • bean says:

            This sounds like the sort of thing that’s going to vary greatly based on what unit you’re in. In a unit with good NCOs, then yes, the loss of an officer isn’t going to be crippling. The NCOs will take over and keep moving. The US had a lot of good NCOs, but also a lot of bad ones. This is also true of Germany and Britain, with the Germans doing a somewhat better job of generating good NCOs than most countries. I’d suspect this was an area where the Soviets were particularly bad. Can’t speak to Japan.

          • dndnrsn says:

            @bean

            WWII or present (or Cold War era for the Soviets)? I have read that during the Cold War the Soviets, depending on conscripts, relied relatively more on junior officers to provide what in the US would be done by professional NCOs. WWII, I believe they had a lack of radios through most of the war. I seem to recall that also through the Cold War the Soviets had fewer radios than the better NATO powers.

          • bean says:

            I know about the Cold War problems. Can’t speak to WWII in particular, but I really doubt the problem just appeared in 1950. On the other hand, you also had the risk of ending up as an Il-2 rear gunner if you weren’t aggressive, so who knows how it all cancelled out.

          • johan_larson says:

            It seems strange that a squad would just fall apart and become completely ineffective if it lost its leaders, whether officers or NCOs, and stranger still that this would be the expected behavior. Maybe if they were completely green, sure. But you’d think decent infantry would have a bit of resiliency and resourcefulness, so they could keep going when the bosses take dirt-naps, though with somewhat degraded performance. Anything else sounds terribly fragile, and ill-designed for an environment where anyone can die at any time.

            I seem to recall Squad Leader, the game, represented this by having squads that broke do nothing but flee, and typically only officers could make them stop fleeing. But a few militaries had Self-Rallying, which allowed squads to stop fleeing and un-break themselves without help from officers.

          • cassander says:

            I believe that this is the article in question. I think it’s less a question of NCO quality and what happens when officers are killed and more a question of basic military culture, a US attitude of “When in doubt, attack!” that doesn’t seem to exist elsewhere.

          • bean says:

            Ohh. Fought with American troops in Afghanistan. For some reason, I assumed this was a WWII thing. That makes a bit more sense.

            I have lots of potential explanations for this, none of them particularly great. Yes, the US military has learned a lot of lessons over the past 3-4 decades, and figured out how to keep them in mind even when we’re not using them. That’s helped a lot. As has more training and better raw material to work with. I’m not sure about the “when in doubt, attack” thing. It’s quite plausible that this is a cultural thing (the US military is heavily borderer, a group absent from most of our European partners) or it could be an airborne thing. Or maybe I just haven’t been paying much attention to ground fighting lately.

    • Noah says:

      I don’t think the claim that the US had much higher population than everyone else holds up.

      Granted, these population counts are weird in some ways–the Japan population includes much more than the Japanese, the Soviet population is higher than US, but much of that was under German control for much of the war, etc. What the US did have was overwhelming economic might.

    • Aapje says:

      What makes up that toughness?

      A key part is a specific sense of honor, which is culturally enforced.

      One reason why Western POWs were treated badly by the Japanese was because their sense of honor required a maximum effort for the cause. Allowed oneself to be captured goes against that.

    • RalMirrorAd says:

      The USSR had its fair share of ideologically committed soldiers, but was far more heterogeneous than Japan so there was probably much more variety in the kind of soldier.

      What makes Japan Unique I think would be the ‘Honor Culture’, the idea of going into battle expecting to die which had past historical precedent.

    • bean says:

      Semi-related question- has the US ever experienced extremely high casualty rates in warfare- and persevered through?

      Yes. Look at any infantry unit that saw more than a few months of combat, and you’ll see casualty rates that are frankly staggering, particularly when you consider how much of an American division was noncombatants. During the fighting in Normandy, it was said that the commander of one division (I think it was the 28th) had three divisions under him: one in the field, one in the hospital, and one in the cemetery. But this was basically true for all of the veteran infantry divisions in both theaters. There wasn’t the large-scale brutality of the Russian Front, but on the small-unit level, American troops did well. The only numbers I have to hand are of the first day at Tarawa, when the Marines suffered 1,500 casualties out of 5,000 men who got ashore. I can look at more when I get home.

      (I’ll agree that the Japanese at least were more courageous on a man-to-man level, but that doens’t necessarily win wars. Logistics does, provided your troops are good enough. And ours were.)

  29. Harry Maurice Johnston says:

    This puzzled me, specifically the second to last sentence:

    Unfortunately, Holder ran out the clock in district court, the Obama administration came to an end without ever complying with the subpoenas, and the American public never got an answer.

    Why would the end of the Obama administration preclude the release of the documents? Surely the DOJ doesn’t shred everything every time there’s a new President in charge?

  30. rubberduck says:

    Continuing the recent trend of sports threads: Anyone here interested in rock-climbing?

    The strongest team by far competing in the International Federation of Sport Climbing competitions, in lead and bouldering (but not speed), is Japan. This is unsurprising: mountainous country, famous work ethic, rich. And the second-strongest team in the IFSC in the same categories is… Slovenia?

    Further down the ranking are the countries that one would expect (France, Austria, Russia, etc), but they are all far behind Slovenia and Japan. But why is Slovenia so strong? I do not understand why it beats out other rich, mountainous European countries like France and Switzerland, which have far bigger populations. Anyone have any idea? My own guess is that any given country only needs a handful of really amazing climbers to dominate the competition and top the rankings, since climbing is a solo sport, but I really don’t know much about the country.

    • Fossegrimen says:

      Slovenia seems to have a strange all or nothing approach to sports; They dominate in ultra endurance cycling, ski jumping and apparently rock climbing but are totally invisible in most other sports. This has puzzled me too.

      • AlphaGamma says:

        They’re decent at rowing (Iztok Čop competed at 6 Olympics and won 4 medals including a gold) though perhaps not as successful as Croatia.

    • Watchman says:

      Is sports climbing taught in Slovene schools? There certainly seems to be a strong climbing tradition in Slovenia, which may have been part of the small state’s self-definition in Tito’s Yugoslavia, in a way which I don’t think I’ve seen in other nations other than locally.

    • fion says:

      I do it avidly, but I don’t really keep up with what’s going on in comps. I knew Japan was strong, but I didn’t know Slovenia was. I have two (not very well thought-through) comments:

      First, climbing is relatively young as a competitive sport. It seems to me that most countries haven’t yet come close to optimising in the way that they have for football (soccer), swimming or gymnastics or whatever. Maybe ten years from now the best countries in climbing comps will be populous, rich countries like for most other sports, but right now it’s an under-explored niche that allows unusual candidates like Slovenia to excel.

      Second, I want to push back a little bit on your comments about mountainous countries. It certainly used to be true that climbers were born in the mountains, but I think most of these young competition climbers started indoors. I’m not sure if the abundance or quality of indoor walls is influenced much by the geography of the country. In short I think you may be overestimating how much the natural rock of the athletes’ home countries affects their competition performance.

  31. Nabil ad Dajjal says:

    So we have a lot of historically knowledgeable people here and I’d like to pick some of your brains about something that I’ve never really understood. How did codes of hospitality like the Greek Xenia actually work in practice?

    In the era before electronic communication, if someone stranger blew in from parts unknown there would be no way to check whether or not he was a thief or a rapist or a murderer before you let him spend the night in your house. And if he did steal from or rape or murder a member of your household there’s a good chance that you’d never be able to get your hands on him if he could get out of your house alive. So a pledge to house and not harm strangers seems downright suicidal, especially given that pre-modern societies are supposed to have been much more violent.

    Obviously there were a lot of morality tales urging people not to turn away strangers, but I can’t understand why anyone would listen to them. A disguised God or king is almost certainly not knocking on your door in the middle of the night but there’s an uncomfortably high chance that a lunatic is. It seems like something else had to be going on.

    • bullseye says:

      Even if the guest isn’t someone important, turning him away could offend his family, tribe, or nation and provoke a violent response. My guess would that that risk would outweigh the risk of letting in a psycho who’d attack unprovoked.

    • Erusian says:

      Hospitality did not require you to let everyone in. It required you to treat those you did according to certain rules. If someone showed up at your door with a bloody knife and gibbered about murdering your children then asked for hospitality, it was not a violation of hospitality to turn them away.

      In fact, there is a common scene in early Greek works where the residents debate whether to extend hospitality. Probably the most famous is between Eteoneus and Menelaus in the Odyssey. Homer subverts the expectation there by having Menelaus immediately demand to take them in on principle. More normally there would either be a tension building debate or a long description of the guests.

      This issue of guest respectability is a major point in the Odyssey. A person has to be worthy of being a guest, which was bound up in all kinds of manners, religious status, and class status. This is why we transition from the suitors to Telemachus in the Odyssey. Menelaus’s comments on how Telemachus is a great guest are meant to make a point about hostship, guesthood, and hospitality. As is Helen’s presence and the fact he’s in a place where another prince committed a gross violation of hospitality.

      And on a cynical level, law codes show that the principle was often not honored. There are a lot of laws protecting travelers from murderous locals and providing for the hunting down of travelers who have committed crimes. This probably points to the hostile nature these relationships could take. Even the Trojan War starts with a violation of hospitality.

      How did they actually work in practice? Well, generally a group of people would be traveling for some reason or another. Traveling alone was rare except for very small distances. They would be stopped at some kind of border. They would then make a request for hospitality, which was probably a bit of relief as another common possibility was a raid. The hosts would either grant or deny it. If they denied it they would hope the people would just go elsewhere but epic poetry shows denying hospitality could lead to a raid. If they agreed, both parties were considered obliged by xenia to treat each other well. They would mingle, eat together, and exchange gifts.

      Interestingly, this was a hereditary and perpetual obligation. People would visit back and forth as a way to cement relationships. They were also considered to have obligations not to harm each other and act in a manner bound by xenia even when they were not guests. While it is something of a convenient excuse not to fight, there’s a scene in the Illiad where Diomedes and Glaucus realize their parents used to be guests in each other’s homes. They say this means they cannot fight each other due to Xenia and engage in the symbolic gift exchange.

      It’s also how the proxenos worked, which was a Greek diplomatic tradition that lasted through the whole of antiquity. Rather than sending one of their own citizens, Greek city states would deputize foreigners in other cities to act in their interests and bind them to the state by hospitality. This was entirely a public affair. The proxenos was still loyal to the city he came from but would visit the city he was engaged as a poxenos for, and would host them in turn in his city. And likewise, the post was often hereditary.

    • Watchman says:

      Hospitality didn’t mean access to everything. I don’t know the Greek examples well, but the post-Roman norm was that hospitality could be given outside the hall, so beggars would be fed outside the door, or in the hall or even in the hall owner’s own chamber. Beowulf was hosted by Hrothgar (in what was a poetic meditation on expectations) in the hall, as a high-ranking ally. If your king visited he presumably got the sleeping quarters judging by the fact one king of the West Saxons, Cynewulf, seems to have been surprised in the bedroom whilst his followers were in the hall (it didn’t work out well for him). As with everything hospitality was linked to status: it’s in this light we have to see royal edicts that foreign traders were to be valued as royal retainers, to provide them status and therefore access to hospitality in territories where they were otherwise unknown and untrustworthy.

  32. littleby says:

    The news article of the day is:

    https://www.apnews.com/fd2226d9e7bf46ca8d2542bdf88c26be

    If this article is to be believed, China is treating opioid addiction with mind controlling brain implants.

    Quote: “This machine is pretty magical. He adjusts it to make you happy and you’re happy, to make you nervous and you’re nervous,” Yan said. “It controls your happiness, anger, grief and joy.”

    I should be horrified — at some level I am horrified –, but what I’m actually thinking is: we live in the future now.

  33. The original Mr. X says:

    One of the criticisms I’ve seen of the last few Game of Thrones episodes is that Daenerys seems to be carrying the idiot ball a lot, pointlessly losing men and dragons and alienating potential allies, all so that the writers can ratchet up the conflict by making the balance of power more in Cersei’s favour.

    So, I thought it would be interesting to hear what people here think would be a good strategy for Daenerys to take over the Seven Kingdoms. Say that you’re one of Dany’s advisers; the time is at the start of series 7, the Queen has just occupied Dragonstone, and she’s asked you for advice on what her next moves should be. What do you suggest?

    • albatross11 says:

      Go take King’s Landing, wipe out Cersei’s army, and if possible, capture Cersei and Jamie. That’s their center of power now, more than Casterly Rock (which Tyrion knows is much less valuable now that the gold is apparently all mined out). Showing up with a much tougher army than Cersei has, giving her minimal time to plan a defense, makes the most sense.

      • Nabil ad Dajjal says:

        This.

        If Daenerys came to King’s Landing with three fully-grown dragons, legions of Unsullied, the Dothraki horde, armies of Dorne and the Reach, and a fleet of Meereenese galleys and Ironborn longships then Cersei would have absolutely no chance to stop her. There probably wouldn’t even be a battle, since all of Cercei’s forces aside from possibly the Lannister army were allies of convenience: I don’t see either Goldcloaks or the Iron Fleet laying down their lives in a doomed battle to protect her throne.

        The cost would be high for the civilians of King’s Landing. Many would burn, and many of the survivors would be brutalized by the Dothraki and other non-eunuch armies when the city was sacked. But the war would definitively end with Daenerys sitting on the Iron Throne with the largest army in Westeros and no serious rival claimants.

        • The original Mr. X says:

          Would there be any way to end the war without large-scale civilian casualties? E.g., if she made a proclamation saying “Look, citizens and garrison of King’s Landing, I have a bigger army, three giant dragons, and a superior fleet, and all your allies have deserted you. There’s no way you can defeat me. I don’t want to harm you, so if you throw open your gates and deliver up Cersei Lannister bound and trussed like a roast hog, we can end this thing peacefully without any raping or looting.” Or even just tried to starve the city out. (Although for propaganda reasons a quick victory would probably be better. Then again, taking the city without major bloodshed would *also* be good for propaganda, so IDK.)

          • Nabil ad Dajjal says:

            I don’t see how starvation is any better than a sack, especially since there’s a good chance that a siege would end in a sack anyway.

            Demanding surrender is a good idea, and the more likely Daenerys winning seems to be the more likely that Cercei’s unreliable allies would betray her. But in order for that to work, she needs to be prepared to take and sack the city: hesitation will be interpreted as her being unsure that she can win.

    • meh says:

      I would tell her that instead of wasting time telling Jon Snow you don’t believe him, just go find out for yourself, since it is only a 20 minute Dragon ride to get North of the wall.

    • A Definite Beta Guy says:

      What Tyrion said. Take the Lannister base of power, besiege King’s Landing with the Tyrell army, wait for someone inside King’s Landing to get hungry or irritated enough to stab Cersei in the back, become Queen.

      You will lose the Reach, but you will still crush most of the Lannister army and you will leave the Reach leaderless. Then besiege King’s Landing with the Unsullied instead and wait for Euron to show up, and start carving lighting up his fleet with Dragonfire. Keep the Tyrell gold locked up in port and inform the Iron Bank that if they want said gold, they better be sending mercenaries to help you.

      Dany isn’t holding the Idiot Ball so much as the writers buffed up all of her opponents in entirely unrealistic fashion. Euron built an entire fleet out of nothing in a matter of months, with super-spears that cripple ships faster than any artillery piece earlier than the 20th century, with accuracy rivaling the precision of modern SAM systems. Somehow Cersei convinced an entire company of mercenaries motivated by solely by money to fight an army full of undefeated Mongol hordes, undefeated slave phalanxes, and undefeated dragons. The Night King’s javelin throwing abilities make NFL quarterbacks look like 2 year old children playing shotput, and his undead wights overwhelmed a castle without siege equipment. And even if you want to give THEM a super-natural edge, the Lannisters and the Tarlys successfully took Highgarden without so much as a freakin’ ladder.

      I mean, there’s only so much you can do when the DM insists your enemies are rolling nat 20s every single time.

      • Lillian says:

        The start of winter is the absolutely worst time to besiege a Westerosi fortified settlement. At that point, they have spent years stockpiling food for the winter, which means that forcing them into starvation is not actually a viable strategy. It will take years for them to run out of food, and meanwhile your army is sitting out there in tents and other makeshift dwellings instead of proper shelter, and feeding them is going to be a problem because all the food has been stockpiled inside the very fortifications you’re trying to besiege. The show seems to have completely forgotten this, but this is the reality established by the setting.

        Tyrion’s plan was bad, and his excuses for why it was the correct plan were bullshit. The only reason he proposed it is because Cersei is his sister, and even though she’s an evil bitch who has mistreated him terribly, he still loves her. He didn’t want to kill her, and because he’s a smart guy he came up with very reasonable sounding and plausible excuses to not to strike the enemy queen with everything they’ve got. But doing exactly that was the right plan.

        A dawn dragon attack on the Red Keep followed by landing all their forces in and around King’s Landing would basically end the war right then and there. With the sun at their backs the defenders wouldn’t see the dragons coming until they were right on top of them. Cersei and her court would perish in the flames, and the rest of the city garrison likely surrenders in the face of the massive force arrayed against them. Without Cersei, there’s nobody left around whom to organize resistance, so Danaerys pretty much wins by default. Like i said in a previous thread, it would have been the ultimate “King me!” moment.

        Frankly the show would have been much better for it if they’d gone for that in the first few episodes of Season 7. The rest of the season could have been spent on Danaerys trying to consolidate after her victory, and learning about the impending ice zombie army threat. She is really good at conquering things, but sucks at ruling, so it would have been interesting to see her try to rule Westeros. Especially if she is shown having learned some lessons from ruling Mereen, but finding new challenges due to the different social structure. Also you know, giant ice zombie army.

    • John Schilling says:

      The plan Tyrion and Varys laid out at the start of S7 was actually fairly good, if you want to win with minimum civilian casualties and rule by something other than fear and terror in the aftermath. Both Euron’s impossible naval capabilities and Tyrion’s implausible ignorance re the Lannister gold reserves are irrelevant -or from another point of view, inevitable. Since Danny isn’t allowed to win until the end of S8, whatever plan Team Danny comes up with in early S7 will be doomed by fiat.

      If we assume it’s just those weaknesses, then there needs to be a decisive engagement with the Iron Fleet before anyone starts moving by sea. Dragons should make that a decisive win. And Tyrion’s attack gets redirected towards the food-producing regions of Tyrell, to the same net effect as Cersei’s armies need to eat even more than they need to be fed,

      If we just want a fast win and don’ care about civilian casualties, then as albatross and Nabil point out, a coup de main against King’s Landing is it.

      • albatross11 says:

        Westeros has been grinding its way through a brutal war of all against all for several years now. Almost certainly the most humane way to wage further war is to win the damned thing quickly, and then reestablish order and deal with the (then unknown to Dany et al) impending ice zombie apocalypse. (Note that until Dany lost a dragon to them, the ice zombies were apparently contained by The Wall.)

    • aristides says:

      I think the other responses are the best strategies, so I’ll approach it differently. Do everything the same through episode 5, spoilers to follow. Next secretly execute Jon and Tyrion, since there is nothing you could do at this point for them to forgive you, and you do not have to worry about another claimant. Then order the unsullied to kill all the Northmen in their sleep after they finish looting and partying. Winterfell will be easy pickings after their army is decimated and Drogon is still alive. Have the Dothraki terrorize the countryside so Westeros lives in fear, while the unsullied serves as your Queen’s Guard. Give what’s left of the castles to your loyalist followers. Eventually get assassinatedby Arya wearing Grey Worms face, because at this point I can’t think of a way to stop that from happening.

    • Dack says:

      Say that you’re one of Dany’s advisers; the time is at the start of series 7, the Queen has just occupied Dragonstone, and she’s asked you for advice on what her next moves should be. What do you suggest?

      Hire a faceless man or three.

  34. nkurz says:

    I recently came across an interesting approach to law enforcement: incentivizing the public to document violations by sharing a percentage of the fines collected. In the example I read about (http://nymag.com/intelligencer/2018/07/on-the-upper-west-side-a-one-man-war-against-air-pollution.html), New York City had laws on the books requiring truck drivers to turn off their vehicles rather than leaving them idling on the streets, with shorter time limits in school zone. Enforcement was poor, and as a result, compliance was low.

    Then the city changed law to give 25% of the fines collected to those who provide evidence of violations. Now a small industry has popped up of people who use their cell phones to record evidence of violations. Some of these are now make thousands of dollars per year.

    Is this a good system? I tend to think so. I don’t like noise, and don’t like pollution, and generally like the idea of having well enforced laws. I guess there are potential downsides, but overall it seems like an improvement. The sentiment I’ve seen in online comments, though, is usually strongly negative. What do people here think?

    • In Periclean Athens, prosecution of crimes was private. A successful prosecutor got a share of the fine–details depending on the particular case. This raised a potential problem. Find someone who is rich and unpopular. Accuse him of a crime. With luck and demagoguery get the jury to vote to convict him (majority vote, not unanimity as in most modern U.S. cases). Pocket the money.

      The solution was that, for many sorts of cases, if the prosecutor failed to get at least a fifth of the (large) jury to vote for conviction, the prosecutor was fined.

      One can imagine similar problems arising here. With a little careful photo editing … .

      • Protagoras says:

        The solution was that, for many sorts of cases, if the prosecutor failed to get at least a fifth of the (large) jury to vote for conviction, the prosecutor was fined.

        Sources from the time seem to indicate that the frivolous prosecutions continued to be a problem despite this and other efforts to discourage them. Which of course gives more cause to be concerned about the prospects for similar approaches like the one described above.

    • The Nybbler says:

      A whole world of snitches? That’s horrible. If our governments (federal, state, and local) could somehow restrain themselves to good and just laws, and not make it so we’re constantly committing violations, misdemeanors, and even felonies that harm no one, things might be different. But that’s not the case, so the fact that an official has to be around to see the violation acts as a check on state power. Make incentives for private snitching, and that check is gone.

      • nkurz says:

        One doesn’t need to take this approach for all laws, only for ones that the public judges to be for the public good. I presume you’d agree that there are some laws that are actually worth enforcing? Might this approach work for those laws?

        Also, I think you need to consider the second order effects. I’d expect that as soon as public officials start paying large fines for breaking lousy laws, there will be considerable pressure to change these laws. In the current system, selective enforcement ensures that those who make the laws rarely bear the cost. If financially incentivizing vigilante citizens helps to even out the enforcement, might this help us move more quickly to better laws?

        • John Schilling says:

          How does “the public” exercise their judgement in this respect? It seems like, if there were a good answer to that, it would have been applied at the level of “do we make all these ridiculous laws in the first place”, and that clearly didn’t work.

        • The Nybbler says:

          One doesn’t need to take this approach for all laws, only for ones that the public judges to be for the public good.

          Sure, under this approach, those will be the ones which result in the most fines.

          I presume you’d agree that there are some laws that are actually worth enforcing?

          Oh, sure, crown me Mad King Nybbler and I could probably find a few. But here in the real world where I’m just a schlub, I’m pretty sure the laws that I find worth enforcing won’t be the ones enforced, and the laws that prohibit me from doing things will. For instance, as I’ve noted before, every time I fly my model helicopters I’m letting myself in for a few $1100 to $20,000 fines. I’m sure a little Gatwick-style “drone” hysteria would get those laws on the list to be enforced by freelance fine collectors.

          Also, I think you need to consider the second order effects. I’d expect that as soon as public officials start paying large fines for breaking lousy laws, there will be considerable pressure to change these laws.

          Sure. To exempt public officials in some way. Either actually making them immune (e.g the Governor of PA being driven at 100+mph on the PA turnpike.. in an official police vehicle), or by officially or unofficially setting the standard to prove a violation MUCH higher in the case of public officials.

          If financially incentivizing vigilante citizens helps to even out the enforcement, might this help us move more quickly to better laws?

          Accelerationism doesn’t work.

      • onyomi says:

        I feel like it creates an additional bad incentive, a dynamic I believe I’ve seen elsewhere: the government does a terrible job at doing the things we expect the government to do; government finally comes up with an idea that amounts to getting the general public to do their job for them; incompetence of the government in this one area becomes less obvious, but neither do they cut spending or lower taxes to compensate the public for the job they’re pretending to do but actually offloading.

        Of course, this may smack of the libertarian version of “don’t give to charity because it slows the day when the contradictions inherent to capitalism that make charity necessary finally come to a head,” a form of argument I’m not very sympathetic to regardless of which side it comes from (might be a form of “accelerationism”). And I’m also certainly not saying e.g. “let’s stop allowing private schools so everyone can realize faster how bad public schools are.”

        At the same time, I feel like a lot of libertarians, sometimes calling themselves “agorists” really emphasize the whole “let’s build alternatives to the state and render it irrelevant!” angle without, as far as I’ve seen, a lot of evidence that the second step in the plan actually happens. As far as I can tell the order tends to be something like: government sucks, private entities fill the gap, government continues to suck (this is the first plan I’ve seen where the general public can earn money for doing the government’s job for it, however, so it has that to recommend it, I guess, though I also am not a fan of the “culture of snitches” aspect mentioned above*).

        *An additional problem is you know it’s a matter of time before someone’s ass gets kicked because someone sees him snapping a photo of his license plate, etc. Getting the general public to enforce this kind of thing also subtracts from the sense in which the impersonal aspect of the law/government allows it to absorb interpersonal ire.

    • ana53294 says:

      Why does anybody leave a vehicle idling, anyway?

      I always switch off my car when I stop, except for red lights. It takes like 5 seconds to switch it back on. You can put the car on neutral, put the handbrake, and let your feet rest. Isn’t that more comfortable?

      I get terrible cramps from holding the clatch. And putting the car on first is what takes me more time; switching it on takes no time at all.

      Is it an automatic transmission thing? I’ve never driven one, as I always used a manual car.

      • LHN says:

        In an automatic you can put it in park and there’s no effort involved, certainly. And while the car’s on you have climate control, your entertainment system, and your phone is charging instead of discharging.

        (You could get the latter two by putting the car in accessory mode, but that’s a couple of clicks extra effort and may start something beeping or blinking.)

      • Douglas Knight says:

        This article gives a variety of reasons that long-haul truckers idle all night, but it doesn’t really address why they don’t turn them off for an hour of unloading in the middle of the city. And many of the details may be different in Europe, such as weather and size of trucks.

        Diesel trucks are different from petrol cars. It says you’re supposed to take 5 minutes turning them on and off, not 5 seconds.

        (The Christian Science Monitor! That’s not a name I see every day. Oh, it’s from 1996.)

        • ana53294 says:

          I’ve never seen trucks idling at night stops in Spain. I guess they do it when it snows, but not as a habit.

          Fuel prices are higher in Spain, though.

          The adjustments described in the article seem like they would be much cheaper than the EV subsidy, and they would reduce a lot of emissions. Why hasn’t this been subsidised as a federal policy? It would both help truckers (small business) and the environment.

        • Aapje says:

          It says you’re supposed to take 5 minutes turning them on and off, not 5 seconds.

          No, it only says that you have to let it cool down by idling the engine for 5 minutes.

        • Gray Ice says:

          Like Douglas Knight said:
          -Diesel engines are harder to start in cold weather.
          -Drivers get used to idling their engines for long periods of time (to avoid problems) and to stay warm and have power.
          – This can e a problem when drivers idle their truck in a residential area.

      • Aapje says:

        @ana53294

        You don’t have to hold the clutch in a manual. You can shift to neutral and then disengage the clutch.

        Note that start-stop systems are increasingly added to cars.

        • ana53294 says:

          Sure, but you still have to put the handbrake on, or keep the foot brake. And you then have to put the car on first. I find changing gears takes more time than switching a petrol car. Now, I was told that trucks take much longer, so I understand why trucks idle – but I still don’t get why you would keep a car idle, unless it’s freezing winter.

          I find that you save no time by idling on neutral. You have two hands – one switches on the car, the other removes the handbrake, simultaneously. It doesn’t save much time.

          • Aapje says:

            A likely reason why people idle a passenger car is to keep the air conditioning working. It requires a running engine.

            Another issue that I’ve found is that it is hard to turn off the engine without turning off the electrics. So if people are listening to the radio and such, they may not want to first switch off everything and then move the key half-way to get electrics back.

            BTW. I’m a bit confused how you easily work the handbrake and the key at the same time. Both are typically on the right, intended to be used with the right hand.

          • ana53294 says:

            I have a British car – the key is in the right, the gearbox on the left. Now that I think of it, European cars do make that uncomfortable.

            Spanish second hand cars tend to be very old, whereas British second hand cars that are less than ten years old are much cheaper. There is a floor price for a car that works that has passed technical inspections, and those beater cars are priced similarly.

            Except for the south of Spain, you don’t need aircon much in the north of Spain. Do you use it in the Netherlands?

            The 5-10 seconds you miss of radio or other between turning off and half turn may make sense in a red light – but if your car is stopped for a much longer period, is that so critical?

          • AlphaGamma says:

            BTW. I’m a bit confused how you easily work the handbrake and the key at the same time. Both are typically on the right, intended to be used with the right hand.

            Hence Porsche’s tradition of the key being on the left (in an LHD car) to allow it to be started more quickly in the traditional Le Mans start.

            (Meanwhile Saab had the key next to the gear lever so it was away from the driver’s knees for safety, and some Saabs required you to select reverse to remove the key)

          • Aapje says:

            @ana53294

            There doesn’t have to be a need for aircon, for there to be a desire for it. More advanced aircons allow the setting of a temperature, so I can see some people leaving it on all the time, if they don’t care much about fuel use.

            And yes, it gets cold enough in The Netherlands to makes aircon very useful in winter and hot enough to make it very useful in summer. In winter, you don’t just have to deal with the cold either, but with fogging of the window. In summer, cars can heat up a lot in the sun, as they are basically green houses on wheels. I use a car sun shade if my car is parked in out in the open on sunny summer days, but it doesn’t remove the usefulness of aircon.

    • brad says:

      They should definitely put this in place for placard abuse. (Why they definitely won’t is left as an exercise for the reader.)

    • kaakitwitaasota says:

      This is the new system the Chinese government has put in place to catch illegal English teachers; fines can be up to a few thousand bucks in some cases.

      It probably works to an extent on teachers who aren’t actually from the Anglosphere; in my opinion you’d have to be a pretty crappy person to turn in some poor Russian girl for running afoul of a rather arbitrary visa law, but there are a lot of crappy people in the business.

      Of course, one issue is that not being on the right visa is a private crime against the state, not a public crime against other people. If you never tell somebody you’re on a tourist visa, they’d have no way of knowing unless they look at your passport.

  35. Le Maistre Chat says:

    Warning, CW:

    Can we have a rational discussion of race and violent crime? Taking homicide as a first approximation for all violent crime, in the US 52.5% of homicide offenders are black compared to 12.6% of the population. While people do sometimes get falsely accused of homicide, blaming this disparity on current racism would logically entail an enormous conspiracy theory: to have equality of outcome obscured by racism would require the police arresting and charging an innocent black person for a person of another race’s homicide at least 38% of the time (4.167x over-representation / ~1/2 of arrests).
    However, there’s an interesting possibility that we can blame the legacy of racism for blacks committing 4.167x violent crimes per capita relative to whites. IIRC, Thomas Sowell proposed that in 1865, Afro-Americans had a culture of violence equal to pre-industrial European and Euro-American peasants and proletarians, and traces the trickling down of bourgeois values down the class ladder to everyone who looked like them… thus excluding Afro-Americans.

    Though on reflection, it’s not clear what going with 19th century racism over genetics accomplishes. People could say it doesn’t matter why blacks are >4x more violent, they have a right to stay away from criminal demographics regardless, and raise the racist class warfare narrative about housing prices and “good school districts.”

    Thoughts?

    • albatross11 says:

      A good source for official crime statistics is BJS. I worked through the homicide numbers a few years ago on Making Light, and got a similar answer–there’s basically no way to make the existing statistics consistent with a claim of blacks and whites committing murder at anything like the same rates.

      This is the sort of information that rarely appears in mainstream media coverage of crime issues, race-and-policing issues, etc. The journalists and editors there are either ignorant of the most basic statistics on crime (you can find official murder stats with like five minutes of Google searching), or they simply don’t think that this is information their readers can be trusted with. The murder rate among blacks is, according to official statistics, about eight times as high as among whites. The rate of *being* murdered is about seven times as high among blacks as among whites. Most murders are within the same race. Nearly all murderers are male, as are most victims, though there’s a substantial number of women being done in by their boyfriends, exes, husbands, etc., per year.

      Looking at those numbers, or just reading the _Homicide in the United States_ report, gives you a radically different picture of the world than the one you get watching TV news coverage or reading newspaper coverage, and even more different from what you’d get from opinion journalists and politicians.

      Now, some people who trumpet these numbers want to stir up hatred against blacks, or justify police misconduct against blacks, or whatever. It’s perfectly reasonable not to want to give those folks fuel. But you can’t have a meaningful discussion of race and crime in the US without knowing the basic outline of what the crime statistics look like. Withholding this information is just making sure that the people who read and watch the news–far more informed than the average person–are doomed to speak and think nonsense about race and crime.

      • littleby says:

        Is this true even after controlling for income and education?

        • Le Maistre Chat says:

          Updating from what HeelBearCub said about lead, the thing we should be looking at is family income and education (education presumably being a proxy for living in a nicer area without leaded soil), not the individual perp.

          • albatross11 says:

            We’re not talking about causes right now, just about bare observable facts. Why blacks have eight times the murder rate of whites is an interesting research question, and you can imagine any number of plausible hypotheses to explain it–poverty, racism, too few fathers, some broken part of American black culture, gang culture in inner cities, etc. But if you want to have a discussion about race and crime and policing in the US, you need to know that bare fact that blacks commit way, way more crime per capita than whites. I’ve watched intelligent, educated people react with genuine anger and outrage (and extreme skepticism) to hearing that said in public.

            People who don’t know that basic fact, and try to understand black/white disparities in arrest rate, or probability of being shot by the police, or incarceration statistics, are doomed to talk nonsense. And yet, lots of apparently smart and reasonable people don’t know that fact, because the organizations they trust to inform them on important issues of governing the country do not see fit to ever bring the fact up–it would complicate the story, or give people the wrong idea, or embolden racists. And so most of the country talks nonsense about these issues, or gets into dumb arguments about well-established basic facts.

          • HeelBearCub says:

            @albatross11:
            Why the crime rate for [subset] is higher makes a huge difference to just about every relevant thing you can discuss abut the difference.

          • albatross11 says:

            hbc:

            But you can’t even have that discussion, or even *start* that discussion, until you know that crime rates differ a great deal. Which is one reason to encourage mainstream information sources to make it easy to find that fact in their related reporting.

            Also, once you know that rates differ, trying to fix it probably requires knowing why it’s happening, but adapting to it in your current life or policies probably doesn’t. Knowing that the murder rate is much higher among blacks than whites helps you understand why the arrest rate of blacks is higher. (Though of course you don’t know for sure that there’s not some component of racism in there alongside the applied statistics.) It offers a pretty obvious explanation for why blacks are disproportionately in prison. (Again, though, that doesn’t rule out racism, it just means that racism isn’t the obvious explanation.) And so on.

          • salvorhardin says:

            I would not assume that family education or income are good race-neutral proxies for living in less lead-contaminated areas, given the history and stickiness of residential segregation in the US.

          • Radu Floricica says:

            @HeelBearCub

            The problem is the asymmetry of discourse. You can say publicly that police are shooting more black people than white, but you can’t say black people commit more crime – you have to say poor people commit more crime.

            It would be perfectly ok – indeed, it would be more productive – to move the whole conversation to the whys and say that police tends to shoot poor males of certain age and poor males of certain age commit more crime.

            But it takes two to tango, and talking about race in one context and socioeconomic status in another makes for a far more complicated conversation than needed.

          • Le Maistre Chat says:

            The problem is the asymmetry of discourse. You can say publicly that police are shooting more black people than white, but you can’t say black people commit more crime – you have to say poor people commit more crime.

            This. There must be a base rate at which being a violent criminal gets you shot by police, unless we disarm the police. If black Americans commit 52.5% of it in the US, we have to predict that a slim majority of police shooting victims will be black under the base rate. That’s not racism, though racism could raise it above the base rate due to cops finding black goons scarier than Euro-American or Latino ones.

          • HeelBearCub says:

            There must be a base rate at which being a violent criminal gets you shot by police, unless we disarm the police.

            People aren’t particularly bent of shape about the violent criminals who get shot. It’s the non-violent, non-criminals that really move the needle.

            Yes, when that needle starts moving it won’t necessarily be very p“discriminating” in what it sweeps along with it. But that’s more of a side-effect.

            And it’s not as if the violent criminals are uniformly shot.

          • The Nybbler says:

            Almost twice as many whites are shot and killed by police as blacks, however.

          • Randy M says:

            People aren’t particularly bent of shape about the violent criminals who get shot.

            But there is very often radically different degrees of skepticism about the violence of alleged criminals, such as Michael Brown or Trayvon Martin (the latter admittedly not killed by police, but still relevant because he was killed in a situation in which his violent behavior may justify his killing while his non-violent behavior would not).

          • acymetric says:

            @The Nybbler

            Isn’t that mostly going to be a function of population size? There are a lot more white people out there to shoot.

            70% of the population accounts for 46% of police shootings.

            12% of the population accounts for 23% of police shootings.

            Black people are also much more likely to be shot while unarmed (just under half of those shot while unarmed were black). Unarmed in this case means “no weapon of any kind”.

          • The Nybbler says:

            @acymetric

            You would expect if police killings were mostly of criminals, and one particular racial group was ~50% of the criminals, they’d be roughly 50% of those killed by police. This is independent of racial population proportions.

          • You would expect if police killings were mostly of criminals

            Even for killings not of criminals. If half the crimes are committed by blacks we would expect about half of the police/criminal interactions to involve black criminals. Black criminals are likely to live in places where most of the non-criminals are also blacks, so non-criminals who get accidentally or mistakenly shot in such interactions are also likely to be black.

          • albatross11 says:

            Nitpick: The available crime statistics show blacks with a way higher murder rate than their rate of most other crimes. I think blacks have a higher crime rate than whites for almost all crimes, but murder is way, way higher.

            One possibiltiy is that this has to do with murder being more consistently reported and investigated than other crimes. (In that case, we’d expect the actual crine rates to be similarly high for other crimes.). Another possibility is that this has to do with some specific thing (urban drug dealing, blacks adopting the poor Southern whites’ honor culture, whatever). (In that case, we’d expect murder to be an outlier, and rates of other crimes to be much less skewed toward blacks.)

          • albatross11 says:

            I believe the *arrest rate* of blacks vs whites very closely corresponds to the rate of police shootings of blacks vs whites. The natural model here is that every arrest has some probability of going terribly wrong and ending up with someone dead.

    • rlms says:

      While people do sometimes get falsely accused of homicide, blaming this disparity on current racism would logically entail an enormous conspiracy theory

      I don’t want to say literally no-one has ever suggested that, because probably someone has. But it’s wacky to suggest it’s common enough to be worth engaging with.

      The actual mainstream left-wing explanation is approximately that racism (past and present) causes poverty and poverty causes violence. Perhaps you are in an extreme bubble and haven’t come across that, but as I understand it the mainstream right-wing explanation is that bad culture causes poverty causes violence, so I don’t really understand why the two explanations you suggest are 19th century racism and genetics.

      The glib mainstream explanation is incomplete on two levels. Firstly, it’s not exactly clear to what extent racism and bad culture are responsible for poverty. Secondly, poverty doesn’t explain the whole of the racial crime gap. On the second point, this is due at least in part to higher levels of crime in more densely populated areas and differing racial geographic distributions. I don’t know what happens when you control for population density, although I expect that would be difficult to do well since there just aren’t really any majority-white densely populated extremely poor areas. And of course this all involves assumptions about the directions causation is going. So it’s not a perfect explanation for everything, but it’s rather more nuanced than a random conspiracy theory, and what you need to engage with if you want to have a “rational” discussion.

      • Le Maistre Chat says:

        The actual mainstream left-wing explanation is approximately that racism (past and present) causes poverty and poverty causes violence. Perhaps you are in an extreme bubble and haven’t come across that, but as I understand it the mainstream right-wing explanation is that bad culture causes poverty causes violence, so I don’t really understand why the two explanations you suggest are 19th century racism and genetics.

        What? The Sowell (at least I think it was him) explanation is a variant of the basic right-wing one that bad culture causes violence (black Americans missing out on bourgeois values trickling down the class ladder, due to racism).
        I’m not sure how you think I should more deeply engage with the Left’s glib mainstream explanation that racism causes poverty, since as you note it’s too incomplete to possibly be right. I can and am looking at it from the angle of racism contributes to poverty and poverty = brain damage from lead, but we need data on whether per capita rates for poor urban whites match those for blacks who are or grew up poor.

        • rlms says:

          I can and am looking at it from the angle of racism contributes to poverty and poverty = brain damage from lead

          Great, but there’s nothing in your original comment to indicate that.

      • RalMirrorAd says:

        Well one big problem with poverty is that it can, at best, explain variation within a country but might not be so good at explaining variation between countries.

        I suspect that a country poorer than the united states does not have an average homicide rate equivalent to the average of the segment of the US population whose incomes are equivalent to said poorer country.

        So how does the target population we are discussing compare with say, eastern Europeans [wrt income]

        • Le Maistre Chat says:

          This is an interesting point. We hear about how homicide rates are so much lower in Europe than the US… so what’s it look like in the poorest European countries?

          • RalMirrorAd says:

            These are the ‘Region = Eastern Europe’ intentional homicide rates according to le wiki.

            According to the same table the US in 2016 was 5.35
            Also beware of very very small countries that can have wildly varied homicide rates like Greenland.

            Country Continent Region Rate #killed Year
            Belarus Europe Eastern Europe 3.58 340 2014 UNSDC/CTS
            Bulgaria Europe Eastern Europe 1.14 81 2016 UNSDC/EUR/CTS
            Czech Republic Europe Eastern Europe 0.61 65 2016 UNSDC/CTS
            Hungary Europe Eastern Europe 2.07 202 2016 UNSDC/CTS
            Poland Europe Eastern Europe 0.67 256 2016 EUR/UNSDC/CTS
            Moldova Europe Eastern Europe 3.19 130 2014 UNSDC/CTS
            Romania Europe Eastern Europe 1.25 247 2016 SDG/UNSDC/CTS
            Russia Europe Eastern Europe 10.82 15,561 2016 NSO/CTS
            Slovakia Europe Eastern Europe 1.05 57 2016 SDG
            Ukraine Europe Eastern Europe 6.34 2,845 2014 UNSDC/TSMNEE/CTS/DMDB

            A better approach might be to just show homicide rate by per capita GDP.

            _______

            More thoughts:
            And sorry in advance if this analysis stereotypes eastern Europe as poor. In pure $ terms most all countries are poorer per capita GDP then the US.

            Also per capita GDP may not be the right approach. The gini coefficient of each country might reveal subpops with very low incomes relative to the average which would still lend the possibility that some kind of relative poverty causes crime. However it does not eliminate the reverse possibility that crime causes poverty, directly or indirectly.

          • Aapje says:

            Some of these countries have a lot of emigration of young people, who are typically the most violent. So this should drive down their murder rates.

      • Conrad Honcho says:

        I understand it the mainstream right-wing explanation is that bad culture causes poverty causes violence,

        I’m more inclined to believe violence causes poverty. There are plenty of places that are poor but not violent. But once violence/crime starts in an area, everyone with any money leaves and no one who has money wants to move in. Hence the area falls to poverty and disrepair.

        I’m also not sure we can talk about murder without talking about the drug trade. If you want to avoid being murdered in the United States, the best piece of advice I can give you is “don’t get involved in the illicit drug trade.”

        • albatross11 says:

          +1

          I suspect a high crime rate is one equilibrium you can fall into as a community/society, and low crime rate is another. Once you’re established in one, it’s going to take a lot of work to get to the other, but the right conditions and policies can move your society in either direction. I imagine this is sort-of like a culture of bribery and corruption—once established it’s very hard to eliminate, but you can establish a culture that has almost none of it.

          One way this likely works is a feedback loop—high crime convinces people with any choices to move away, which drains the tax base and pulls the best students (parents with something on the ball, and also parents who care about their kids’ education) out of the local schools, and also drives away businesses who find it uneconomical to do business in places where they get robbed once a month. Few people considering where to locate their company offices think “hey, how about someplace where we’ll need armed guards and secure parking to keep our employees safe?” As jobs move away and stores close, there are fewer and fewer alternatives to crime to make a living.

      • RalMirrorAd says:

        I would rank the explanations as follows; (approximately…)

        New Far Left: The cause is systemic racism and discrimination
        Old Far Left: The cause is the legacy of slavery
        Center-Center-Left: The cause is poverty
        Traditional Right Wing: The cause is black culture
        Economic Right Wing: The cause is the great society
        Far Right Wing: The cause is due to differences in genetics

        Note that ‘Cause’ here can refer both to the level and the gap

        Lead was also brought up as a cause, but that’s rather niche.

        _______________________________________

        As far as the ‘Why can’t the gap be discussed’ — I think it’s because:

        1. The possibility of a genetic difference becomes immediately apparent in a way that talking about the arrest rate gap does not

        2. The ‘commission rate’ gap explains the incarceration gap. One would naturally have far more sympathy for people who were incarcerated well out of proportion to their rate of offense then a group that was over-incarcerated but at the technically ‘correct’ rate.

        • Plumber says:

          @RalMirrorAd,
          I’d love to see a top level post of your Left and Right factions/”Tribes” schemas.

          • Dan L says:

            I remember seeing your earlier post on the factions of the Left, and it made me think of this article at the time.

          • Plumber says:

            @Dan L, 
            That was interesting reading.

            Thanks! 

            I found a similar typology of elected Republicans by the same author, and I’m struck by how much more united Republicans are compared to Democrats (who are more divided by age than Republicans).

            I was struck by this statement on ‘Conservative Democrats’ in your link: “…This wing, I think, will punch above its weight in the national debate about where the Democrats are headed — because these Democrats will likely be those pushing loudest for it to avoid the policy stands of the Very Progressives and the Super Progressives. And they will have a compelling argument — by being elected, Edwards was able to expand Medicaid to more than 400,000 people in a very red state, a real policy change that Ocasio-Cortez can’t make…”, as because I’m 50 years old I just don’t expect to live long enough to see any harvests from the “moving the Overton window” actions of the “progressive” wing of the Democratic Party, and (ironically) I expect the “Moderates” to achieve more actual progress that I may live to see.

            In any case if the 2020 Democratic presidential nominee loses, or is single term, I expect the party to swing hard from whichever wing they’re perceived to be.

          • Dan L says:

            @ Plumber:

            I usually prefer data-driven stuff, but Perry Bacon Jr. has earned my respect on his qualitative reporting – it’s usually pretty good stuff.

            The quip goes: “Democrats need to fall in love, Republicans just fall in line”, and while there’s some truth to the idea that Republicans are inherently a more unified party I think it’s definitely true that it’s easier to have a pole to rally around when the party holds the presidency. I’m really interested to see what happens to the Republican party the next time they run against an incumbent D president, in however many years that may be.

            I’m endlessly fascinated by how much more effective politicians seem to be when they’re conspicuously centrist, especially when they’re in executive roles. There’re a lot of reasons that may be, but so far I definitely agree that the conservative wing of the Democratic party has way more power to actually get their priorities adopted in the next 6 years than maybe any other faction of either party. Mm, that might be a bit much – they’re definitely undervalued right now, though.

            But man, I am not looking forward to the Progressive wing losing their minds in the case of a Biden loss in 2020.

          • Clutzy says:

            The quip goes: “Democrats need to fall in love, Republicans just fall in line”, and while there’s some truth to the idea that Republicans are inherently a more unified party I think it’s definitely true that it’s easier to have a pole to rally around when the party holds the presidency.

            That’s probably because, as Steve Salier says, the Dems are a coalition of fringe interests that don’t overlap. Blacks and Hispanics are at odds on immigration, both are more anti-gay than your average American, rich liberals interests are generally not aligned with either group as well.

            I’m endlessly fascinated by how much more effective politicians seem to be when they’re conspicuously centrist, especially when they’re in executive roles.

            I simply think that this is because its easier to hire people that have experience at execution. Trump has had trouble governing because there are only like 10 people who have ever been in mid-upper level government positions that think like Trump. OTOH, a president Romney would have 10000 to choose from. Ditto President Sanders vs. President Clinton. Obama also had issues governing early on because he campaigned far more centrist than he wanted to govern once elected. The GOP House actually helped him a lot in that way.

          • Dan L says:

            @ Clutzy:

            That’s probably because, as Steve Salier says, the Dems are a coalition of fringe interests that don’t overlap.

            Zero overlap would be putting it too strongly, I think, but yup. There’s definitely an asymmetry there, and if I was god-king of the Republican party I’d be exploiting it [well beyond what is possible without an unincentivized incentivizer].

        • RalMirrorAd says:

          @Plumber what do you mean?

          • Plumber says:

            @RalMirrorAd,
            I”d like to see more on what defines the (your words) the “New Far Left”, the “Old Far Left”, the “Center-Center-Left”, the “Traditional Right Wing”, the “Economic Right Wing”, anf the “Far Right Wing”, for a typology with public figures and/or beliefs.

          • RalMirrorAd says:

            Well ‘Old/New’ split relates to the specific to the question at hand:

            1. Ostensibly, civil equality was granted in the 1960s
            2. the US has communities of African Migrants/Refugees who were never subject to NA Slavery or the pre civil rights era legal system.
            3. Similar migration has occurred in europe

            This makes certain older kinds of attribution less compelling. Attributing it to something “systemic” makes it harder to measure except in reference to the things that it is believed to cause, harder (potentially impossible?) to correct by changing a law, and naturally harder to disprove.

            And that is what is meant by new/old.

            A more moderate type might try to treat the problems as a subset of generic social ills like poverty. This might be more politically palatable for some people but disappointing or disingenuous to others. It also clashes with what some people including myself have pointed out that the gap cannot be fully explained at least in terms of income, although there may be better ways to look at poverty such that the quality of life of a person in a poorer country is better than the QOL for a poor person I a rich country.

            Blaming the problem on anti-poverty legislation has been a staple of right wing discourse for awhile though overton window shifts have made it a bit dangerous to say. The most extreme and disheartening hypothesis has grown in notoriety due to the internet and to a lesser extent the fact that the less extreme explanation is being driven from polite discussion regardless, but there is something of an intellectual leap you have to take to get from one to the other which is why I separate them.

      • albatross11 says:

        Plenty of people seem willing to claim that differences in incarceration and arrest rate between races are due to racism. This isn’t crazy–there’s visible racism among many people involved in the process. But it’s not consistent with the available data, at least not as a complete explanation for those differences.

    • DinoNerd says:

      Can we have a rational discussion of race and violent crime?

      Probably not.

      In general, discussions of large-non-homogenous groups and their statistically more common traits wind up with people insisting either that (all) those folks are bad, or that they personally are good because of their membership in the statistically-better-in-this-sense group.

      For similar reasons, we can’t have a raitonal discussion of Americans (of all races) and violent crime, as compared either to Northern Europeans or residents of war-prone dictatorships.

      As a Canadian, my bias is to see Americans as a bunch of gun-happy lunatics, shooting each other on a whim. I’m sure I can find statistics showing more gun violence in the US than Canada; I wouldn’t even have to fail to adjust on a per capita basis to “prove” this point. OTOH, because you are concerned with race, and American races in particular, I conjecture that you are American, and would find such a discourse both pointless and insulting.

      So getting back to the burning issue of races and violent crime – I’m not going to pick a short person for my basketball team, even if he’s 100% Masai. And I don’t actually expect my American neighbours to shoot me, in spite of their statistical propensity for gun crime – but that’s probably because I’ve lived in the US, and know real Americans, and so have direct personal experience that most of them are not actually crazy.

      In general, though, I find that dicussions of statistical traits of this kind usually wind up being ‘about’ this sort of conclusion.

      • albatross11 says:

        DinoNerd:

        If that is true, it is equally true for all us/them trait discussions. You can’t make that argument with respect to black/white difference in crime rate and then want to talk about how whites are more often racist than blacks, or men are more often violent then women. (I mean, you can, but not if you want to be consistent.)

      • quanta413 says:

        As a Canadian, my bias is to see Americans as a bunch of gun-happy lunatics, shooting each other on a whim.

        Have you checked who is shooting each other a lot? Utah is full of gun nuts, but they’re not even close to the top of American homicide rates.

        If you compare like ethnic groups to like ethnic groups, the U.S. is not as different from Canada as it would appear in statistics that don’t disaggregate. Although Canada is hardly a disarmed populace compared to Britain.

    • HeelBearCub says:

      Everyone here should be at least mentioning the lead crime hypothesis every single time these conversations come up.

      For example, here is Kevin Drum’s run down of a recent study showing how lead in top soil is disparately distributed, and the resultant behavioral issues are subsequently disparately distributed.

      – Overall, about 3.5 percent of boys report cognitive difficulties.
      – In counties with high lead concentrations, this increases to 7.5 percent.
      – In urban areas with high lead concentrations, it increases to 11 percent.

      The reason I’m highlighting this study is not because it shows that lead poisoning affects cognitive ability. That’s well-known and uncontroversial. However, it also shows specifically that lead in topsoil is a significant factor in producing those cognitive problems. And if topsoil lead affects cognitive ability, then topsoil lead also affects the likelihood of being prone to violent crime later in life.

      If you aren’t already familiar with Kevin’s extensive writing on the subject, it’s persuasive and broad. I can dig up multiple good examples if need be.

      • Le Maistre Chat says:

        – Overall, about 3.5 percent of boys report cognitive difficulties.
        – In counties with high lead concentrations, this increases to 7.5 percent.
        – In urban areas with high lead concentrations, it increases to 11 percent.

        Whoa, those are significant numbers, and a big oversight.
        So we could blame criminality on lead brain damage, a poverty issue which current racism makes racially-tinged, and not just the legacy of racism making black culture worse than working-class white.

        • albatross11 says:

          The lead/crime hypothesis looks interesting, but I don’t know how much it explains here. If lead exposure in the topsoil is the driver of black/white differences in crime rate, then we should see whites and hispanics who grow up in the same neighborhoods as blacks have the same murder rates. (Similarly if differential lead exposure explains the IQ difference–the difference should be by neighborhood rather than by race.) I haven’t looked into this, but I’m pretty skeptical that this is what you find when you look there.

          One natural experiment we can look at–there are neighborhoods that rather quickly shifted from majority-white to majority-black during the white-flight era. There are also neighborhoods that have shifted back during gentrification. Black and white children were born in those neighborhoods, only a few years apart. Do they have approximately the same crime rate, once income is accounted for?

          I do know that the black/white IQ difference isn’t eliminated (or even close to eliminated) by accounting for parental income/poverty, and is also not eliminated (or even close to eliminated) by looking only at kids in the same neighborhood.

          • HeelBearCub says:

            Here is a study that controlled for everything else (income, race, etc.) and still found “every 100-unit increase in meantopsoil lead content was associated with a 1.05 (95% CI: 1.03, 1.08) increased risk for violent crime events.”

            Clearly crime is multi-factorial in its causes. I don’t think anyone is disputing that. But the “ZOMG, crime wave” trends seem like they may very well be driven by lead, and exacerbated by everything else.

          • HeelBearCub says:

            And here is something he wrote from 2013 that specifically looked at lead exposure differences by race.

            Blood lead prevalence over 30 mg/dL among white USA children fell from 2% in 1976–1980 to less than 0.5% in 1988–1991, as prevalence over 30 mg/dL among black children plummeted from 12% to below 1%.

            One thing to note from that article is that “neighborhood” is not the same thing as “building”, as another source of exposure is old lead paint.

          • quanta413 says:

            Lead probably contributes to crime, but I am highly doubtful that it explains much of a gap in homicide rates between populations at the same timepoint. The nationwide homicide rates over the past several decades fluctuate by roughly a factor of 2 across the whole U.S. The average blood level of lead in children back in the 1970s was roughly 10x as high as the average is now. Black children do suffer from more toxic exposure to lead on average than white children, but exposure for the average black child in 2000 was still roughly 5x below the average of white children in the 70s (follow the links in the first link to the CDC NHANES report to see blood lead levels for the mid 90s to early 2000s).

            The homicide rates on the other hand, differ across groups by roughly 7x (a much larger difference than the fluctuations we see over time), and current homicide rates counting either victims or perpetrators who are black haven’t come down to the rate of the white population of decades ago even though the level of lead in blood in the black population is on average well below that of the white population decades ago.

            I don’t have an explanation for the gap though, and I dunno if one would be very helpful if it didn’t involve some other sublethal toxin (which I think is unlikely to explain most of the gap since most of the people in the U.S. are much less poisoned than they used to be). Sublethal toxins have relatively clear and distinct policy responses. The only other plausible policy responses I’m aware of all basically boil down to “better police”, and while the exact definition of better police is definitely debatable, it seems like that would pretty much always help.

            I’d bet that if you measured the homicide rate of a group like recent Igbo immigrants, that they’d have a notably lower homicide rate than aggregating over the entire non-hispanic black population. You can also look at homicide maps in single cities and see huge variations in homicide rates in small areas. I haven’t looked for a while, but last time I did the differences could be pretty striking. I’m not sure if much of the difference in homicide rates across groups is driven by the number of outlier subgroups within each group, but I wouldn’t be surprised if the effect was significant.

          • albatross11 says:

            Is there any data on whether blacks are more, less, or equally susceptible to lead exposure leading to depressed IQ or higher crime/violent events?

      • bullseye says:

        I can’t find it now, but I read a magazine article saying that mothers pass on lead to their children even if the lead exposure was years before childbirth. Lead collects in the bones, and the mother’s body pulls material from her own bones in late pregnancy. So not only is lead in your own environment bad for you, lead in your mother’s childhood neighborhood is bad too. The article didn’t address crime, but presumably it’s relevant.

        • Le Maistre Chat says:

          I can’t find it now, but I read a magazine article saying that mothers pass on lead to their children even if the lead exposure was years before childbirth. Lead collects in the bones, and the mother’s body pulls material from her own bones in late pregnancy.

          Jesus.

          • albatross11 says:

            If that’s true, shouldn’t it have messed up the nice curve from Drum’s original article about lead and the 90s crime wave?

          • HeelBearCub says:

            Not really, if the effect is of much smaller magnitude. I think it would just mitigate the attenuation.

      • HeelBearCub says:

        I said I could dig up examples, but here is a list of most of what Drum has written on the issue. There a huge number of articles, covering many studies.

        The most persuasive general trend, in my mind, is that in country after country we see crime rates rise and fall as their children’s blood lead levels do. Pretty much everywhere sees a dramatic drop in crime rates … not at the same actual time, but rather 20 years after whenever they ban leaded gas.

        • HeelBearCub says:

          The one about the rise in rural crime in the late 1800s and early 1900s is one that cuts away from the thinking that crime is a predominantly urban issue.

          Also, it explains why barns are red.

          • salvorhardin says:

            Is there anything in Drum’s analysis that responds specifically to the critique Barry Latzer offers in the links Atlas cites below? Key sentences seem to be:

            “The same “lead-free” generation that engaged in less crime from 1993 on committed high rates of violent crime between 1987 and 1992. Incidentally, there was plenty of lead in gasoline in the 1950s and early 60s when violent crime was low.”

          • HeelBearCub says:

            I’m not aware of Drum interacting with Latzer.

            Drum makes the point over and over that crime is multi-factorial. The lead-crime hypothesis is attempting to explain the sharp rise in crime we saw, and then the drop. But the idea that drops in poverty would also be associated with drops in crime is something Drum is already accepting as likely. He acknowledges that poverty plays a large role in crime.

            As to your point about cohorts, here is a post directly addressing some of the issues around cohorts.

            But the point about high lead in the 50s completely misunderstands the point about lead. Lead exposure in childhood results in developmental issues that result in high crime rates for the already most criminal age group, 20 year olds. People who were in their 20s in the 1950s grew up in lower lead environments.

            The expectation is that crime rates are growing along with lead exposure in childhood 20 years previous, not instantaneously.

          • broblawsky says:

            “The same “lead-free” generation that engaged in less crime from 1993 on committed high rates of violent crime between 1987 and 1992. Incidentally, there was plenty of lead in gasoline in the 1950s and early 60s when violent crime was low.”

            I don’t think that’s a very good criticism.
            a) People who are dead or in jail can’t commit crimes. Logically, there should be a certain level of burnout in the lead-afflicted population; without new lead in the environment, there won’t be enough replacement to sustain the existing level of criminality.

            b) Lead exposure doesn’t instantly turn people into criminals. It takes time for lead to accumulate in the environment and then for that lead to accumulate in children’s bodies, and then a little longer for the lead-afflicted children to have an opportunity to become criminals.

          • ana53294 says:

            Of the people who were of prime crime age (15-30) in the 50s, many of them would be dead.

            If young men with aggressive tendencies and impaired cognition (but not enough* to avoid the draft) get conscripted, they are much more likely to die.

            *And those who were too stupid would probably be very easy to catch.

      • INH5 says:

        Does Drum ever address why homicide rates for Hispanics show the exact same trends, even though a huge portion of of Hispanics in the 1990s were Mexican immigrants (39% of American Hispanics were foreign-born in 1994), and Mexico didn’t start phasing out leaded gasoline until the 1990s?

        Here’s something else: starting in 1994, poverty rates for all races in the United States, but especially for blacks and Hispanics started to steeply drop, not leveling off until the end of the decade. What’s more, this drop seems to have resulted in permanent improvements: black poverty rates had been steadily above 30% for decades before 1994, but after the ’90s they never reached those levels again, even during the worst years following the Great Recession. I strongly suspect that if you could isolate a demographic of disproportionately poor white people, you would see similar trends.

        1994 is also the year that violent crime started to steeply drop, and that drop also leveled off around 2000 (though it continued to decline at a slower rate until the mid-2010s). The timing matches way too closely for a coincidence to seem plausible, and it’s hard to believe that the phasing out of lead 20 years prior could cause such a large drop in the poverty rate (especially since, again, Hispanics show the same trends despite a lot of them being immigrants). So I really think that whatever caused the mid-90s drop in poverty rates probably also caused the mid-90s drop in crime.

        Granted, that still leaves the question of why crime rates didn’t go up again with poverty rates after the Great Recession, but I think that that can likely be chalked up to any combination of: better police/forensics technology, more security cameras, video games making people stay inside, Netflix making people stay inside, etc. Or maybe the fall in poverty rates didn’t directly cause the drop in crime, but was an effect of something that also caused crime rates to fall. This stuff is complicated, but, again, I think the timing is just too perfect to ignore this.

        As for what caused those poverty rate drops in the first place? It clearly wasn’t simple economic growth, as the 1980s saw plenty of economic growth that didn’t translate into similar drops in either poverty or crime rates. I’m not nearly qualified to answer the question of what was special about the ’90s, but plausible candidates include the computer revolution, the expansion of the earned income tax credit in 1993, and NAFTA (which came into effect on January 1, 1994). It seems like it might be a pretty important question for economists to look into.

        • Conrad Honcho says:

          If we’re looking for “what changed in the mid 90s” then perhaps the start of the era of mass incarceration? While not all people who use/sell drugs commit violent crimes, almost all people who commit violent crimes would also use/sell drugs. If you start locking up an awful lot of the population of non-violent offenders that removes them from society before they can become violent offenders.

          • INH5 says:

            1. Incarceration rates really started to take off in the 1980s, not the mid 90s, so this explanation doesn’t fit the timeline very well.

            2. I don’t think this can explain the drop in poverty. Even if we assume that every single prisoner would be below the poverty line and they aren’t counted in poverty statistics, there simply aren’t enough people in jail for the numbers to work out.

          • Conrad Honcho says:

            1. From those graphs, sure, the rates started climbing in the 80s but it’s not until 1990 that we’ve even locked up a million people. Absolute numbers count more than per capita in this sort of thing because it only takes one low-trust person to ruin a high-trust area.

            2. My read on crime and poverty is that crime causes poverty, not the other way around. There are plenty of places that are poor, but not crime-ridden. However, once crime starts in a place, everyone with money leaves and no one with money wants to move in, leaving the area to fall to poverty and disrepair. Lack of capital and jobs prevents the poor people in those areas from becoming not poor. So incarceration of future criminals -> less crime -> less poverty.

          • baconbits9 says:

            From those graphs, sure, the rates started climbing in the 80s but it’s not until 1990 that we’ve even locked up a million people. Absolute numbers count more than per capita in this sort of thing because it only takes one low-trust person to ruin a high-trust area.

            This sounds like a just so story, incarceration rates double from ’74 to ’85 and triple from ’74 to ’92, and is a larger rate increase (400 people per 100,000 vs 300) than the jump from 92-96, but the last 300 is hugely meaningful and the first 400 isn’t at really at all?

            Edit: Discussing numbers from the male incarceration rate here, but general trend holds for total incarceration rate obviously.

          • Conrad Honcho says:

            The types of criminals incarcerated changed in the 90s. And I said absolute numbers matter more than per capita rates. Eventually you hit a tipping point wherein those most likely to commit violent crime have been locked up and a lot fewer violent crimes happen.

            Is it a Just So story? No more than every other explanation for broad, nationwide trends. The real answer is probably too complex. Was it mass incarceration, dropping lead levels, easy access to abortion 20 years beforehand, the rise of indoor entertainment, better policing and more security cameras? Yes.

          • baconbits9 says:

            Is it a Just So story? No more than every other explanation for broad, nationwide trends.

            No, the lead hypothesis has broad data as well, and has implications that can be tested.

    • albatross11 says:

      Scott made a post about racial bias in the justice system awhile back.

      Also, I think Nancy linked awhile back to this Washington Post article by Radley Balko (behind a paywall if you block ads) detailing many reasons to think that blacks and whites are treated differently by the criminal justice system.

    • INH5 says:

      The Rise and Fall of Violent Crime in America by Barry Latzer is very informative. See also this interview in The Atlantic with the author.

      I haven’t read the book, but just from the Atlantic article I see a significant hole in his arguments regarding black migration to the North in the post-war era: any explanation involving factors specific to the black population seems implausible due to the fact that Oregon, where less than 2% of the population was black during the period in question, experienced almost the exact same rise-in-the-1960s and-fall-in-the-1990s trends in crime.

      • albatross11 says:

        I have a half-baked theory that the intensity of white flight and urban decay was largely driven by a spike in crime rates across the board, probably caused by something entirely unrelated to integration but happening at the same time. And I suspect that still poisons some of our discourse wrt race to this day. Blacks were going to have higher crime rates, more poverty, worse school performance, etc., whenever integration happened, but having it happen at the same time as a large spike in crime probably made everything worse….

        • Aapje says:

          This is a good article about the issue.

          The writer argues that black people already had very high crime rates, with southern states having the highest crime rates in 1933, with most of that crime being black-on-black. There was a conference in 1904 in Georgia about black crime which concluded: The amount of crime among Negroes in this state is very great. This is a dangerous and threatening phenomenon. It means that large numbers of the freedmen’s sons have not yet learned to be law-abiding citizens and steady workers, and until they do so the progress of the race, of the South, and of the nation will be retarded

          Note that back then black leaders were also worried about this. W.E.B. Du Bois was secretary of this conference.

          The argument is that urbanization caused many black people to move to northern cities & that the government tried to integrate these migrants into mixed race neighborhoods (perhaps as a Stalinist-like ploy to politically weaken non-WASP neighborhoods by breaking up ethnic homogeneity). At the same time, banks were doing redlining.

          These mixed race neighborhoods didn’t work out because of different culture and due to white anger at black crime, which caused white people to react with slurs & violence against blacks, which in turn created racially motivated violence by black people. With the continued influx of black people and white people having & taking the option to leave for safer white neighborhoods, the remaining whites were then outnumbered. So once white flight started, it became unstoppable. So this resulted in segregation.

          Then presumably, redlining and defensive policies from white people in the neighborhoods were no black migrants had come in, further solidified this, especially as the politicians stopped with their pro-integration efforts in response to white outrage.

          Busing of black people to white schools then caused (more) white flight to suburbs again later on, as we’ve discussed earlier.

          Note that white flight also happened in my country. As migrants came to the big cities, a lot of white people were unhappy with the incoming culture and fled for suburbs and nearby cities. These now most often vote for far-right parties, suggesting traumatization.

          In fact, the common left-wing explanation probably at least in part mistakes the direction of causality, with cultural conflicts between ethnic groups (including, but not limited to anger over crime rates) driving discrimination and such, rather than just the other way around (although it can and does often become a self-reinforcing cycle). If one then looks back at the situation, it is easy to mistake the effect for the cause.

      • Douglas Knight says:

        Oregon and Canada and France and …

  36. johan_larson says:

    You are invited to propose a big-budget summer blockbuster that dramatizes the work of a vital but often neglected part of the military: the logistics corps.

    • albatross11 says:

      So this would be pictures of military vehicles continuing to run because they have enough fuel and spare parts, soldiers able to keep shooting back because they have bullets, soldiers with enough to eat and drink every day, etc.?

      • johan_larson says:

        Yes. But make it exciting. $100 million worth of exciting.

      • Douglas Knight says:

        That sounds like an attempt to illustrate the value of the logistic corps. I think that you could make dramatic movie of failures of logistics that would demonstrate the value. But that wouldn’t illustrate the work of the corps.

    • CatCube says:

      The Red Ball Express would be a place you could start–it’d be something you’d have to have a good writer to make work, but the sleeplessness and fear of attacks from German planes would be possible conflict points.

      • Le Maistre Chat says:

        Agreed. Using Red Ball Express soldiers as protagonists would get you pretty images of US military systems and Nazi German aircraft, auto chase scenes, and a black dudes vs. sky Nazis narrative.

      • Doctor Mist says:

        Been done. Haven’t seen it, so I don’t know if it was any good.

        Larry Gelbart and Gene Reynolds tried a TV comedy along the lines of M*A*S*H; while it gets good ratings on IMDB, I gather it was not a success.

    • metacelsus says:

      Maybe transport flights from India to China in WW2?

      https://en.wikipedia.org/wiki/Air_Transport_Command