Open threads at the Open Thread tab every Sunday and Wednesday

Bottomless Pits Of Suffering


A friend on Facebook recently posted the following dilemma, which of course I cannot find right now so I have to vaguely quote my recollection of it:

Would you rather the medieval Church had spent all of its money helping the poor, rather than supporting the arts. So that maybe there were fewer poor people back in medieval times, but we wouldn’t have any cathedrals or triptychs or the Sistine Chapel?

I was surprised to see so many people choosing the cathedrals. I mean, I guess this question is kind of unfair, in that it’s really hard to figure out what it means, moral value wise, for there to have been less suffering in the past. This is especially true if you choose to believe Robin Hanson – as always a decision that starts a mini-civil-war between the rational and intuitive parts of my brain – when he says we should give much more weight the preferences of past individuals.

I think maybe choosing the cathedrals is so appealing because they’re right there, you can touch them, but the starving peasants are hidden all the way in the past where you can’t see them. So it feels like you’re being asked to sacrifice something you really like for something that you would otherwise not have to think about.

This is one of the biggest and scariest problems with utilitarianism. Utiltarianism is at least kind of easy when it’s asking you to trade off some things in your normal world for other things in your normal world. But when it asks you to make everything you consider your normal world unambiguously worse to help some other domain you would otherwise never have to think about, then it starts to become unintuitive and scary.

Imagine a happy town full of prosperous people. Every so often they make nice utilitarian decisions like having everyone chip in a few dollars to help someone who’s fallen sick, and they feel pretty good about themselves for this.

Then one day an explorer discovers a BOTTOMLESS PIT OF ENDLESS SUFFERING on the outskirts of the town. There are hundreds of people trapped inside in a state of abject misery. The pit gods agree to release some of their prisoners, but only for appropriately sumptuous sacrifices.

Suddenly the decision isn’t just “someone in town makes a small sacrifice to help other people in town”. Suddenly it’s about the entire town choking off its luxury and prosperity in order to rescue people they don’t even know, from this pit they didn’t even know was there a week ago. That seems kind of unfair.

So they tell the explorer to cover the lid of the pit with a big tarp that blends in with the surrounding grass, so they don’t have to see it, and then go on with their lives.


The developing world is sort of a bottomless pit of suffering if some First Worlder didn’t expect it to be there. But I think most people do expect it to be there, most people are happy to help (a little), and it doesn’t really confuse or alarm us too much when we are reminded they still exist and still need help.

But what about nursing homes? Most of the doctors I have talked to agree most nursing homes are terrible. I get a steady trickle of psychiatric patients who are perfectly happy to be in the psychiatric hospital but who freak out when I tell them that they seem all better now and it’s time to send them back to their nursing home, saying it’s terrible and they’re abused and neglected and they refuse to go. I very occasionally get elderly patients who have attempted suicide solely because they know doing so will get them out of their nursing home. I don’t have a strong feeling for exactly how bad nursing homes are, but everything I have seen is consistent with at least some of them being very bad.

Solving this would be really expensive – I am perpetually surprised at how quietly and effortlessly we seem to soak up nursing home costs that already can run into the tens of thousands of dollars a year. Solving this would also produce no visible gain, in that bedridden old people are very very bad at complaining in ways anyone else can notice, and if we don’t want to think about them we don’t have to. If we as a country decided to concentrate on decreasing abuse in nursing homes, we might have to take that money away from important causes in our everyday visible world, like welfare and infrastructure and education funding. We would have to take limited Public Attention And Outrage Resources from causes like human rights and gay marriage and what beverages the President is holding while he salutes people. I think everyone agrees it’s a lot easier not to think about it, and nobody can make us.

Prisons are an even uglier case. Not only is prison inherently pretty miserable, but there seems to be rampant abuse and violence going on, including at least 5% of prisoners being raped per year. Every couple of weeks there’s a new story about how, for example, prisoners are gouged on phone bills because someone can do it and nobody is stopping them, or how they’re kept in cells without air conditioning in 110 degree weather in Arizona because no one has any incentive to change that.

Now the reason this is so ugly is…well, a lot of this is due to prison overcrowding. And a lot of people have very reasonably suggested imprisoning fewer people – ending the drug war would be a good start, but the past thirty years have also seen a momentous lowering of the threshhold for imprisoning people in general and a ballooning of America’s prison population. Which is awkward, because the last thirty years have also seen an unprecedented drop in violent crime.

It would be absolutely lovely if this were confirmed to be the result of some very clever policy like reducing lead exposure, or even if Levitt’s theory about abortion were proven true. But the least convenient possible world is that the recent drop in crime is mostly due to the recent rise in imprisonment and the recent lengthening of prison sentences – everybody with even the slightest bit of criminal tendency is already safely locked up [EDIT: strong argument against this]

Think about what a moral nightmare that would be. Sure, you can do something about the bottomless pit of suffering where people are packed together into 110 degree cells and raped for ten or twenty years – but it’s going to raise crime back to the horrible 1990s levels we’re all pretty relieved to have escaped. Or you can just whistle, pretend not to notice, and continue to enjoy nice low-crime 21st century society.

And then there’s a broader worry.

Conservatives like to talk about how much better we all had it back in the 1950s with traditional this and traditional that, and how you can just tell from listening to stories from people of that time. or reading media from that time, that things were a lot calmer and more pleasant.

And the left likes to talk about how we are widening the circle of empathy and bringing in new and finally starting to pay attention to the concerns of downtrodden groups.

What if they’re both right? What if progress since the 1950s has been about opening one bottomless pit of suffering after another, trading off the well-being of the nice prosperous town for getting people out of the pits, and then moving on to another pit somewhere else?

I mean, this is kind of the standard view of history. Except that in the standard view, conservatives tack on “But really, the bottomless pit wasn’t so bad, and the sulfurous flames gave you a nice, warm feeling inside.” And leftists tack on “but in the end, everyone including the people in the nice town benefitted from the increased understanding and diversity this created, so really history was just this series of obvious win-win propositions that everyone was just too stupid to figure out, until now.”

Although there has been a lot of interesting argument against the conservative proposition that things in the nice town have gotten worse since the 50s – some of which I have participated in, it seems important to note that even if the proposition is 100% correct, progress might still have been morally correct.


A lot of the paradoxes of utilitarianism, the things that make it scary and hard to work with, involve philosophers who compulsively seek out bottomless pits and shout at you until you pay attention to them.

Utility monsters are basically one-man bottomless pits.

Pascal’s Wager (or Pascal’s Mugging, if you prefer) splits the universe into a billion Everett branches, then points out that one of these Everett branches is a bottomless pit and asks the others to make sacrifices to help it.

A lot of the addition paradoxes treat a pool of “potential people” as a bottomless pit.

This seems to be the easiest way to break utilitarianism – point to a bottomless pit, real or imagined, and make everyone in the world lose utility to solve it, forever. It’s not always easy to come up with solutions that successfully rule out these problems, while preserving our intuition that we should continue to worry about people in nursing homes or jails.

Contractualism scares me a little because it offers too easy an out from bottomless-pit type dilemmas. It seems really easy to say “All of us people not in jail, we’ll agree to look out for one another, and as for those guys, screw them”. You would need to have something like a veil of ignorance, or at least a good simulation of one, to even begin to care.

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

458 Responses to Bottomless Pits Of Suffering

  1. suntzuanime says:

    Isn’t “All of us people not in jail, we’ll agree to look out for one another, and as for those guys, screw them” basically the ethical system that obtains? I mean, there’s some ambiguity as to where you draw the reference class boundaries, so you find people who care very much about the welfare of prisoners, slaves, fetuses, livestock or rocks with faces painted on them, but there’s nothing stopping you from drawing the boundaries with them outside, and for many categories, most people do. It seems kind of weird to point to the huge problem with an ethical theory being that it produces the sort of ethical system we observe most people to follow.

    • Grognor says:

      Do you expect that people just naturally follow the One True Morality?

      • suntzuanime says:

        I deny the meaningfulness of the question.

        • MugaSofer says:

          Then why discuss it at all?

          What possible use could we have for an ethical theory that tells us to do exactly what we were going to do anyway, biases and all?

          • suntzuanime says:

            Hmm, I dunno, what possible use could anyone have for a theory that tells them exactly what people are going to do. Sure seems useless.

          • peterdjones says:

            That would be a minimally useful
            descriptive theory, but Mugasofer was asking about ethical, and therefore normative., theory. Normative theories that echo back what you were going to do are pretty useless.

          • suntzuanime says:

            To the extent that there is a theory that can be constructed based on reality, it will have to be a descriptive one, not a normative one. You can’t derive an Ought from an Is. All ethical theories are equally “true”, it’s just that some are contingently more agreeable to human beings than others.

            Possibly where we’re getting hung up is the difference between endorsing an ethical system and holding an ethical system. Humans are hypocrites, and will say one thing while doing another, if they can benefit from it. Contractualism, from this perspective, is scary to Scott because it seems like it might make his pretending to care about the downtrodden too transparent. Better to stick to something safer like virtue ethics, where you’re basically *required* to do the sort of things that virtuous-looking people do.

          • MugaSofer says:

            Suntzuanime … enjoyable though Hanson-style cynicism is, you shouldn’t expect anyone here to fall for it. -1 rationality point.

            In any case, our goal structures are a part of reality. A hard part to isolate, but still a real one.

          • peterdjones says:

            For “minimally” read “maximally”.. damn tablet.

          • suntzuanime says:

            -1 point for correct predictions, I can name one poor lost soul here who has in fact fallen pretty hard for Hansonianism.

        • peterdjones says:


          Not all norms are ethical, and you can rigorously attach truth values to many normative claims. Humes argument absolutely does not mean there are not true normative claims. You seem to be thinking of the “…is true” predicate as placing something on the fact side of the fence. It would be better to think in term of a descriptive/normative distinction, with both kinds statements being potentially true.

        • RCF says:

          How is it not meaningful? I can see someone denying that “The One True Morality” is meaningful, but if you take that position, then you won’t expect people to follow it (because it doesn’t make sense to expect something that isn’t meaningful).

    • Carinthium says:

      There is a difference between descriptive and prescriptive ethical theories. There are plenty of online arguments for why utilitarianism is superior to deontology which can be used to say that even though people act that way, they are irrational to do so.

      • suntzuanime says:

        A prescriptive ethical theory is just a descriptive ethical theory of the ethics the theorist happens to hold.

        • Carinthium says:

          Define ‘hold’. Do you mean ‘hold’ as in a description of the moral feelings of the theorist, or ‘hold’ as in believing that the ethical theory is right?

          More broadly speaking, the problem with not having a descriptive-prescriptive distinction is that it leads into the fallacy of “X is a description of my ethical feelings. Therefore I should do X” ignoring arguments such as Coherent Extrapolated Violition and contradicting one’s own long-term self-interest.

          If you disagree with the concept of prescriptive ethics, you should attack that rather than attacking somebody who is clearly attempting prescriptive ethics as if it were descriptive ethics.

        • DanielLC says:

          It’s an ethical theory that an expert holds. That’s the difference between folk physics and general relativity.

    • RCF says:

      So, if one criticizes 18th century ethics for condoning slavery, that would be “kind of weird”, because it would be pointing to the problem being that it produces the ethical system that most people followed?

      • suntzuanime says:

        What happened with the emancipation movement was simply a redrawing of the boundaries, not a change in the system. They freed the slaves and beat the horses, and saw no contradiction in this.

        • RCF says:

          I notice that not only does your post not address my question, it also equates black people with horses.

          • suntzuanime says:

            This is sophomoric debate-team bullshit. To mention two things in the same sentence is not to equate them.

          • RCF says:

            That is quite unkind, so according to blog rules, you need to be sure it’s true and necessary, which it is not.

            Certainly, your literal words did not equate black people and horses. But the implicature of saying “they saw no contradiction” is to assert that there is a contradiction, that beating black people is no more clearly wrong than beating horses is. That you see no difference between black people and horses is an entirely reasonable inference from your post, and I have reported your post for so rudely attacking me for pointing that out. If you’re going to compare black people to horses (or, if you’re going to be really pedantic, beating black people to beating horses), at least have the decency to admit it.

            If we’re identifying sophomoric debate-team bullshit, you know what else fits in that category? Doing X, and then when someone says you’re saying Y, saying “Doing X is not saying Y”, when the other person did not says “Doing X is saying Y”, and is in fact basing their claim that you’re saying Y on facts other than you merely did X.

          • suntzuanime says:

            lmao you admitted I didn’t equate black people with horses, so what? You think it’s kind and relevant to claim I did? Give me a break with your hypocritical citing of the comments policy.

          • MugaSofer says:

            I concur with RCF. If it’s a “sophomoric debate-team” tactic, then it’s one I have accidentally duplicated the results of simply by reading your comments.


          • RCF says:

            I really don’t think that characterizing “your literal words did not equate black people and horses” as me admitting that you did not equate black people with horses is an honest reading of my post. I was clearly making a distinction between the literal, explicit denotation, and the clear meaning. If you insist that only the former matters, then I really don’t want to deal with a racist liar.

        • blacktrance says:

          This being a contradiction is a contentious claim. It’d be a contradiction if the general principle was that beating anything is wrong, but who says that’s the principle?

          • suntzuanime says:

            Yes. The general principle is that beating some things is wrong and beating other things isn’t. This was the general principle before, and all that’s changed is what things fall in what categories. Hence, what happened with the emancipation movement was simply a redrawing of the boundaries, not a change in the system.

          • blacktrance says:

            If the general principle is “beating some things is wrong and beating other things isn’t”, and we set aside the issue of how to figure out what things it’s okay to beat, there really is no contradiction – it’s an internally consistent principle.

          • suntzuanime says:

            Yes exactly! That’s my point!

          • RCF says:


            You are simply identifying one thing that the two situations have in common, and claiming that that makes them identical. It’s like saying “bats and lizards are both animals, therefore there is not difference between them”.

          • suntzuanime says:

            I never claimed that makes them identical and I wish you’d stop saying I did, since you know it’s false.

        • ADifferentAnonymous says:

          I think what RCF was trying to say before getting distracted is that the drawing of the boundaries is really important. This post is all about where you draw the boundaries. Arguing that all theories with ‘protected ingroup, unprotected outgroup’ structure are basically equivalent is almost like arguing that all theories where you maximize a utility function are basically equivalent.

  2. Douglas Knight says:

    Lots of smart people lying about crime is no excuse for you to promote their lies. Canada had exactly the same crime trend as America – doubling 1965-1975, halving in the past 30 years – and didn’t change its prison rates at all.

    Although promoting Levitt’s theory, which is pure fraud, is even more reprehensible.

    • suntzuanime says:

      That doesn’t make it a lie necessarily. It’s suggestive, but Canada could have done something else that worked but wouldn’t be replicable in the States. This is where I’d usually put a comedic hypothetical example but I don’t know enough about the sociology of Canada to do so.

      Also “lie” requires an element of intent usually, while the smart people may simply have been unaware of your Canadian data or unpersuaded by it.

      • Douglas Knight says:

        It is important to shout about Canada, so that people cannot lie about their ignorance.

      • RCF says:

        “It’s suggestive, but Canada could have done something else that worked but wouldn’t be replicable in the States.”

        What Douglas Knight is responding to is the claim that lower crime rates is evidence for the hypothesis that incarceration works. You are confusing “X is an argument for Y” and “Y is true”. The fact that Canada could have done something else is an argument for “incarceration works”, it is not an argument for “the lower crime rates show that incarceration works”.

        This is a very subtle Dark Arts tactic. As far as I know it does not have a name, but perhaps it should.

        (Note: by calling it “Dark Arts”, I am not accusing you of intentional malice, only saying that this is fallacious reasoning.)

        • Douglas Knight says:

          Yes, I could have been clearer.

          Also, if incarceration works and Canada did something else, then we are not in the inconvenient world in which incarceration is the only option.

        • Thomas says:

          I see two arguments:
          1) There are fundamentally dangerous people in the world, and failing to account for criminality and evil is as foolish as failing to account for hurricanes and earthquakes.
          2) Mass incarceration was *an* effective method of reducing the observable impact of human criminality and evil.

          Canada’s dropping crime rate without rising incarceration implies that the United States (being geographically close and economically comparable) might have benefited from a similar trend, and the mass arrests did not reduce the number of murders: they are merely an uptick in police violence that fills the gap left by the reduction of civilian violence.

          Or it is evidence that social programs and relative racial homogeneity are effective remedies to inherent criminality, but mass incarceration is also equally effective.

          Or it is evidence that the United States and Canada are fundamentally different countries, and the effective remedies for criminality will likely be different in each country.

          Or it is evidence of a bias in definition and reporting: violence is violence, and why should anyone care whether you call it ‘crime’ or not? A vigilante killing the man who killed his brother leaves two murders on the books; the state of Arizona executing the murderer leaves one. Police who shoot unarmed men don’t get counted at all; men who shoot police often get double-counted as terrorists. If you add the incarceration rate to the kidnapping rate, the United States crime rate has not really shown a meaningful downtick.

          “Crime rates are negatively correlated with incarceration rates in the United States” is absolutely strong evidence for “higher incarceration rates lead to lower crime in the United States”
          “Crime rates are not negatively correlated with incarceration rates in Canada” is absolutely strong evidence against “higher incarceration rates lead to lower crime in the United States”
          “Despite their similarities, The United States and Canada are very different countries, and it is prudent to give lower weight to evidence which comes from Canada when it comes to American law enforcement policy” is a good argument against “People are just lying when they claim that mass incarceration reduces crime rates because Canada didn’t see any correlation between incarceration rates and crime rates over the last 50 years”

      • DanielLC says:

        It’s awfully suspicious for both methods to work equally well. Also, we have no evidence that incarceration helped, as opposed to, say, making things worse. Maybe Canada did something else that also made things worse.

    • Scott Alexander says:

      Thank you. I wasn’t aware of that.

      • CaptainBooshi says:

        It’s not just Canada, either. There’s been a huge increase and then drop in crime across much of the first world. At least, I know it’s happened in much of Europe, Australia, and Japan.

        I’m actually a bit surprised you didn’t know about this, because I thought that was one of the most impressive bits of the lead-causing-crime hypothesis, and you have clearly read a fair bit about that. If I recall correctly, the relationship between lead and crime held true in every country, despite however many other differences those countries had.

        • Ronak M Soni says:

          From the parallel discussion on Levitt I went to the wiki page, where I found the statement

          if abortion rates cause crime rates to fall, crime rates should start to fall among the youngest people first and then gradually be seen lowering the crime rate for older and older people. In fact, they argue, the murder rates first start to fall among the oldest criminals and then the next oldest criminals and so on until it last falls among the youngest individuals.

          If it’s true, doesn’t it also refute the possibility that lead caused the drop?

          • CaptainBooshi says:

            I wasn’t able to find the specific paper being mentioned, so I can’t really address the argument being made. I think the claim is that the big crime drop over the last few decades started among the oldest and then trended younger. Is that really what they’re saying? Are they taking into account that the amount of criminals in a particular age group always drops as that group ages?

            Also, I saw just in the last few months data which indicated in the past few years the only age group which saw a relative increase in incarceration is people in their middle ages, which would indicate a particularly violent generation working their way through the system, so I’m not sure the claim they’re making is even true.

            The lead-crime argument is also different in progression than the abortion argument. The change in law for abortion was one sudden change, whereas lead was a much more gradual change, both in it’s increase and decrease in use (because of the multiple ways lead was being used).

            Basically, if their claim is true, it may be evidence against the lead-crime theory, but there’s enough difference between the two theories I’m not entirely sure, and I’m also not sure that their claim is actually true.

          • Douglas Knight says:


            I think that the wikipedia description is misleading and that they mean “young” and “old” only in a very narrow sense. They claim that Levitt’s analysis is driven just by the crack epidemic, but that he mislabels the ages of the those murderers. They were actually the cohort right after legalization, so they should lead to the opposite conclusion.

            Thus this says little about the lead theory, as Booshi says. I do think that proponents of the lead theory wrongly try to take credit for the decline of the crack epidemic, but the longer term trend is more plausible. It would be good to apply this methodology to the lead data.

          • Army1987 says:

            [massive speculation] Maybe adults who can’t afford to maintain children by legal means and aren’t allowed to abort will resort to crime to maintain their children?

        • Ryan says:

          I can never find proper information on the following:

          Do the studies control for age of the population?

          If not that seems like the most obvious explanation. Who are criminals? Men age 15-35, for the most part. What happened in all of those countries? After WW2 ended the GIs came back, got married, and pumped out babies. And then from the 60’s to the 80’s the men born in that era committed a lot of crimes. They didn’t have nearly as many children as their parents, and thus the next generation didn’t manage to produce as many crimes.

          • Douglas Knight says:

            Right, we want to screen off the best predictors, so the correct metric is crimes per (male) youth. Complicated analyses don’t use it. This is terrible. But if you just want to eyeball the data, you can find it, or at least “age-adjusted” rates. It doesn’t make that much of a difference.

    • g says:

      I hadn’t heard about Levitt’s theory being “pure fraud”. Tell us more?

      (I had a quick look at the Wikipedia page about the theory, in the hope that if it’s widely thought to be fraudulent there should be some mention there of why, but nothing there seems to indicate it’s likely to be fraud.)

      • Douglas Knight says:

        As wikipedia says, it was a programming error. It goes on to say:

        The response acknowledged the mistake, but showed that with different methodology, the effect of legalized abortion on crime rates still existed.

        The arrogance of doing data mining in public is breathtaking.


        An unrelated point: Sailer pointed out an error at the time, an error that invalidates not just Levitt but almost all cohort analysis of crime (including lead), that the crack epidemic murders were committed by much younger people than in other times.

        • Ilya Shpitser says:

          This particular topic is not a mindkill kind of topic for me, and furthermore it would not surprise me if Levitt hypothesis was simply wrong.

          However do you think Levitt’s response to problems in his methodology unusual? This seems like what academics generally do in this situation — correct the problem, rerun the analysis, and hope the original conclusion still holds. Would you prefer people just retracted papers instead?

          It does not seem entirely fair to jump on Levitt for doing this.

          • Douglas Knight says:

            Maybe all academics are frauds. I’m willing to accept that. I’m just not willing to sit by while people pretend that Levitt is better than Marc Hauser. Yes, I would like Levitt to retract every paper he has ever written.

            The right thing is to correct the code, rerun the analysis and accept the result. No, people try to get access to the truth, not hope that their lies survive sunlight. He almost-but-not-quite admitted that the effect is not there in the original data, but pulled out new data and new code that no one has read, and that gets the result, and declares that the original data doesn’t matter.

          • Ilya Shpitser says:

            Ok, I updated appropriately, thank you.

          • g says:

            (this is actually a response to Douglas.)

            So I had a quick look at the thing by Levitt and Donohue that’s referenced by the Wikipedia statement about “… with different methodology …”.

            What they claim is: (1) just fixing the programming error leaves the results still there and still significant, but with reduced magnitude; (2) if they also measure arrests per capita rather than total arrests, as proposed by their critics, the estimated effect drops enough to lose statistical significance; (3) if in addition to this they also make what they allege to be methodological improvements, then the estimated effect gets big and significant again.

            Is making methodological improvements here data-mining and cheating? I don’t think so, for two reasons. (1) It’s not cheating because the issue only arises in response to a methodological change proposed by their critics (looking at per-capita rather than total arrests). (2) It’s not data-mining because they didn’t (obviously, at least) cherry-pick some tweak designed to maximize the effects, they used an alleged improvement they already suggested (giving reasons) in an earlier publication.

            To summarize. The critics said: There’s a programming error, and also you made a methodological mistake by looking at total rather than per-capita arrests, so that the (boring) effect of abortion on cohort size is part of what you’re measuring. L&D said: Fair cop on the programming error, but while fixing it reduces the estimated effect sizes it still leaves them significant at the level we claimed. Fair cop, also, on the methodology, and indeed when we fix that the effects get smaller and distinctly less significant. But if we’re changing that then let’s also take the opportunity to swap in the more accurate abortion estimate we proposed in an earlier paper so that we’re not trying to detect correlations with something that’s mostly noise; when we do that, it turns out that the effects are once again large and significant.

            That doesn’t seem terribly unreasonable to me, and in particular nothing here seems to warrant calling L&D’s theory “pure fraud”.

          • Douglas Knight says:

            so that we’re not trying to detect correlations with something that’s mostly noise

            In other words the original paper was purely due to two fortuitous errors. I say data mining and I say fraud.

            I was wrong to say that they pulled out the new data in response to these critics. As you said, they used the new data in an earlier paper. Yes, it is reasonable for them to use the new data and say that this criticism invalidates the analysis in first paper but only weakens the analysis that half-appeared in the second paper.

            But the analysis only half-appeared. I believe that the second paper simply substituted the new data into the old model. But in this response, the analysis is more complicated. The complications seem well-motivated, but they introduce new parameters for tweaking.

            the issue only arises in response to a methodological change proposed by their critics

            Why does that excuse it? At first I thought you were saying that the critics proposed a problem with the metric of crime and while that parameter was up for grabs, they suggested an even better one. But, as you say, they respond to criticism of their crime metric by changing their abortion metric. “Change on negative results by critics” is “change on negative results.”


            If someone has been caught spinning noise into gold, can he be rehabilitated and his future results trusted? I think this was Ilya’s question. This is a difficult problem. I am extremely suspicious of the same person trying to establish the same result that was originally noise.

          • g says:

            Why does that excuse it?

            Because if you’re fiddling with the methodology anyway (for the excellent reason that someone else has pointed out something wrong with it) it seems pretty defensible to fix all the methodological problems you know how to fix, not just the one pointed out by your critic.

            In the present case, L&D said what result they got by making the methodological change (total -> per capita) recommended by F&G; then they said what result they got if they also made the methodological change (better abortion proxy) recommended by their own earlier paper.

            I’m having trouble seeing how that’s cheating, let alone “pure fraud”.

            I repeat: Of course it’s possible that in fact L&D were simply lying, or falsifying things in some subtler way (e.g., trying lots of adjustments to the models and then publishing whatever gets the best-looking results). So, hypothetically, what do you think researchers should do in the following situation? They think of an interesting new hypothesis; they investigate it crudely and find some evidence for it; both they and others find flaws in their investigation, influencing their results in both directions; when they fix all those flaws the original finding seems to remain (though if they fix only the flaws that made their results better, and not the ones that made them worse, the original finding goes away).

            Because that seems to me to be the situation if they aren’t simply lying, and it doesn’t seem a priori a very implausible situation, and it seems to me that if that’s what happened then what L&D did was a pretty reasonable response. But maybe I’m missing something.

        • peterdjones says:

          Data mining is perfectly legitimate, Presumably, you mean something like cherry picking,

          • Douglas Knight says:

            I’m sorry that you don’t know anything, but that’s not my problem.

          • Matthew says:

            I was genuinely puzzled by your data mining comment above until seeing peterdjones’ comment. He’s not being pedantic; he’s providing a useful clarification. Your reponse is neither true, necessary, nor kind and has been treated accordingly.

            ETA: Rude-but-not-profane version still fails all tests, and still reported.

          • Douglas Knight says:

            Matthew, my comment provided a lot more information than Peter’s. I hope you are capable of learning from it, but I doubt it.

          • David Barry says:

            I usually interpret ‘data mining’ to mean the analysis of large datasets to find real relationships between variables. But I have seen economists (e.g., here) use ‘data mining’ to mean something like “trying different regressions and analyses until you find one that gives you the answer you want”.

            I think this unfortunate clash of definitions is the disagreement between Douglas and peterdjones.

          • MugaSofer says:

            I, too, interpreted “data mining” to mean “sifting through the data until you find something that fits your conclusion”. Clearly this is an ambiguous term.

          • Douglas Knight says:

            It’s not a clash of definitions, but a valuable red flag when people self-describe as data miners.

          • Emile says:

            Douglas Knight: just to be clear, when you talk about “Data Mining” you mean what’s covered here, right?

            (because you seem to be talking about something else AND insisting your definition is the best AND insulting people who are used to the primary definition AND not making much of an effort towards being either clear or charitable)

          • Rand says:

            Douglas: Here are the Google Scholar profiles of Christos Faloutsos (CMU), Jiawei Han (UIUC) and Jure Leskovec (Stanford), three of the foremost researchers in what may broadly be termed Artificial Intelligence:




            All three identify their specialty as “Data Mining”.

            I think you want to backtrack slightly, maybe apologize and say what you intended in unambiguous language.

          • RCF says:

            “(because you seem to be talking about something else AND insisting your definition is the best AND insulting people who are used to the primary definition AND not making much of an effort towards being either clear or charitable)”

            peterdjones said “Data mining is perfectly legitimate”. Such a broad statement, without any qualification, strongly implies a rejection of any meaning of than his own, so I think that your accusation of “insisting that your definition is the best” is misplaced. And the issue was not merely that peterdjones was “used to” what you allege is the “primary” definition, but that he considered it to be the only definition. So, while you have a point as to DK being more rude than necessary, peterdjones’ post was ignorant and arrogant.

            For an example of the original meaning of “data mining”, see

    • Anonymous says:

      Is that the case? This chart appears to show violent crime staying pretty steady (or in fact slightly increasing) for Canada since 1980 but increasing then decreasing significantly for the US (presumably decreasing alongside the incarceration explosion in the 90s):

      • Douglas Knight says:

        For precision, I should have said “homicide” (source) but I just don’t trust other statistics.

        Your chart cuts off 1975-1985, when Canada saw a large decrease in murders. (So I should have said 40 years, not 30 years.) Kieran Healey has many countries. One difference between America and others is the crack epidemic. KH’s loess curve smooths that out. But the way the crack epidemic moved from city to city makes it pretty clearly exogenous.

        • Anonymous says:

          So the things you said, (after accusing others of lying), turn out not to have been true, but hey, your larger point still holds, you can come up with some other data than what you used to support it! Someone up thread might have called it “data mining”, not that that’s what that term actually means.

    • Anonymous says:

      Couldn’t this important thing be said without the bile?

    • xhxhx says:

      Canada’s adult incarceration rate increased 52 per cent between 1960 and 1995. Not as dramatic as the trend in US incarceration rates, which quintupled, but not ‘no change’ either.

      I am otherwise in complete agreement with Douglas Knight.

      • Douglas Knight says:


        vs 5x for America.

        I think I had only seen data starting in 1975 (at the homicide peak) and rounded down the 25% increase to nothing. I don’t know what to make of the 20% increase 1960-1970. The three periods 1960-1975 are weird. The first is before the crime wave and Canadian incarceration increases, while American falls. The second is the beginning of the crime wave. America stays flat, while Canada rises. That is a natural response to the crime wave, but it doesn’t seem to have helped Canada, which had the same 2x homicide increase over that period and the next. In the third period, both are flat.

        • xhxhx says:

          Statistics Canada’s current series on adult incarceration only goes back to 1978. You might have been reading something that relied on it. The Historical Statistics of Canada has a series that goes back to 1870, but I wasn’t able to access it.

          This comparison study uses the finer-grained incarceration figures from the Historical Statistics. It makes the interesting point that Canada, England and Wales, and the United States all had roughly equal incarceration rates in 1970. And then the United States broke away from the pack.

          The chart on p. 168 is really something. Jesus. That chart, when read with the charts of homicide rates later on, more or less make your point for you.

  3. Carl Shulman says:

    “But the least convenient possible world – and one I have heard a lot of smart people support – is that the recent drop in crime is mostly due to the recent rise in imprisonment and the recent lengthening of prison sentences – everybody with even the slightest bit of criminal tendency is already safely locked up.”

    The fall in crime happened worldwide. Incarceration may have contributed in the US, but most arguments that it was dominant are parochial and don’t account for the broader picture.

  4. Andy says:

    Contractualism scares me a little because it offers too easy an out from bottomless-pit type dilemmas. It seems really easy to say “All of us people not in jail, we’ll agree to look out for one another, and as for those guys, screw them”.

    Yep. Or any kind of “you’re not in our ingroup, so we don’t care about you.” Reminds me of a This American Life story a few weeks ago about a public school board taken over by Orthodox Jews, who gut public schools in order to free up more funding for yeshivas, and they’re so well-organized that it’s impossible to beat them electorally.
    Link And the Jewish school board members asserting anti-Semitism every time someone disagrees with them, which lowered the cost for genuine – and violent – anti-Semitism for emerge with a vengeance. So now the Jewish school board members – which is every single one by now – have absolutely no incentive – nothing to gain – from listening to or negotiating with the gentile majority.
    EDIT: I don’t think this had anything to do with them being Jews, before this devolves into a terrible terrible argument, I merely bring it up as evidence of what happens when a group has a strong in-group/out-group mentality and has to live in a much bigger place. I’d also look at Chinatowns as another example, or Catholics in the late 19th/early 20th century sending their kids to private school to keep them Catholic, and the Protestants of the same period trying to get Catholics into public schools to turn them into Protestants.
    I think this is kinda what scares me about Libertarian schemes of the “step 1, privatize everything” brand – if one in-group takes control of critical infrastructure, they can let a bottomless pit of suffering form wherever they want, and if you’re in one…. you’re screwed if they decide to turf it over rather than address it.

    • Anonymous says:

      You give an example of the tyranny of the majority, and then you say that the free market scares you?

      • Andy says:

        Yep. Because the free market can be dominated by the majority, and all the costs for infrastructure et al can be shoved off onto a despised minority, while blaming them for any problems that crop up.
        This link lays out, in an unfortunately fnord-ridden way, what happens when a semi-libertarian intellectual movement can privatize a justice system and push its costs onto the people who use it – which ends up being the people least able to pay the court fees whenever the court system could use some extra revenue.
        Edit: Fixed link

        • Anonymous says:

          I don’t know what you’re talking about because you left out the link, but the justice system is very different because it is all about externalities, and almost nothing to do with a consumer. Versus schools, where maybe the externalities make you demand subsidies, centralization, and brainwashing, but where from the point of view of the parent, the free market provides what they want. Which is very well demonstrated by your example of the Hasids.

        • cassander says:

          >Yep. Because the free market can be dominated by the majority, and all the costs for infrastructure et al can be shoved off onto a despised minority, while blaming them for any problems that crop up.

          That is not how markets work, at all. if 55% of people like red cars and 45 like blue, there is still a ton of money in selling them blue cars. As for shoving the costs of maintaining society onto a minority, that sounds like a political process, not a market one.

        • MugaSofer says:

          To give a concrete, if extreme, example: the majority could buy guns and force the minority to build their infrastructure by hand.

          Or, if that was illegal, they could buy votes and have the army do it for them.

          Or, if that was somehow also illegal [?!] they could threaten to boycott the minority – selectively raising prices on minority goods, for example – in order to price-gouge the minority on the contract for the minority to build them infrastructure.

          (Obviously, in all these examples the PR firms would not phrase it like that.)

          Oh, hey! And now that I think about it; that’s a majority of currency, which need not correspond to a majority of people. Especially since a small starting advantage can snowball into a large one by opening up investment opportunities.

          • RCF says:

            I really don’t think that “A majority might vote in a government that oppresses the minority” is a valid criticism of libertarianism.

      • Mary says:

        The thing about the free market is that it’s a lot harder to tyrannize the minority if you must go after thousands if not millions of decisions. If you want to stop a product being sold in the free market, you have, say, buy up all the supplies of a necessary component, or all the manufacturers, or what have you, and people can learn to synthesize that or start a new factory.

        • peterdjones says:

          You can have markets in slaves, so economic freedom is no guarantee of personal freedom.

          • RCF says:

            What you wrote is blatant bullshit. If you had instead written “Economic freedom of slave owners doesn’t guarantee personal freedom of slaves”, that would be true, but insanely vacuous.

    • pneumatik says:

      Civilization is what happens when everyone chooses to cooperate in the real-world iterated prisoner’s dilemmas that we face every day. When people start defected, either as individuals or groups, eventually is falls apart.

  5. BenSix says:

    …philosophers who compulsively seek out bottomless pits and shout at you until you pay attention to them…

    Indeed. The conservative has no inherent objection to opening up bottomless pits but would maintain that the ground is unstable and if one pokes about under it one might cause an earthquake.

    (This might be to take the introduction too seriously but one could choose the cathedrals on unselfish grounds, believing them to have been important to the formation/maintenance of civilisation – and, indeed, a source of awe and inspiration to men and women over centuries.)

  6. DrBeat says:

    I think the issue of prisons is less about a bottomless pit of suffering that we can ignore or not ignore, and more about how impossible it is to solve a problem with a solution that is contrary to people’s immediate emotional impulses. People think criminals are Bad, and thus they should be hurt; if you have a solution that says things would be better if we stop hurting criminals so much, this means you are Bad.

    The suffering of people in prisons isn’t an isolated suffering pit because A: most of our population is aware of and actively encourages this suffering at every opportunity they have to do so by constantly supporting any measure that claims to be “tough on crime”, and B: these prisons are incredibly efficient at turning human beings into career criminals who will never create anything but more suffering, and then unleashing them on the rest of the non-pit-occupying world.

    • peterdjones says:

      Touch of typical country fallacy there, I suspect. There are countries that have reformative justice systems. Punitive regimes aren’t inevitable, they require a certain ecosystem of mass media, religion and so on to push opinion in that direction.

      • DrBeat says:

        He was talking about American prisons, so I was talking about American prisons.

        Reform-minded prisons were put in place by populations that didn’t have that immediate emotional reaction of disgusted anger and desire for proxy violence, at least as far as I can tell.

        • Someone says:

          Do you think the prison-makers (presumably various elected officials) in some countries are more frightened by crime than in other countries? Why?

          I assume that US lawmakers are about as protected from crime as, say, Swedish ones. (Though I supposed Gabrielle Giffords is a counterexample.)

          • DrBeat says:

            Uh… Yes, obviously? Fear is irrational to begin with. Why on Earth would you expect people to have measured, proportional amounts of fear? Have you ever met anyone?

          • Anonymous says:

            Fear is irrational? all fear?

            Don’t be silly. Fear is a logical reaction to sources for harm.

        • peterdjones says:

          If someone else has climbed out if a pit, or not fallen into into it, it was never much if a pit.

          Scott does thus about since a week…complaining about the terrible inevitability and dreadful unfixability of problems that only exist in the US,

          • DrBeat says:

            The people making the prisons aren’t the people in the pit. I don’t think you got the analogy.

            And prisons are not horrible worldwide and unfixable worldwide. But prisons in America are horrible and unfixable. The fact that there are prisons in other places that are not horrible and unfixable doesn’t change the fact that America’s prisons are; it is not an axiom that all prisons ever forever are endless pits of suffering, but the ones in America are.

          • RCF says:

            “If someone else has climbed out if a pit, or not fallen into into it, it was never much if a pit.”

            That’s an incredibly idiotic and offensive thing to say.

            “Scott does thus about since a week…”

            That’s not valid English.

            “complaining about the terrible inevitability and dreadful unfixability of problems that only exist in the US,”

            No, he doesn’t.

          • Shenpen says:

            Only in the US? In pretty much all of Latin America and the “third world” in general.

            Rather the issue is that people make this mistake to put the US and Denmark into the same category, call this category “first world” just because both nations are rich, and expect things to work the same way.

            But in reality being rich or poor, developed or developing is simply not a proper way to categorize nations, because even large parts of the “first world” were poor as fuck 50-100 years ago (think Norway before the oil or Finland before Nokia or the Appalachia part of the US).

            I would say the US has very much the same problems as Brazil, because the US _is_ very similar historically, culturally to Brazil, just much richer.

            This does not mean they both are somehow “worse” than say Denmark, they are just different. I would say, they are more chaotic and more get-go, both in good things (entrepreneurialism, inventing things, having fun) and bad things (violent crime, political corruption, inequality).

          • peterdjones says:


            There’s no fact that America’s prisons are unfixable. Legislation can be reversed.

          • peterdjones says:


            I half agree with you. I don’t think all first world countries are equivalent, and I do think capitalism is more than one thing … I generally tend to agree with sentences of the form “x is more than one thing “.

            I wasn’t aiming at the connotation “Ha ha, you’re bad, we’re good”, I was aiming at the connotation “You’re too pessimistic about the prospect of fixing some of the the things you don’t like”.

            Incarceration rates in particular aren’t some inevitable downside to an innovative, highly active economy. They were largely brought about by specific pieces of legislation.

      • RCF says:

        “There are countries that have reformative justice systems.

        I’m sure that if I ask you to defend that statement, you’ll resort to motte-bailey nonsense.

        “Punitive regimes aren’t inevitable,”

        They are if a society isn’t suicidal.

        • DrBeat says:

          Prisons in Scandanavia don’t seem to be that much of a rhetorical bailey. I mean, they are there. We can see them.

          You may be assigning your own connotations to “reformative” and “punitive”, but while the dude is wrong about their significance to the argument, reformative prisons that are not suffering pits do exist. Scandanavia is the archetypal model, where prison is a matter of “taking a non-functional member of society, and making them a functional member of society”, rather than “taking someone who has caused suffering, and causing them to suffer.”

          Severity of punishment has very, very little impact on crime, because criminals are not making rational calculated decisions, they are making dumb impulsive decisions. CERTAINTY of punishment has a much greater impact; ie, it matters if they think they can get caught, not what happens if they get caught.

          • Chris H. says:

            Oddly (or not, if one takes politicians’ incentives into account), this is probably far less true of white collar crime, which is far less likely to receive draconian sentences. I stand by the view that if we’re going to execute anyone, we should start with white collar criminals whose crimes (a) cause 1 or more fully non-consensual human deaths, either individually or statistically (e.g. through approving unlawful chemical dumping that causes cancers) and/or (b) cause financial damage that exceeds the actuarial value of a median American human life.

          • Ken Arromdee says:

            All things done on large scales have the potential to cause large numbers of deaths, and we don’t think it’s morally significant in those contexts. Building a shopping mall probably causes a number of deaths from people driving to the mall, and maybe causes a reduction in deaths from helping the economy. Depending on which factor predominates, either building it or not building it makes someone die.

            True, you’re only talking about illegal activity, but in practice, what you suggest would mean that any sort of illegal large-scale activity would be treated as equivalent to murder. 10 people run 10 illegal protests, misdemeanors. 1 person runs 10 illegal protests, it’s life in prison for doing economic damage equal to one life. Also, the larger the company you’re protesting, the more damage your protest does, so the more likely you are to cause damage that goes over the economic threshhold.

          • DrBeat says:

            “Oddly (or not, if one takes politicians’ incentives into account), this is probably far less true of white collar crime, which is far less likely to receive draconian sentences.”

            Why would you believe this? White collar criminals believe they will not get caught, and make efforts to conceal their crimes, even if the punishments are a slap on the wrist. That seems perfectly in line with the pattern of severity of punishment having little effect, and certainty of punishment having a lot of effect.

          • Anonymous says:

            @Ken: I should probably re-emphasize “white-collar crime,” by which I mean essentially (a) fraud, embezzlement, naked shorting, and the like and (b) willful complicity in corporate crime.

            — Chris H, who forgot to log in

          • Chris H. says:

            @DrBeat: I think that the level of impulsivity involved in the typical mortgage fraud, say, which requires multiple purposeful actions over a period of days or weeks, generally performed while sober and collected, is significantly lower than the level of impulsivity involved in the typical drive-by or convenience store robbery gone wrong. White collar criminals are in general at least attempting to be rational, albeit often rather poorly. A few public hangings of wayward stockbrokers might at least shift their cognitive biases in a more useful direction.

          • RCF says:

            “You may be assigning your own connotations to “reformative” and “punitive” ”

            No denotations have been presented to me.

            “reformative prisons that are not suffering pits do exist.”

            Not being suffering pits is what I would classify as the motte. Certainly, there are countries with prisons better than those of the US, but no country’s prisons are perfectly reformative, or perfectly non-punitive.

            “Severity of punishment has very, very little impact on crime”

            Is this assertion based on empirical fact?

            “because criminals are not making rational calculated decisions, they are making dumb impulsive decisions.”

            That wording treats “criminals” as a remarkably homogeneous group. Criminals quite often make choices based on severity of punishment, such as not taking a gun with them when engaging in burglary, because they can get simple burglary pled down to a misdemeanor, but taking a gun makes it a felony (especially if they use it, and why bring it if you’re not going to use it?)

          • Matthew says:

            I think the semantic divide here over the term “punitive” may revolve around whether you consider “forcibly separating someone from society” to be punitive. I’m pretty sure what DrBeat means by “punitive” is “inflicting suffering specifically for the purposes of retribution, over and above what is required merely to protect society or to condition better behavior in the prisoner.” Scandinavia genuinely appears not to do this any more. You may wish to take a look at some of the articles that came out after the Breivik murders on the conditions under which he (as a mass murderer, presumably receiving the worst treatment their system offers) is imprisoned.

          • Ken Arromdee says:

            @Ken: I should probably re-emphasize “white-collar crime,”

            It seemed like you said white collar crime because you wanted all crimes to be treated on par with each other, and thought that white collar crimes being treated too leniently is an example on that. If so, then it applies to non-white-collar crimes for which enforcement is lax as well as white collar crimes.

            If, in fact, you just wanted to selectively increase the punishment for white collar crime and did not care about treating crimes on par, then it doesn’t cover protests, but I find it questionable on an even more fundamental level (and it didn’t sound like you were asking for that, anyway)..

            And we just don’t treat loss of life this way in other contexts. Every time the president makes a bad policy decision that causes $X million of damage to the economy, that is not equivalent to a homicide.

  7. Carinthium says:

    One possible solution to the problem is to embrace selfishness. It’s far easier to act selfishly in life whilst pretending to care than it is to open so many bottomless pits. Ergo a person aiming the former is likely to come closer to perfecting it than they are the latter. It’s also a philosophically consistent posistion.

    Noting against potential arguments that although there are plenty of arguments for believing in morality, those tend to be of the form of ‘It is tenable to believe in morality’, NOT the form ‘There is something rationally wrong with you if you believe there is no such thing as right and wrong.’

    • peterdjones says:

      What problem? The problem of finding resources to deal with bottomless pits? The problem of motivating oneself? The problem of fitting them into a moral framework?

      One of these days I might meet someone who doesn’t believe in right and wrong as applied to themselves, but it hasn’t happened yet.

      • Carinthium says:

        The problem that when bottomless pits exist a person is torn between their morality and the fact they don’t want to destroy their own lives in the name of moral rightness.

        As for your second claim- that’s an ad hominem. The theory I use is that belief in moral right and wrong is a cognitive bias which should be classified like any other cognitive bias. Rationalists often have tendencies to form beliefs in certain ways through cognitive bias yet resist them through their sense of rationality. If that isn’t inconsistent in the sense that matters, why should resisting a sense of right and wrong be?

        This is equally so for a person in regard to themselves and in regard to others.

        • peterdjones says:

          You say “ad hominem” like it’s a bad thing. If a certain opinion is never held sincerely, and cannot be held sincerely, that is a problem.

          > why should resisting a sense of right and wrong be?

          You need to establish that right and wrong actually are biases.

          • Carinthium says:

            The question here is this- are we discussing what a person ‘pragmatically should’ believe, or what is true? If we are discussing what is true, then what humans think or are capable of thinking is irrelevant. Hence why ad hominem is a fallacy.

            I should also point out that psycopaths and sociopaths are highly plausible candidates for not believing in right and wrong, though without both empirical evidence and long philosophical consideration it’s hard to figure it out for sure.

            As for why belief in moral right and wrong are cognitive biases, I’ll give several arguments from different POVs:

            1- ASSUMING moral right and wrong do not actually exist, there is no way to classify them except as biases which makes any sense. This may sound obvious, but I want to make sure this point is very clear in case you or somebody else tries to bring up some other categorisation.

            2- Just world bias is a known phenomena- a tendency to believe in something which clearly does not exist (a morally just world). This shows there is already plenty of precedent for moral right and wrong as a bias.

            I should also point out that many biases involve struggling against a systematic false assumption. Even Daniel Kahneman, an expert neurologist, is helpless against overconfidence biases regarding task completion times. Furthermore, just because you rationally know you’re far more likely to die in a car crash than on a plane doesn’t stop people getting anxious about it (or not all, anyway).

            Human history is also a plentiful source for superstitions coming into being. It may be far more under control, but humans have cognitive problems which lead all but the most enlightened human cultures to believe in an afterlife, souls, and spiritual beings which clearly don’t exist.

            I would like to emphasise here that our ONLY source for the idea that moral truth exists is the human mind, which has already shown itself to have numerous cognitive flaws.

            3- A Moral Truth Theory can be divided into two kind- Moral Facts (simple enough) and Moral Intuitions (morality exists in the sense that it exists in human heads).

            A Moral Facts Theory postulates entities unnecessarily contrary to empirical methodology and Occam’s Razor. I can’t see any plausible reason it’s better than spirituality.

            A Moral Intitutions theory has trouble not dissolving into a Cognitive Bias theory. You would need to account for differing intuitions still making a single account of what is right and wrong, have a means to account for intuitions which contradict each other, and need to explain whether culture does or does not factor into it.

            To illustrate a key part of the dilemna, I give as an analogy the Euthryphro dilemna, particularly it’s early part (though the latter is also a dilemna you need to solve). Socrates points out to a person who proposes as a definition of pious as that which the Gods love that the Gods love different things.

            This all seems to be constructing a highly unnecessary construct given the alternative if one’s perpsective is based merely upon knowing truth.

          • suntzuanime says:

            No. You are illegally crossing the Is-Ought Divide. Cognitive bias relates to is-beliefs; ought-beliefs are fundamentally distinct, and to treat them as another form of cognitive bias is epistemologically naive.

          • Carinthium says:

            Hume’s establishing of the Is-Ought Divide sucessfully establishes that you cannot establish the existence of an Ought from the existence of an Is rationally in any way.

            However, this is perfectly compatible with the basic proposition I argue for- “Ought” does not exist at all. I am therefore explaining how the apparent evidence for “Ought” can be explained as “Is”.

          • suntzuanime says:

            No, the very claim “there Is no Ought” is a violation of the divide. You do not understand that of which you speak.

            EDIT: I guess there’s a sense in which “there Is no Ought” is an obvious consequence of the Divide, instead of proscribed by it. But either way, it’s not the Solution To All Ethics; it’s either nonsense or trivial.

          • MugaSofer says:

            Suntzu is right, I’m afraid.

            Imagine you met a paperclip maximizer on the road, as you do, and fell into conversation.

            You’ve looked up at the sky for words of fire, and you didn’t see “Thou Shalt Maximize Paperclips” written in the stars.

            You’ve read many lengthy treatises on logic, mathematics, and physical laws, and in none of them was the law, “The maximum number of paperclips possible shall exist in any situation.”

            You’ve looked for an instruction manual for the paperclip maximizer stating “WARNING: will attempt to tile the universe with paperclips if not prevented”, but there was no such manual – for they had all been turned into paperclips.

            You tell all this to the paperclip maximizer.

            Do you expect it to stop maximizing paperclips?

          • peterdjones says:


            “The question here is this- are we discussing what a person ‘pragmatically should’ believe, or what is true?”

            In normativity, the two coincide. What you practically should do to win at chess is truly the optimum chess strategy.

            STEM types have a typical limitation or difficulty or bias inapproaching normativity or ethics in that they tend to assume that the truth of a statement requires an object corresponding to the statement. But normative claims are claims about what is allowable or optimal,and can have quite robust truth values that have nothing to with a special ontology. The ontology of chess is carved pieces of wood, the ontology of ethics is human behaviour.

            “If we are discussing what is true, then what humans think or are capable of thinking is irrelevant.”

            The point is not what people in general believe, but what specific individuals believe. Why should
            I believe in As claim for X, if A does not in fact believe it?

            What would be your argument against someone who .claimed to disbelieve in an external reality?

            “Just world bias is a known phenomena”

            That’s not the right bias. You .can believe in good and doing good without believing in karmic  rewards.

            “I would like to emphasise here that our ONLY source for the idea that moral truth exists is the human mind, which has already shown itself to have numerous cognitive flaws.”

            Our only sour ce for the belief  in an external word is the human mind, blah blah….according to the external world sceptic. Whatever evidence you show to that kind of sceptic, they just insist it’s a figment of their own mind. Which is why you need to haul the sceptic up with the “ad hominem” argument that they behave as though there is an external world, and therefore do not really believe what they say they believe.

            In short, you need to make sure that your ethical scepticism does not involve  a universal solvent, that it does not remove truths you want to have, along with the ones you don’t,

            “3- A Moral Truth Theory can be divided into two kind- Moral Facts (simple enough) and Moral Intuitions (morality exists in the sense that it exists in human heads)”

            That is not how moral truth theories are divided ,that is how moral-thingy existence theories are divided.

            There is an important third category, where moral truth does not pre-exist, but is constructed by agents, in such a way that they can converge. A specific example is the contractarianism frequently expounded in this very blog.

            Under that kind of theory, the “existence” of ethical truth just means it is possible to solve a specific certain set of problems under certain set of constraints. Everyone knows what winning at chess is,
            So “there are true facts about how you should play chess” translates to “given certain constraints, it is possible to discover norms that optimize your probability of winning at chess”. Likewise, ethical truth would be the solubility of a certain set of problems involving voluntary control of behaviour to achieve globally desirable outcomes.

            In other words, ethics works like economic,is or decision theory.

            No additional entity is required to exist, in any non metaphorical sense.

            Note how that effects Eurythro problems. You might encounter a better proof of am ethical claim than the ones you have, but it would be proof, not some brute fact or arbitrary opinion, and it would therefore justify it’s conclusion,as proofs do.

          • peterdjones says:

            Hume’s intuitive, handwavey argument only strongly establishes that you can’t derive ises from oughts in short, logistic arguments,

            The contrary handwaving, intuitive argument has it that values are fixed, in some complex way, by the totality of facts, so that you can’t replay the history of the universe, with no fact changed, and have the Holocaust come out as a good thing.


            Not sure what your point is. If Clippie were an ideal epistemic rationalist, the ones inability to persuade Clippie of an ethical truth might mean there are no ethical truths, or an unbridgeable is-ought gaps … but Clippie is only an instrumental rationalist…and we know that arguments about ethics can be made.

          • Carinthium says:

            How on earth can the posistion that an Ought must exist be established from Hume’s arguments for the Is-Ought dividie even if you do interpret it that way?

            Your argument implicitly equates ‘The paperclip maximiser wants to do X’ with an objective Ought.

            The problem with this is that an Objective Ought of the sort humans instinctively believe in is supposed to be independent of human preferences.

            Your analogy about normativity only works if morality, like chess, has a clear end goal. But you have a dilemna depending on your choice of end goal.

            If the end goal is selfish for the individual concerned, why adopt ‘moral’ behaviour to fullfill it? If it isn’t, why should an individual care?

            No matter how you phrase it, there are going to be times when at least some people will care about something else more than they do the end goal of morality you postulate, whether selfish or selfless.

            If you’re not going to reduce normative claims to objects, than how to make sense of them as having objective existence?

            Ad Homimem:
            This is where we disagree. Hypothetically speaking, I could be dealing with an individual, A, who makes a case for something yet who does not believe it. A’s argument could, however, be rationally completely compelling.

            I actually consider the skeptic quite a major philosophical problem, which may be refutable but would take more than 3000 words at least to properly refute.

            Just World Bias:
            It is not identical to the belief in morality as a bias, but it is a very similiar bias. I am using it because it establishes a precedent for morality being classified as a bias.

            Human World:
            Ad homimem doesn’t become a legitimate argument no matter how many times you insist it’s convienient. In addition, regarding the skeptic it is circular as if skepticism is right you have no reason to believe it is impossible to believe in skepticism.

            Your entire system is flawed because of the concept of ‘truths you want to have’. A good philosophical system in any area will produce truth, regardless of what people want to believe is true. Otherwise, it’s just a product of human bias.

            Your Theory of Moral Truth:
            By your lights it makes far more sense not to talk about moral truth, but about moral ‘truths’- i.e. moral beliefs individuals happen to have.

            You also have a problem with placing the instinctive belief humans have in objective moral truths as anything other than a cognitive bias.

            Your system of moral ‘truths’ can be reduced to desires of individuals, moral or otherwise. So why are we even talking about morals? Why not simply get rid of morals and speak of things in terms of wants?

            This would also solve a lot of ethical dilemnas in the same spirit Parfit solved personal identity- by dissolving the question.

          • MugaSofer says:

            >an Objective Ought of the sort humans instinctively believe in is supposed to be independent of human preferences.

            Whether something is “the Paperclipping thing to do” is independent of the existence of actual Paperclippers.

            Similarly, an action could still be said to further human CEV, even in a world of Paperclippers.

            Of course, human CEV is much more complicated than Clippy’s values. It requires a lot more work than simply counting the number of paperclips to determine how “good” or “bad” a situation is. Still, we struggle on regardless.

          • peterdjones says:


            I see the goal of morality as voluntary modification of behaviour with regard to the desires of others.

            Not everybody will be motivated by “my” morality, and that has nothing to do with the non existence of moral truth,

            My theory is constructivist, which means that if you need morality you can construct it, and if you don’t you needn’t. You need to show that morality is unconstructable, or that you don’t need it. To do the latter, you need show that you are not going to free ride.

            Ad hominem is more than one thing.

            If skepticism is right, and I don’t believe in skepticism, there is noone who believes in skepticism.

            Heuristics for optimizing some X are not restricted to philosophy. It is not a devastating critique of constructivism that you won’t discover it’s truths without the appropriate motivation, since that is true of many other things,

            It’s truth, not truths, since it’s about coordinating the behaviour of groups. Note “which they can converge on”.

            Not all desires are morally relevant, since many can be satisfied without infringing on the ability of other people to satisfy their own desires, and therefore without the need to create coordination mechanisms.

        • MugaSofer says:

          “The theory I use is that belief in moral right and wrong is a cognitive bias which should be classified like any other cognitive bias.”

          [patronizing voice]Do you have any evidence for this theory that you’d like to share with the class? [/patronizing]

          • Carinthium says:

            I’ve given plenty of evidence in my discussions above already. Apologies for missing this post up until now, but I would have thought it was obvious.

            Let’s use a law metaphor here to clarify- treating it like a legal precedent what is and isn’t such a thing.

            Just World bias is a case of a systematic human delusion being treated as a bias. We aren’t supposed on your principles to rationalise the world so it is just, so why do we rationalise it so morality exists? Superstition also fits, if less well.

            Pragmatically speaking, if moral truths don’t exist (which I’m going to establish in the other thread we debate in), how else are we supposed to classify morality?

        • Illuminati Initiate says:

          Morality is not a cognitive bias, it’s a preference. Is preferring the taste of strawberry to vanilla ice cream a cognitive bias? Preferring a world without involuntary suffering (for example) over a world with involuntary suffering is like preferring strawberry over vanilla. Of course, for most people the preferences called “morality” are much stronger than food preferences, to the point that many are willing to cease to exist to make the universe closer to their preferred moral state. (by the way there is no objective reason to desire your own well-being, any more than there is to desire the well-being of others. That does not mean either of those are irrational, they are arational preferences.)

          • Carinthium says:

            Of course there is no objective reason to desire your own well-being. But there’s a severe problem with your argument.

            Most people have a belief in the existence of a moral Right and Wrong- hence why they are persuaded by moral arguments rather than dismissing them as preferences (the fact people have been persuaded by arguments about right and wrong is an empirical fact all-too often denied).

            There is also plenty of psycological evidence that on an instinctual level humans see certain things as objectively Morally Right or Wrong. Which is plain false.

          • Ghatanathoah says:


            There is no objective reason to prefer to own a couch that ways 80 pounds over a couch that weighs 90 pounds. But that doesn’t make the weight of my couch subjective. Regardless of what your preference for couch weights is, or if you have one at all, you should still be able to objectively agree with me on how much my couch weighs. Something can be objectively 90 pounds or objectively 80 pounds, regardless of what your preferences about how much it should weigh are.

            If there was a person who preferred to have 80 pound couches over 90 pound ones (and preferred this as an end in itself, not for any instrumental reason), and I discovered that they had misweighed their couch and bought a 90 pound one instead, I would be able to talk them into buying a new couch using a rational argument about objective facts.

            Similarly, there is no inherent reason to prefer doing Right to doing Wrong. But that doesn’t mean Right and Wrong aren’t objective. If a person has a preference to do Right, but is mistaken about what is Right, I can persuade them to change their behavior with moral arguments.

          • Carinthium says:

            Your first problem is that you define moral right and wrong in terms of human intuitions. You thus run into the same problems of Euthyphro trying to define piety in terms of what the gods love. The fact humanity is unanimous on almost nothing makes this even harder.

            Your second problem is attempting to establish that a human being’s CEV involves doing Right. Human beings have a systematic delusion in objective Right and Wrong. Remove that, and it is questionable they would care about what Right action is.

    • DanielLC says:

      From the perspective of someone that is not selfish, that’s not a solution. Becoming selfish would be good for me, but horrible for people in those bottomless pits. That is a net loss, so if I were to find a way to become selfish, I would avoid it.

  8. ShardPhoenix says:

    Maybe it’s just me being overly neurotic, but it seems like the nursing and prison issues can in principle be solved through enlightened self-interest alone. I don’t expect to end up in prison but I know there’s some probability of that, so even from a purely selfish point of view I wouldn’t want prison to be excessively harsh. And it’s quite likely that I’ll end in a nursing home or equivalent someday, so I wouldn’t want those to be excessively unpleasant either.

    The fact that neither of these problems have been solved yet perhaps suggests that others don’t think like me however (or it may be more a matter of ignorance).

    • Anonymous says:

      Much of the point of the post, particularly the last paragraph is that it’s really to make this argument about prison. But unless people are micro-locally selfish (which doesn’t match much behavior) people should care about the problems in nursing homes, so if you could produce that enlightenment, maybe the prison problem would be solved too.

    • Note that Americans don’t generally save nearly enough for their retirement either.

    • Ghatanathoah says:

      I think the problem with this plan for prison reform is that there are a large portion of American who think that criminals are fundamentally different in character from “normal people.” So they don’t think there’s any chance of them going to prison because they’re a good, normal person, whereas criminals are monsters. There are entire corners of the blogosphere specifically devoted to reinforcing this belief. See Steve Sailor and all the right-wing bloggers who dedicate all their time to complaining about the “social pathologies” of low-IQ people in loving detail.

    • DanielLC says:

      That only works insomuch as you expect to go to prison. Using that logic, someone less likely to go to prison would be harsher to criminals. Since it’s good to be viewed as someone who is unlikely to go to prison, people will be harsh to criminals to signal that they won’t become criminals themselves. Since they’re being compared to others who are trying to send the same signal, it causes a runaway effect where someone could easily have a harsher view on crime than someone who never considered that they might become a criminal but didn’t try to signal.

  9. social justice warlock says:

    I mean, this is kind of the standard view of history. Except that in the standard view, conservatives tack on “But really, the bottomless pit wasn’t so bad, and the sulfurous flames gave you a nice, warm feeling inside.” And leftists tack on “but in the end, everyone including the people in the nice town benefitted from the increased understanding and diversity this created, so really history was just this series of obvious win-win propositions that everyone was just too stupid to figure out, until now.”

    These are both obvious bullshit of course, in the Harry Frankfurt sense – things we would expect people to say regardless of whether they were true. Evidence for it being false can be found in that the hard right and hard left, who can say “yeah, I don’t give a shit about [pit|townsfolk],” generally acknowledge a difference of interests (though this is truer for the left than right.)

  10. Multiheaded says:

    First Scott is impressed by actually existing socialism, then he walks away from Omelas… what’s happening?

    Also yeah, I have often said much the same thing about history and pits, mostly when yelling at reactionaries. Yes, it’s curious how this framing makes conservatism to appear the easier, convenient and morally dubious way while leftists (aspire to) care about virtue for virtue’s sake, when in the popular imagination it’s vice versa

    • social justice warlock says:

      Disentangle liberals from leftists and the confusion disappears. (NRx, which incorrectly diagnoses the former being driven by the latter, has revived a Right tradtion (ironically a torch held by liberals during the period of the cold war) of seeing Virtue(“-signalling”) as the problem.)

      • Multiheaded says:

        Yeah, sure, I agree but…

        Hmm, firstly “driving” is not necessarily the same as “constraining the overton window”. As in, “Rawls was just a wishy-washy would-be leftist” is incorrect, but “Rawls couldn’t hope to get away with half the things Locke had said, and felt some leftward pressure” is probably an important obsevation.

      • nydwracu says:

        Don’t we have two thousand years of tradition saying that virtue and virtue-signaling are two totally different things?

        (Have there been any psychology studies along those lines? I think there have, but I don’t know what to search.)

        • social justice warlock says:

          I would call virtue a subset of virtue-signaling.

          • Chris H. says:

            That’s rather curious. Would you call heterosexuality a subset of heterosexuality-signaling?

          • ADifferentAnonymous says:

            The virtue-is-virtue-signalling theory is basically that the evolutionary advantage in being virtuous comes from being treated as a virtuous person, so what’s actually selected for is virtue-signalling. Sometimes actual virtue is an effective way to signal virtue.

            To make the same argument about heterosexuality, you’d need the premise “the evolutionary advantage in being heterosexual comes from being treated as a heterosexual person”, which doesn’t appear especially true. If you demonstrated that being known to have homosexual relations was disadvantageous and that avoiding this was the only advantage to not having such relations, then you could indeed draw that conclusion.

          • Chris H. says:

            Well, that at least explains where the notion comes from, but it still strikes me as confusing the explanation for a phenomenon with the phenomenon itself.

            To pursue my counter-argument further: I am a heterosexual man, and want to have sex with women, despite not wanting (at least at present) to produce children. Evolutionarily, it’s fairly clear that the reason I have an instinctive desire to have sex with women is that it increases substantially the probability that I will produce children. But it would be quite mad, on the basis of that fact, to say that my contracepted and/or non-PIV recreations are “a subset of seeking to reproduce.”

    • Princess_Stargirl says:

      I really don’t get how people think ruining an entire wonder city for the sake of one person is a good idea :(. They seem to think that perpetuating the Omelas situation is actually immoral not a wondrous triumph. Some of them even claim they would walk, not just work to free the child. (walking is even more pointless as it helps no one). I am always genuinely horrified by this reaction.

      • Matthew says:

        Walking is a ridiculous action, but freeing the child, including through violence if necessary, is not. In the story, the only purpose of the child’s misery is to provide a basis of comparison for everyone else’s happiness. It like positional goods taken to a bizarre abstract extreme. It’s not like the city is being powered by er tears.

        • Really? I had no idea that was a common interpretation of the story. Everyone I’ve ever heard talk about it seems to have been under the impression that the city was literally powered by a forsaken child.

          • Incidentally, there’s a notable literary precedent to “The Ones Who Walk Away from Omelas”: the scene in The Brothers Karamazov leading up to the story of the Grand Inquisitor.

            “Imagine that you are creating a fabric of human destiny with the object of making men happy in the end, giving them peace and rest at last, but that it was essential and inevitable to torture to death only one tiny creature—that baby beating its breast with its fist, for instance—and to found that edifice on its unavenged tears, would you consent to be the architect on those conditions?”

          • Matthew says:

            Whereas today is the first time I have ever encountered the “literally powered” interpretation.

            I’m not kidding. I have discussed this story in high school English class. I have discussed this story in creative writing classes. I have never encountered the version where people aren’t simply being positional sadists before.

          • Anonymous says:

            What proportion of assigned stories do you read? Did your class all use the same brand of cliffs notes? It’s really important to know what the teacher uses.

          • Matthew says:

            What proportion of assigned stories do you read? Did your class all use the same brand of cliffs notes? It’s really important to know what the teacher uses.

            I assume this is directed at me, in which case the verb tense is off by decades now. But the answer was 100%. I definitely did not fall into the “school seems pointless” cluster that seems common among LW (particularly among autistics, which I am not). As my high school was tracked by ability, the other people there were also (presumably) neurotypical high achievers and I would be surprised if they didn’t also read everything. College was self-selecting; people who wouldn’t want to read that stuff would choose some other class.

            The assumption that the teacher relies on some brand of Cliff Notes is amusing to me. (My mother teaches literature; she reads the Cliff Notes and Spark Notes and such just so she can compose essay questions that aren’t answerable by taking shortcuts.) Most people who teach literature care about literature.

          • Anonymous says:

            I don’t understand how a whole class could read that sentence above and collectively misunderstand it. Perhaps you don’t recall the discussion correctly?

          • Matthew says:

            Someone (maybe Orson Scott Card in his book on writing speculative fiction) noted that one large difference between science fiction and other genres is that metaphor often doesn’t work in science fiction — because so many things are possible, people take descriptions literally.

            It sounds to me like you read this story with people who took it as a a work of science fiction, whereas I read it with people who took it as a work of literary fiction. To me (and clearly the typical mind fallacy is a powerful thing) it was obvious that the causal connection was metaphorical and the people were wrong, empirically as well as morally.

            Basically, my interpretation was that the people saw this child powering their prosperity in the same way that ancient pagans saw burnt offerings powering their good harvests.

          • Quite Likely says:

            Yeah I agree with anonymous. The positional sadist theory isn’t totally crazy, but it is definitely not the obvious interpretation compared to literal dependence on the child’s suffering for the city to go on.

          • Anonymous says:

            For most purposes, such as the moral questions that started this thread, whether they are correct about the human sacrifice yielding good harvests is irrelevant; all that matters is what they believe. But believing that the misery yields good harvest is very different than believing that the misery gives them a point of comparison to make them feel better.

          • suntzuanime says:

            Does it even actually matter whose interpretation is correct? Whether it’s positional goods or pain energy, we’ve got a good thing going here and it’s really rude of you to spoil it just so you can moralize about, like, human rights or whatever.

          • James says:

            Both the “positional goods” and “powered by a forsaken child” interpretations occured to me on reading the story; I was frustrated by the ambiguity.

            I suspect that the vagueness of the causal connection between the child’s suffering and the city’s prosperity is actually a deliberate ploy designed to amplify the repugnance of the situation. It increases our feeling of arbitrariness, of “why-oh-why must it be this way?”. If there were a properly well-specified mechanism by which the child’s suffering led to everyone else’s happiness, I suspect it would be easier to argue for.

            It’s a decent thought experiment, but like Daniel Dennett suggests in Intuition Pumps, it’s worthwhile investigating its power by tweaking some of the variables and noticing how the resulting intuitions change.

          • Protagoras says:

            The “powered by” interpretation was the first that occurred to me, but various other stories of how suffering is somehow necessary (the savage in Brave New World, some common responses proposed by Christians to the problem of evil) eventually led me to the interpretation on which Omelas is a place where the argument that suffering is somehow a necessary part of a happy system is taken seriously, and they’ve found a way to contain the suffering to the tiniest minimum while still gaining the same benefits.

        • Hedonic Treader says:

          I always find it absurd when people pretend to be horrified about the Omelas bargain. In every actual society, children are forced to suffer for a lot less and it’s completely legal.

          • Carinthium says:

            Could you clarify what you’re talking about here? My guess would be education which people think acceptable for the child’s benefit, but I’m not sure.

          • Multiheaded says:

            It’s a matter of the difference in causalty and motives. Not necessarily hypocritical at all.

          • Hedonic Treader says:

            You can’t have a real (reproducing) society without making
            1 – all children suffer a little,
            2 – many children suffer considerably, and
            3 – a few children suffer extremely severely.

            Hypothetically, humanity could coordinate to ban all reproduction. If the political will was there. People will then say, “But this is ethically bad because humaaaan goood!” – which is equivalent to accepting Omelas.

            Multiheaded, yes you can play the framing game, but the pleasure-victimization tradeoffs are ultimately equivalent.

          • Carinthium says:

            If you’re just talking about ordinary matters however, couldn’t it be argued that it isn’t so much about suffering per se as the ratio of happiness to suffering? I.E it’s acceptable to make a child suffer to some extent as long as they have a certain amount of happiness proportionate to it?

          • Hedonic Treader says:

            Carinthium, in any real-world society (*) there will be quite a number of children who will experience more suffering than happiness, some of them severely so. If you argue that others will feel more pleasure than those children will feel pain, you are endorsing the Omelas bargain. Even if you don’t know in advance what will happen to which individual child, stochastically you know it is inevitable.

            (*) modulo speculative future utopias, post-enhancement etc.

          • Chris H. says:

            @Hedonic Trader: “People will then say, “But this is ethically bad because humaaaan goood!” – which is equivalent to accepting Omelas.”
            Consequentially? Yes. Deontologically? Perhaps, but probably not. Aretaically?

            Not in the slightest.

          • grendelkhan says:

            If you do the math on Omelas, it’s way better than what we have. (Twenty-five thousand happy, prosperous citizens for every starving child, whereas we manage less than twenty-two hundred in our oh-so-moral world.)

            I always thought the point was to shine a light onto how bad our own tradeoffs were, but apparently the idea is to get some good solid moral outrage going at those awful utilitarians who would do The Horrible Thing.

          • Multiheaded says:

            Oh come on, this and the “positional goods” interpretation are both fighting the hypothetical way too much from opposite ends. Neither really belong in the actual moral philosophy debate here.

          • Carinthium says:

            Hedonic Treader- To be clear, I wasn’t sure and was just checking. But on reflection, you’re right.

          • In the real world, the idea of parental rights increases the odds of children being horribly tortured by parents.

            Admittedly, a policy of taking children away from bad parents doesn’t work reliably either.

        • Deiseach says:

          Well, this is definitely a diference between genre fiction readers and literary fiction readers.

          If the point of the suffering child is to make the people of Omelas appreciate the good things they have by the contrast (“look how good you have it, you could be a lot worse off, so be happy!”) then they don’t need their own suffering child – just pointing out the misery in the city over the border, or the port by the sea, or the town up in the hills, would work.

          That’s the “eat your vegetables, children are starving in Africa” version. So if it’s just using a scapegoat, then sure: opening the door of the cell and letting the child out works, because all the good things will still continue without the suffering.

          But if (say) Omelas gets its blessings from a pact with a malign god who makes sure harvests are bountiful, the citizens are healthy, and no envious foes conquer the city by war, provided that a delicious source of mortal misery and suffering from an innocent is the sacrifice, then by freeing the child you destroy the city.

          So those who walk away are taking the middle path: they do not destroy the city and all their fellow citizens by freeing the child by force, but they no longer buy their privileged lives from its suffering. They go out and earn their own living and make or suffer their own joys and losses like the rest of the world.

          If there isn’t a real price to be paid in freeing the child, then there is no dilemma; setting the child free and refusing to imprison another in its place still leaves Omelas happy, beautiful, peaceful and free. The citizens pay no price for their virtue in giving up their lives of plenty, nor do they suffer righteous punishment by finally enduring the suffering of all the generations of tormented children that went before; the revolutionaries pay no price in shed blood and misery, nor bear no guilt for harming the other innocent children who now have to grow up in a fallen Omelas.

        • RCF says:

          If the child is there for positional goods sake, why wait until people reach adulthood to show them the child? And why do the ones that object to it walk away? If all of Omelas’ success can be achieved without the child, except for the positional good, why don’t the ones who object create their own no-tortured-child version of Omelas?

      • Carinthium says:

        This is clearly a case of utilitarianism vs deontology/virtue ethics. For most virtue ethical systems and some deontological systems walking makes perfect sense.

        I don’t understand why you aren’t horrified by deontology and virtue ethics instead of the Omelas situation.

        • MugaSofer says:

          >I don’t understand why you aren’t horrified by deontology and virtue ethics instead of the Omelas situation.

          I’m curious; how do you know they aren’t horrified by deontology and virtue ethics?

          Personally, I’ve always been horrified by both walking away from Omelas and deontology and virtue ethics.

          • Carinthium says:

            Because I figure they would have said so if they were horrified by that, since the connection should be obvious.

      • Chris H. says:

        “I really don’t get how people think ruining an entire wonder city for the sake of one person is a good idea :(.”
        Because it is what a human being of decent moral character would do.

        Understand that this is not rhetoric, but a literal description: From the perspective of virtue ethics, this is one of the two central moral questions: In such-and-such a circumstance, what would an individual of decent (or good, or noble) moral character do? The other is, of course, having such-and-such an indecent (or bad, or ignoble) moral character but desiring to change it, what course of action should an individual pursue in order to cultivate the right sort of character.

        For any notion of what it means to be fundamentally decent that is not facially absurd, freeing the child is what a decent person would do. Thus, for any virtue ethicist who views decency as all or part of virtuous character, such an action is self-evidently righteous.

        When you get right down to it, strict consequentialism is inhuman, in a rather precise sense, and additive utilitarianism is inhuman in almost every sense. Why should it surprise you that most humans reject it?

        Walking, on the other hand, is stupid under both consequentialism and any reasonable form of virtue ethics. But there are probably some internally coherent deontological (or hybrid deontological/consequentialist) moral frameworks in which it would be viewed as the correct choice. I hold deontology in almost as low regard as consequentialism (it’s also inhuman, but less so, at least in some realizations), so this is irrelevant to me, but not to everyone.

        • Carinthium says:

          Interesting. I’ve never dealt with a virtue ethicist on this site before. I think I’ll take the opportunity to challenge the system.

          First problem- What does it mean to have a good moral character? Merely good character by our culture’s standards? Appeal to some mysterious universial standard? Or what?

          Second problem- An extenstion of the first- demonstrating that your concept of Good Moral Character is entirely internally coherent.

          • Chris H. says:

            I get the impression (perhaps incorrect) that most commenters here are connected to Less Wrong in some fashion, which I am not.

            As regards your first problem, I would imagine that different virtue ethicists would answer that question differently, and I’m not sure that there’s any clearly dominant reply — I don’t think that virtue ethics really has a “default” form in the same way that consequentialism has additive utilitarianism.

            Speaking for myself, I would say that a good character is in principle one that is well-conformed to the objectively correct standard of what constitutes a good character. In practice, what I regard as a good character is one that is well-conformed to my moral intuitions, tempered by experience.

            I am rather agnostic about whether objective moral facts actually exist or not. In practice, I think it is both useful and commendable to act as if they not only do exist, but can at least be approximated in this fashion — still recognizing, of course, that one’s approximation is likely to be imperfect, and may at times be in need of revising.

            It is worth noting that although my standard of good character is in some sense relative (to my own moral intuitions), it is not relativist: I do not regard an individual’s character as good insofar as it is well-conformed to their intuitions, but insofar as it is well-conformed to mine.

            An objection will likely be raised that I am relying entirely on moral intuitions that cannot be proven correct, at least by any means that anyone has discovered thus far. (Even the process of tempering my intuitions by experience is ultimately a process of tempering them against each other, not against something external.) This is true, but also irrelevant, because it is true of every other moral system as well. Consequentialist systems also depend on moral intuition, they just front-load that dependence: In order to accept strict additive utilitarianism (for example), you must accept, on ultimately intuitionist grounds, that the morally correct thing to do in all circumstances is to act in such a way as to maximize total utility, summed across all agents. This is a huge leap to make, but once it is made, any remaining leaps are tiny in comparison (e.g. questions about how to define utility in such a way as to allow it to be summed across multiple agents). In contrast, virtue ethics requires no similarly enormous leaps of intuition, but it does require a much larger number of moderate jumps.

            Regarding your second problem, I’m fairly certain that my concept of Good Moral Character is not entirely internally consistent. But it is the best approximation I presently have to a system that is both internally consistent and true (if there are actually objective moral facts, and my means of approximating them are effective), or at the very least both internally consistent and conformed to my (suitably tempered) moral intuitions (in any event).

            If I observe a contradiction or incoherency, then there are really two choices available to me. One is to resolve it by adjusting my standards — this is essentially the sort of thing I am talking about when I talk about tempering my intuitions through experience. The other is to decide that my current standards, inconsistency and all, seem to me to come closer to being right than any revision that I can come up with that removes the inconsistency. There’s no contradiction in that; it merely means that I have not succeeded in finding a sufficiently good resolution to the problem at hand, let alone the optimal one.

          • Susebron says:

            How big of an intuitive leap is utilitarianism, though? Saying “I think the best thing to do is to make as many people happy as possible” isn’t that big of a leap. Some of the consequences of that statement aren’t intuitive, but they’re not as big as you seem to be saying.

          • Chris H. says:

            @Susebron: It’s an enormous leap, because it’s not human.

            A normal human being cares more about their spouse and children (as applicable) than about their cousins, more about their cousins than about more distant kin (and/or countrymen as symbolic kin), and more about their kin (and/or countrymen) than about strangers. A normal human being has tastes, and would like to see a flourishing of styles (of clothing, art, music, architecture, literature, etc.) they have a taste for, even if other styles would please more people in total. And a normal human being has things they care about morally that are not about happiness at all.

            Utilitarianism allows for none of those things. (Or, at least, for acting on none of those things.) You could make a typical human being into a better utilitarian simply by selectively removing much of what makes them human, without adding anything in its place.

            Zoomed in, this means that accepting utilitarianism is a huge leap because it requires rejecting so much that one instinctively believes and acts on. Zoomed out, it is huge leap because there is something quite extraordinary about the assertion that the right way for humans to behave is, fundamentally, as if they were not human.

          • Susebron says:

            I think you’re talking about… I’m not sure what to call it. Simple utilitarianism? Anyway, most utilitarians (or at least the ones in the comments here) tend to support a more complicated view, based on generalized human values as opposed to simple happiness (yes, I know I said happiness, but it’s a bit difficult to summarize human values). And maybe it’s inhuman to directly think about human values on a meta level, at which point I don’t think that inhumanity is a sustainable objection.

          • Chris H. says:

            There is a reason that I said that “strict consequentialism is inhuman, in a rather precise sense, and additive utilitarianism is inhuman in almost every sense.” Selecting a better form of consequentialism is undoubtedly an improvement (potentially an enormous one), but it’s still inhuman because humans are particular and strict consequentialism* is incompatible with particularism.

            To be perfectly clear, there is nothing inhuman about taking universals into consideration, nor is there anything incompatible with virtue ethics. What is both inhuman and incompatible with virtue ethics is to say that only universals may be taken into consideration.

            *Of the sort that anyone here is actually discussing, i.e. agent-neutral consequentialism.

          • Carinthium says:

            I’m a moral nihilist myself in the sense I agree that moral beliefs and intuitions exist but consider any attempt to form them into a moral truth beyond “I want” pointless, incoherent, or both.

            My attack would be this. If your system is based merely on your own intuitions, it is merely a type of “I want” so why bother with ethics instead of being a moral nihilist? Why treat ethical desires as stronger than other ones?

            If by contrast your system is based on objective moral facts, what evidence do you have for their existence?

            Your posistion on contradictions can be more effectively resolved by a moral nihilist who converts their moral beliefs into simple desires.

            This is because there is no contradiction between a conflict betwen two desires, and humans already have memes for resolving conflicts between two selfish desires.

            In a very, very broad sense I could be called a virtue ethicist in that there are certain traits I want to have, if not necessarily for moral reasons. But these aren’t rooted in moral intuitions, but in “I want” as a first principle.

          • Chris H. says:

            Is that really moral nihilism? It seems to be treading awfully close to the border with existentialism (albeit of a less nauseated flavor than is traditional), if not actually stepping over it.

            Which, I suppose, my views arguably do as well, from the other end. Existentialism/”I want” is a sort of a fallback position for me, insofar as I believe that I am approximating an objective (though hard to discover) moral truth, but I am certainly expressing a particular subset of my desires.

            My evidence for the existence of moral facts is largely intuitionist; whether I should view this as evidence at all is a matter of some controversy, but I can hardly expect you to (c.f. Thomas Paine’s remarks on revelation). If you want to put things in formal Bayesian terms, I suppose that you could say that my belief in objective moral reality is chiefly the product of a strong prior.

            I’m not sure that I’d accept that moral intuition is merely a type of “I want” in any event; I think that “I feel that I/someone else should” is sufficiently distinct in character to merit its own category. But this is sense is unnecessary to address your central question (as I understand it): Why do I privilege (or try to privilege — I would not assert that I am truly moral, only that I strive to be) my moral desires over others? As I see it, there are two reasons:

            First, I am inclined to do so by my psychology. That is to say, I have a strong instinctive second-order desire which says that I want to want what is right (as I see it) more than I want the other things that I want, and moreover to choose what is right over the objects of my desires even if I do not desire it more strongly. I think this is a very common human trait, albeit one that many people tacitly attempt to neuter by convincing themselves that what is right neatly coincides with their non-moral desires.

            The second is that I see it as part of a noble character to regard the desire to be virtuous as more than a mere common desire, and to prize virtue over the satisfaction of non-moral wants. Consequently, to hold the moral views I do as mere desires would be intrinsically dissatisfying to me. Thus, for me to be better off doing so, it would not be sufficient for this merely to provide a better set of decision criteria in individual cases of conflicting wants; it would have to provide a sufficiently better set of criteria to make up for its own intrinsic cost.

            Of course, all of this is an analytic treatment of a phenomenon that I don’t normally think of in quite those terms. My more immediate reaction to the question is more along the lines of “because that would immoral and ignoble; the very thought is an ugly one.”

            Also, I agree that there are means of resolving conflicting desires that don’t require privileging those desires that are moral in nature, but I’m not convinced that those are actually more efficient, at least for me. Almost any sort of privileging is, to a large extent, efficiency-boosting insofar as it simplifies the comparisons that need to be made.

          • Carinthium says:

            I hesitate to call any belief system existentialist which denies the existence of free will.

            Thomas Paine’s remarks on revelation (assuming I looked up the right ones) are rationally unfounded as they depend upon God’s existence and are thus irrelevant.

            The facts about your psychology (which don’t apply to me, admittedly because my psycological desire to be rational has usurped ‘pride of place’ in a pseudo-moral manner) are good enough unless there is no actual moral truth to work towards. If there is no such truth, the desire makes no sense.

            The noble character argument has problems thanks to the problem of ‘what if no moral truths actually exist’?. It works only if you are willing to erect an arbitrary line of ‘moral’ and ‘non-moral’ truths with no basis in order to satisfy your desire, which is either deluding yourself or unsatisfactory.

            Efficiency isn’t everything. An efficient means of desire resolution can leave a person unsatisfied.

          • Chris H. says:

            I have profound difficulty understanding how someone can actually function who truly does not believe in at least a compatibilist version of free will. (Personally, I’m at least locally compatibilist, and agnostic about compatibilism in general, but with a strong-bordering-on-absolute prior on the existence of free will, whatever the proper definition of that may be.)

            I was referring to this:

            It is a contradiction in terms and ideas to call anything a revelation that comes to us at second hand, either verbally or in writing. Revelation is necessarily limited to the first communication. After this, it is only an account of something which that person says was a revelation made to him; and though he may find himself obliged to believe it, it cannot be incumbent on me to believe it in the same manner, for it was not a revelation made to me, and I have only his word for it that it was made to him.

            I don’t think that attitude requires a belief in God, only a belief in the possibility of God. In terms of rationality . . . some time I have to write up an essay on the the reasonableness of certain forms of fideism, when placed in a proper Bayesian context. This is probably a good time (or at least, an amusing one) to mention that I’m a practicing animist and Sun-worshipper.

            Back on the debate that we’re already having, I think that the facts about my psychology are still relevant even if there is no objective moral truth to work toward — if I want to overridingly want what is right, and I observe myself to do so (however incorrectly), then I obtain satisfaction from the “knowledge” that I have done what I ought. Although that “knowledge” is ultimately false, the satisfaction that it produces is real.

            Moreover, one could reformulate a weaker version of the same second-order desire that would still be coherent in the absence of objective moral facts, and that would still be a desire that I hold: I want to want what I believe is right more than I want to achieve my non-moral desires.

            I agree that an efficient method may produce unsatisfactory results. But it is my strong empirical observation that prioritizing virtue, conceived in roughly the way that I in fact conceive it, results in my being happier and, in fact, more likely to achieve high-value non-moral ends, as does holding various beliefs about objective morality.

            That’s not actually evidence for the existence of moral facts. But it is a compelling reason, in my view, for me to hold a strong prior in favor of that same existence.

            And although I only have strong empirical evidence of this as relates to myself, it seems to me (acknowledging the possibility of bias here) that I also have somewhat weaker evidence that this is true much more broadly, though not necessarily universally.

            [Courtesy note: I am going home and going to sleep now, and so will not be responding further tonight. I will probably check this comments section at some point tomorrow, but am uncertain when.]

          • Carinthium says:

            I’m rather tired myself, but I think I should make some final remarks.

            On Free Will:
            Some people say that you have to act as if you have free will even if you don’t, but this isn’t quite true. Let me use an analogy here purely for clarification purposes- an A.I can act perfectly well whilst believing it’s own actions to be deterministic. What is it about a human that makes them different?

            When I was young, admittedly I simply worked around it by treating myself as if I had free will whilst believing I didn’t. But as time went on I learned to speculate on myself as if I were somebody else in order to determine better what influenced me. The metaphor I used was being at the head of a loose federation and politicking within it. This helped me get around it.

            On Revelation:
            In most Biblical cases, a revelation was usually backed up with a known record of correct prophecy or miracles to get around Paine’s problem at the time. However, he is at least partially correct.

            That being said, Paine’s argument is one for the unreliability of beliefs from revelation. It cannot in any way demonstrate that a revelation is authentic to the person who originally recieves it.

            On Moral Facts:
            Rationality is divided into two things- instrumental and epistemic. Your posistion is instrumentally rational (in that you achieve your ends) but not epistemically rational (in that you lack true beliefs). That being said, saying you must be epistemically rational in this case fails the amoralist challenge.

            Your weaker version is debatable on the basis of C.E.V. Since you aren’t familiar with LessWrong, I’ll simplify by saying that a person’s Coherent Extrapolated Violition is a construct representing what they WOULD want if they knew all facts that might possible affect what they desire.

            The question is- IF you truely did realise, on both an intuitive brain level and an explicit level, that there was no such thing as objective Right and Wrong, would you still want what you want? Admittedly I can’t answer that in your case, so I admit you might have a point depending on the answer.

            A person could ignore CEV in theory, but I suspect you can’t because it’s very counterintuitive. Take the resulting case: A wants to open a door, not knowing there is a bomb on the other side set to blow them up and kill them if they do. B tackles A out of the way of the door. Without CEV, you’d have to say that B was acting in an unfriendly way. The possible substitutes are ultimately incoherent.

            I suppose you could make this work by classifying things you ‘want to want’ into a category linked only by your wanting them, but this is so far from being proper philosophy or rational as to make it only a matter of personal desires.

          • Chris H. says:


            On Free Will:
            Are you familiar with the compatibilist understanding of free will? Because your objections seem to be largely to libertarian/incompatibilist notions of free will. It seems reasonably likely that you believe in what a compatibilist would call “free will,” but, being incompatibilist yourself, do not regard it as free will. If so, the semantics are largely irrelevant to my basic question of whether someone can really fully accept hard determinism in the strictest sense and still function as a human being.

            On Revelation:
            I agree that Paine’s argument does not support the validity of first-hand revelation. It does, however, support the idea that the validity of first-hand revelation is irrelevant to the (in)validity of revelation at second hand. Hence my remark: “[W]hether I should view [my intuitions regarding the existence of moral facts] as evidence at all is a matter of some controversy, but I can hardly expect you to (c.f. Thomas Paine’s remarks on revelation).”

            On Moral Facts:
            We agree that my position is instrumentally rational but not necessarily epistemically rational, but I would argue that that’s because the notion of epistemic rationality doesn’t really apply to the question of priors anyway: Epistemic rationality is how you get (correctly) from priors + experiences to posteriors, but the ultimate source of priors is something (or, really, several somethings) else entirely.

            On CEV.:
            If I consider the scenario in which there are no objective moral facts, but I believe that there are, then looking at the scenario from the outside, I would say that I still want for myself in that scenario to want what I believe (falsely) to be objectively morally correct more than I (still in that scenario) want to achieve my non-moral desires. Even if it not more moral than the alternative, it is in my estimation nobler, and even if nobility of character of character is not objectively righteous, it is still something that I prize.

            By the same token, I am inclined to say that even if I knew my subjective sense of virtue to have no objective backing, that I would still want to (actually) privilege it over my other ends — again, because it is a worthy thing to do so, according to that same sense of virtue, and even if I have no objective reason to value that sense, I still do.

            I am not sure that I entirely agree with your argument for CEV. I do agree (probably*) that B ought to tackle A out of the way, but not for quite the same reason. I am inclined to say that in preventing A from achieving his immediate aim, B wrongs A (in a conceptually local sense), but in preventing A from being prevented in achieving his broader aim (to continue living), he does him a service which exceeds the wrong, and as the two are intrinsically linked, he does A no net wrong at all.

            I am not quite certain what you mean by your last paragraph.

            *I say probably because my instinct here seems like it might (but probably does not) conflict with my instinct about predictable but unpredicted negative outcomes, e.g. with regard to starting up a methamphetamine habit. I will have to give the matter some more thought. But the alternative actually has me more strongly in conflict with CEV, so this particular uncertainty is irrelevant to this particular discussion.

          • Carinthium says:

            On Free Will: I disagree with the compatibilist that the actual state of affairs can be defined as ‘free will’, yes.

            I’m not sure what you mean by hard determinism here, but it’s irrelevant as the question here is what is true. Whether it is possible for humans to believe it and still fucntion as human beings is logically irrelevant to its truth.

            On Revelation: Fine, you’re right there.

            On Moral Facts: We agree on instrumental rationality and disagree on epistemic rationality, partially because I maintain that it is completely contrary to epistemic rationality to have a willful delusion.

            I’ll move onto the next point as I don’t think epistemic rationality affects priors per se, but indirectly does through C.E.V.

            On C.E.V:
            I’m skeptical instinctively, but it’s hard to say as I don’t know you as a person. Given this, I’ll say that your C.E.V posistion is fine and perfectly defensible IF your factual claims are true.

            That being said, even then it doesn’t change the fact that morality has a lot of coherence problems.

            On my argument for CEV: Come to think of it, you actually have a point about that. But there is another plausible argument.

            By definition, if a person truely understood the situation properly they would follow their CEV (already established). The only reason they’re held back from their CEV, whatever it is, by definition must be their own ignorance or delusions.

            Hence, to the extent a person is acting against their own CEV isn’t that effectively a form of insanity?

          • peterdjones says:


            “Some people say that you have to act as if you have free will even if you don’t, but this isn’t quite true. Let me use an analogy here purely for clarification purposes- an A.I can act perfectly well whilst believing it’s own actions to be deterministic. What is it about a human” that makes them different?

            The actual argument is not that some paralysis will fall over your limbs if you lose free agency…
            it’s more about acting as an agent, justifying your actions as an agent. Even by reductive accounts, an agent , to be an agent, needs to make some sort of choice or decision between courses of action, according to goals…knee jerks, or rolling in the direction you were pushed don’t count.

        • MugaSofer says:

          >Because it is what a human being of decent moral character would do.

          This seems obviously false to me. I certainly would not trust anyone who valued an entire city full of people so lightly.

          Yes, any decent person would *feel bad* about the child.

          But – and if I’m wrong, I would be really interested to see your evidence – it seems to me that only someone pretending to be good would actually free the child, let alone walk away from Omelas.

          They would have to be deliberately distracting observers from the city in order to signal concern for the child. That isn’t morality; that psychopathic.

          • Chris H. says:

            Your reply, as far as I can see, only makes sense under strict consequentialism, which I would presume that you follow. (Even then, not all forms of consequentialism would necessarily lead to your conclusions.)

            Virtue and the desire to signal virtue to others are distinct. A genuinely decent human being would free the child because they cared about doing right by the child, not out of a desire to signal that they cared about doing right the child.

            A very different topic, but a similar distinction. I am a heterosexual man. Given the opportunity to have sex with an attractive woman of suitable age (and barring any number of reasons not to), I will do so, not because I want to signal that I am sexually attracted to women, but because I am.

          • Ghatanathoah says:

            @Chris H.

            I see this as a scenario where virtue ethics and consequentialism should be in complete agree.

            A consequentialist would not free the child because inflicting misery on a ton of people in the city is a bad consequence.

            A virtue ethicist would not free the child because a virtuous person would not inflict misery on a ton of people in the city.

            If “knowingly and willingly inflicts massive harms on other people” is virtuous I don’t want to be virtuous.

          • MugaSofer says:

            Chris, there are also *fake* signals of heterosexuality – gay-bashing IIRC is weakly correlated with closet homosexuality, probably for exactly this reason.


            Ah … now that I think about it, destroying Omelas might also indicate a combination of [virtue + foolishness.]

            If you genuinely forgot about the city above you, or don’t realize the connection in time, you would try to free the child.

            Walking away, however, seems only plausible if you’re faking a virtue-signal (or engaged in some very complex deontological/TDT shenanigans.)

          • Chris H. says:


            You seem to regard consequentialism as an essential part (perhaps even the essence) or virtue. Given such a notion of virtue, it is hardly surprising that virtue ethics and consequentialism should agree, not only in this case, but in virtually all others.

            But I reject consequentialism, not only in itself but as a virtue as well. A virtue ethicist (of a type sufficiently similar to my own) would free the child, because it is the decent and honorable thing to do. That it would cause misery for the citizens is unfortunate (at least as pertains to those among them who are below the age of majority), and would be a cause for a regret, that the honorable thing to do could not be done without such consequences. But it would not be sufficient reason not to do it.

          • Chris H. says:

            @MugaSofer: To be sure, fake signals of heterosexuality exist. But that only reinforces my point: Heterosexuality is distinct from signaling heterosexuality. Virtue is distinct from signaling virtue.

            The city above is irrelevant to the question of whether one ought to do the decent and honorable thing. It is regrettable, to be sure, if one cannot do the decent thing without a great deal of misery resulting, some of it even for people who are undeserving, but it is ultimately beside the point.

            We are in agreement about the plausible reasons for walking, though. That’s somewhat interesting.

          • g says:

            Chris, it is not decent and honourable to carry out an action that results in thousands of people’s lives being plunged into suffering.

            It seems that you are rejecting not only consequentialism (i.e., the doctrine that consequences are all that matters morally) but the very idea that one should consider consequences at all: once you’ve categorized an action (by whatever occult means) as “the decent and honourable thing”, it simply doesn’t matter whether it results in millions of people being tortured, or the extinction of the human race, or what: it’s The Decent And Honourable Thing, and it therefore must be done. And to me, that is in itself indecent and dishonourable, and a virtuous person would not endorse such bullshit and call it ethics.

            Evidently you disagree. Would you like to tell us why?

        • Deiseach says:

          Walking is the middle way. You no longer live your life of ease at the cost of another’s suffering.

          On the other hand, you don’t decide for others that now they must suffer for your virtue, e.g. one child in misery versus all the other happy children who grow up loved and safe in beautiful, peaceful, wealthy Omelas.

          Revolutionaries would say “burn it all down, never mind who bleeds, the continuning injustice cries out to heaven for vengeance and if your child must now go hungry, must weep, must blister their hands toiling in the weed-choked fields, why should they be spared the common lot of all humanity?”

          Walking away can be seen as cowardly (why not stay and overthrow the city by force?) or it can be seen as an imperfect answer in an imperfect situation (do the needs of the many outweigh the needs of the one?) – by walking away, you give up what you have taken unknowingly (until you are initiated into the price that has to be paid for your security and wealth), you are no longer complicit by that much – but you don’t force the choice on your fellow-citizens either, by forming a revolutionary cabal, storming the Bastille prison, freeing the imprisoned child and overthrowing the government – which, if the suffering genuinely is what protects Omelas, also means that the men, women and children of this generation and future ones will suffer, be poor, be prone to sickness, misery and the chances of war, and will lose their happiness and peace.

          Those who walk away make the choice for themselves and for no-one else. Whether this is satisfactory is still ambiguous, even in the story.

          • Multiheaded says:

            Yes, this is a very correct description. Also, to reiterate, in Le Guin’s other stories, outright revolution is seen as at least conceivable; that’s just not what this story concerns itself with.

          • Chris H. says:

            I think that you (and many others, perhaps most) and I have very different notions of what it means to be complicit. You see complicity in accepting benefit from an injustice, while I see it in being able to prevent that injustice and not doing so.

            To put it another way, walking does not demand that the citizens suffer for your virtue (in freeing the child); instead, it demands that the child suffer for your “virtue” (in not making the citizens suffer for your virtue (in freeing the child)).

          • Ken Arromdee says:

            If you walk away, you no longer live at ease at the cost of another’s suffering. But the total amount of suffering remains the same. Responsibility for that child’s suffering is now divided among N-1 people instead of N people. In other words, if you walk away, you’ve reduced your share of the child’s suffering at the cost of increasing everyone else’s share. So you’ve harmed yourself for no gain.

            In the real world, if you walk away from causing someone suffering, the suffering is reduced, at least on the margin (if you don’t buy cheap sweatshop-made goods, the market for sweatshops is marginally lower, and using sweatshops becomes marginally less profitable, so decent wages becomes marginally more likely. “If I don’t buy it, someone else will” is wrong–you’re affecting the market by not buying it). Omelas lacks this–walking away doesn’t reduce the suffering at all.

        • grendelkhan says:

          Because it is what a human being of decent moral character would do.

          From a consequentialist point of view, we trade off a lot more suffering for our happiness in the real world. As a virtue ethicist, does that mean that a human being of decent moral character would tear down the system of the real world?

          Or is there… I’m not very good at making this sound like I’m not making fun of your ethical system, but I’ve been influenced by respected virtue ethicists making really stupid arguments… in war, for example, there’s some sort of Catholic doctrine which implies that if your intent really matters here; if you drop bombs which happen to kill civilians, even if you knew they would, but that wasn’t your primary intent, then it’s ethically good, but if you intended for even one civilian to die, then the whole thing is an evil enterprise. (It’s all very Ender’s Game, now that I think about it.)

          Does your evaluation of the right thing to do in Omelas have any bearing on the right thing to do in the real world? It seems that you’re obliged to do some pretty out-there things. Am I missing something?

        • A side note: freeing the child is certainly not enough to have a large chance of improving it’s situation.

          You’d be obligated to either take of it or find someone else to take care of it.

      • James says:

        It’s hard to make any definite claims because so much in the story is underspecified, but it’s possible that walking isn’t pointless. It sends a very costly signal of disapproval; if enough people did it, then it could cause things to change.

        • CAE_Jones says:

          Walking decreases the population. Do it loudly enough, and in large enough groups, and it might catch on to the extent that the population will shrink. If this succeeds, it reduces the utility cost of helping the child enormously.

          Also, the ones who walk are generally going to be inclined toward egalitarian ethics but also toward prosperity. If enough of them are capable enough, they could contribute to an alternative society that, while not Omelas-level superhappy, would be attractive enough to help with the incentives it would take for this gambit to work.

          It will, however, require several generations, so in the mean time, a great deal of suffering is permitted.

          However, selection pressures would likely result in Omelas eventually being populated by isolationist sociopaths, and people in the city-of-walkers might very well forget about their entire purpose for leaving, now that there is more distance at work. And we’re right back to Scott’s original post.

          • Multiheaded says:

            However, selection pressures would likely result in Omelas eventually being populated by isolationist sociopaths

            Well, we might argue that at some point this would constitute just cause for simply destroying the city; really not sure if this possibility is at all part of Le Guin’s intent. In The Dispossessed, the Odonian anarchists were actually waging a world revolution, and accepted resettlement to the other planet as part of a peace offer after Laia Odo’s death.

            and people in the city-of-walkers might very well forget about their entire purpose for leaving, now that there is more distance at work

            Omelas itself makes all citizens acknowledge its raison d’etre and show their informed consent. So presumably the negation of this would be part of the other place’s founding myth, although of course such things get twisted for political convenience (as certainly happened on Anarres!)

            Also, yes, I maintain that much of virtue ethics and deontology is what humans use as folk TDT, so at the very least it’s unwise to dismiss them out of hand.

          • Null Hypothesis says:

            Doesn’t that logic also cut the other way though? If you increase the populace, you increase the happiness and prosperity in the world for no additional cost. – you make the suffering more efficient in producing happiness.

            Therefore, by attracting immigrants, increasing population, and expanding Omelas, you make the child’s suffering that much more valuable.

            What I find interesting, is that Scott’s post basically talked about people disregarding the suffering of strangers. Many people are speaking about the suffering of the child at the heart of Omelas as being such a phenomenon – people accept the suffering of someone else for their prosperity. The child is ‘the other’.

            But when you consider, in the rest of the world, the countless number of children that suffer significantly as a consequence of life, whose suffering could be ended if they were brought to Omelas, or the children who will suffer if they leave Omelas – then you’re privileging this one child’s suffering over the ‘other’ children in the world. Arguing for continuing the city is ‘othering’ the child – but arguing for ending it is familiarizing him, and ‘othering’ all other children, that wouldn’t suffer should Omelas be maintained.

            If you’re not allowed to value different human’s suffering differently, (if you were, you could morally ignore the bottomless pit) then the only answer you’re left with is the minimize total, absolute suffering.

            In my opinion anyway, the world is nothing but a massive number of Omelases spread about. Sometimes an Omelas has a child suffering only for the benefit of a small tribe. Or even just a family. Or even one-sole adult. Some larger ones contain thousands or hundreds of thousands of people. The larger the community, the less immoral the society – the more good they do with the suffering they cannot rid themselves of. Omelas Prime houses millions of prosperous people at the same cost of a single tortured child.

            The best solution, is for Omelas Prime to expand and subsume the entirety of humanity – freeing all the children at the hearts of all the other Omelases around the world.

          • Multiheaded says:

            Le Guin almost certainly did not intend for the hypothetical to be stretched this way.

      • Peter says:

        Walking away: assuming the “positional sadism” reading is the wrong one, I think there’s still an argument to be made for this – one you could try putting in consequentialist terms at that. Simply: how does this all work? It’s a long time since I heard the story read, but I came up with the notion that there’s a Being With Too Much Power[1] involved, perhaps a bargain has been offered. It’s easier for me to imagine that than to imagine some magical happiness juice being extracted from the child or something. Walking away is refusing to give the BWTMP what she wants, if enough people do that / if enough people are committed to acting in such a manner, then the BWTMP might change the bargain, or BWTMPs in future won’t offer/extend such bargains. It’s a bit like hostage situations; if you make it a policy to not give the hostage takers what they want, maybe you can prevent them.

        [1] If you like to break the fourth wall, how about the author? A non-fourth-wall breaking BWTMP would probably work better though.

      • Protagoras says:

        I’m consequentialist enough to take it as further evidence of the general moral superiority of the people of Omelas over the typical people we encounter in our world that nobody in Omelas does in fact ruin the wonder city for the sake of one person. Even the ones who object just walk away.

        • Matthew says:

          general moral superiority of the people of Omelas

          Not buying it. Let me set aside my “superstitious sadists” interpretation and accept arguendo the “literally powered by tears” version.

          Now the Omelasians are reactionaries. Society is stable, and a hierarchary with only one person at the very bottom is pretty great, and well, you can’t make an omelet, etc., etc. There is no mention anywhere in the story of anyone even considering the possibility that there might be ways to achieve prosperity that don’t involve the abject torture of one individual, let alone researching the issue.

          Maybe they aren’t serving Moloch, but they’ve cut a deal with him.

          • suntzuanime says:

            Society is stable and a hierarchary with only one person at the very bottom is pretty great and you can’t make an omelet &c. &c.

            This isn’t the right blog comment section to be using reactionary as a scare term. If we have to sacrifice one child to Moloch, we got off pretty light.

          • Matthew says:

            I wasn’t using it as a scare term. It was a straightforward description of a moral worldview that many consequentialists disagree with.

            Sacrificing one child is better than sacrificing many, but that’s no excuse to stop there and not even be looking for ways to sacrifice zero.

          • Multiheaded says:

            Okay, substituting “insufficiently morally egalitarian” for “reactionaries” should dissolve this problem. Basically, part of an egalitarian moral outlook IMO is assuming that the negative moral value of suffering increases exponentially, and thus, if the increase is sufficiently sharp, it’s worse to have a single child in neverending extreme misery than a billion children having, like, a hundred very bad days each throughout their otherwise ordinary lives.

          • Protagoras says:

            Er, where in the story does it say that the people of Omelas aren’t and never have been looking for a way to sacrifice zero? The story is pretty clear that they haven’t found one yet, but I can’t find even a hint of a suggestion that they’re not looking.

          • My interpretation of Omelas is that it’s the least convenient possible world.

            Why is the utopia powered by a forsaken child? Because it creates a moral dilemma: which matters more, the child or the utopia? And if you live in the utopia, what do you do?

            Why is everybody in the city brought to see the child? Because that way, nobody can claim that they would have acted differently if they had known what was going on, or if they understood the full horror of it rather than just hearing about it in the abstract like we hear about faraway horrors. Everyone who chooses to live a happy life in Omelas does so with full knowledge of what makes that happy life possible.

            Why do some choose to walk away? To make clear that there is another option. The child did not choose to participate in this bargain, but the people of Omelas did—not just on a societal level, but every single one of them, individually (except the ones who walked away). That leaves no question about who bears moral responsibility for the situation.

            Why aren’t we told anything about how this bargain is possible? Because that would presumably introduce another agent (some kind of demon, maybe?) to bear that moral responsibility, and it must lie with the people of Omelas alone.

            Why doesn’t anybody free the child? Because if they did there’d be no Omelas, and no story. Yes, this is a cop-out; oh well.

            And finally, why don’t they look for another option? Because, again, that would remove the moral dilemma without answering it, when the point is to find the right answer.

            Of course these are all out-of-universe reasons rather than in-universe ones. I’m not bothered by things in fictional works that don’t have good in-universe explanations as long as there’s a sensible out-of-universe one, but if the out-of-universe explanation is also unsatisfying then I get annoyed. This may be unusual. My in-universe explanation for why they didn’t look for another option is that they did look for one, and probably still are looking, but it’s generally accepted that no other option is likely to be found.

          • houseboatonstyx says:

            Protagoras says:
            I can’t find even a hint of a suggestion that they’re not looking.

            I can. They make sure that everyone sees the child, but don’t offer an option of “Join our research team and help us find a solution.”

          • houseboatonstyx says:

            I’m not bothered by things in fictional works that don’t have good in-universe explanations as long as there’s a sensible out-of-universe one, but if the out-of-universe explanation is also unsatisfying then I get annoyed.

            The city rulers forcing everyone to see the child merely to make the citizens guilty, seems ‘authorial intrusion’ through the fourth wall. To protect suspension of disbelief, they should have some sort of in-universe motive as well. Perhaps the demon has made that part of the deal? Or it’s all some sort of psychological experiment, like Milgrim’s — in which case walking away could mean “This can’t be true, it must be some kind of hoax.”

            At which point the reader reaches Zen enlightenment.

    • Ken Arromdee says:

      My subversive interpretation of Omelas is that the child is a taxpayer. The country has a system of taxation which is more uneven than some (i.e. only one person is taxed, but at a very high rate), and like conventional taxes, the fact that the tax hurts some people is justified by the benefit to others.

      Of course, if every individual receives more benefits from taxes than they pay, this would not be true, but I don’t believe that for a second.

      • Carinthium says:

        That makes a lot of sense. The ordinary person would probably argue it was the child’s duty, and differentiate the situation by claiming an upper limit on how much a person can reasonably be allowed to sacrifice OR arguing that sacrifice must be ‘fair’ (in the sense that people sacrifice only a portion of what they have, not all of it, or that the sacrifice be balanced as a proportion of happiness lost. It helps that people choose to work hard to earn greater than average income whereas they don’t choose to be the child in this case).

        Justifying either is rather difficult, I agree- the former is arbitrary, and the latter is incoherent.

    • James says:

      I’m finding it weird that this post should pop up today, when I *just yesterday* read The Ones Who Walk Away From Omelas on the advice of a friend with whom I’d discussed utilitarianism.

    • pneumatik says:

      Walking away from Omelas is harder than it sounds. Everyone reading this probably resides in Omelas. The suffering child is making our smartphones one component at a time in factories with suicide nets underneath the windows. He is recycling the rusting hulks of the ships that bring the phones to us, hulks run aground on a beach where the child attacks them with poor equipment and no safety gear. We eat food that the child harvested all day in the sun. All this so that we can sit on the Internet and debate philosophy.

      The alternative is worse, of course. If everyone walks away from Omelas then there will be no Omelas anymore, but the child will still be locked in the basement.

      • Chris H. says:

        There is, of course, a third option, which is seek to free the child. What that means in practice in most situations in our actual world is a complicated question.

        Walking away is a bad solution, because it amounts to passive complicity, but that doesn’t mean that active complicity is the answer.

        • pneumatik says:

          The real world can’t afford to free the suffering child because the world he would be freed into literally depends on him. No electronics would be affordable without him, which means no modern world. Food would be expensive and only locally available. It’s literally not an option.

  11. blacktrance says:

    Contract theories don’t say that you can’t or shouldn’t help these bottomless pits, they only say that the extent to which you should care about them must be justified to you in a way that you would rationally accept them. This may be because helping others makes you happy, or because it reduces crime, or because it would be in your self-interest to have adopted a rule of helping bottomless pits from behind a veil of ignorance. This isn’t a bug of contract theories, but a feature – because you’re an individual moral agent, whatever you morally do must be justified to you in particular, not to some disembodied ideal observer as in standard utilitarianism.

  12. The Anonymouse says:

    If people do tend to prefer cathedrals to starving peasants a millennia away (and I suspect they do), the logical followthrough to that would have powerful implications to current questions, such as the funding of welfare programs versus manned spaceflight.

    • Carinthium says:

      If people do prefer cathedrals to starving peasants, that implies that current people have a bias towards people around in modern times. Why does that necessarily mean respecting the preferences of people in the far future?

      I figure most people already get this, but relative to cathedrals manned spacecraft are far less likely to be beautiful things our descendants appreciate. That being said, that doesn’t change the fact implications still exist.

      • MugaSofer says:

        >relative to cathedrals manned spacecraft are far less likely to be beautiful things our descendants appreciate

        Not if they live on the Moon.

    • Jaskologist says:

      I realize this is fighting the hypothetical a bit, but do we have stats on how many people were starving back in the cathedral days? Was the church actually choosing between a cathedral for many and starvation for a few?

      • MugaSofer says:

        I understand the medieval Church was, in fact, doing a great deal of charity work as well as building cathedrals. So presumably this dilemma must have come up at least once.

  13. I think that it’s possible to choose the starving peasants over the cathedrals while being, if not completely selfish, then at least unwilling to make significant personal sacrifices solely for the benefit of people who can’t be easily empathized with (like anonymous peasants who’ve been dead for hundreds of years). I find it plausible that today’s society might be much better off if the medieval church had chosen to help the starving peasants, because of flow-through effects.

    Of course this doesn’t in any way address the bottomless pit idea, since that takes place in the least convenient possible world.

  14. Auroch says:

    I can’t find it either, but it was definitely Brienne Strohl who posted it.

  15. AJD says:

    Utility monsters are basically one-man bottomless pits.

    Or to put it another way, Omelas is a reversed utility monster?

    (Hmm, I see Multiheaded already made this connection. Oh well, posting anyway)

  16. Hedonic Treader says:

    The pit gods agree to release some of their prisoners, but only for appropriately sumptuous sacrifices.

    Beware perverse incentives. If you pay torture threateners, you will cause more torture threats.

    • Chris H. says:

      True, but there are (in principle) no perverse incentives if you only pay people (or gods) to stop torturing who demonstrably began (and, if unpaid, will continue) torturing for reasons unrelated to the hope of being paid to stop.

      • Hedonic Treader says:

        Also true. I sometimes hand vegan chocolate out to beggars, friends and passersby.

        But even then, I usually don’t mention I do it to reduce animal suffering, and I worry there’s indirect backlash either way (after all, they’re the kinds of people who would torture, so if you transfer resource control to them, people who would torture gain in power and you lose in power).

        • Deiseach says:

          Vegan chocolate? Is that real chocolate, or that carob stuff?

          Because if it’s the ‘just like real chocolate honest’ faux-chocolate, then YOU MONSTER!!!! Deluding poor hapless folk into thinking they’re getting delicious chocolate and instead you’re giving them that thin brown yuck!


          • Hedonic Treader says:

            Ha ha ha 😀

            No, it’s just normal dark chocolate that afaics contains nothing out of animals.

          • nydwracu says:

            Vegan chocolate as opposed to…? I have a bar of Lindt 90% here, and the ingredients are: chocolate, cocoa butter, cocoa powder processed with alkali, sugar, bourbon vanilla beans.

            (The Adventist store used to have carob malt balls, and they weren’t bad, though I can’t imagine the early Adventists approving.)

          • Anonymous says:

            Most dark chocolate contains a bit of dairy. For example Baker’s 70% contains milk solids

      • DanielLC says:

        False. If the pit gods are good at predicting you, they will capture more prisoners based on how much they predict you will be willing to sacrifice to free them. If they gain no benefit from prisoners and predict that you won’t pay ransom, they won’t take prisoners.

    • DanielLC says:

      You’re anthropomorphising the pit gods. They aren’t going to capture prisoners in hopes of getting a reward. They just release prisoners when offered a reward.

  17. Halfwitz says:

    Shouldn’t the “utility monsters” of any sufficiently-advanced moral theory be indistinguishable from its prescribed world state? A paperclip maximizer isn’t fretting about utility monsters.

    • suntzuanime says:

      Yeah the “utility monster” paradox boils down to saying “what if there were someone we really cared about a lot, that would be awful because we don’t actually care about them that much”. A paradox born of hypocrisy.

      • Hannah Lewis says:

        The utility monster objection is a challenge to particular ways of assigning how much we care about various things, not to the idea of assigning care. If you’re thinking of assigning moral weight to happiness-moments or desire-fulfillment or suchlike and you’ve so far justified this by appealing to intuitions involving humans, it’s useful to consider whether the justifications still work out when exotic agents get involved.

        If you think you care about happiness alone and you think you care about people equally, considering someone who just feels happiness way more intensely than anyone else could hope to forces you to decide which of these principles is more dear.

        • Halfwitz says:

          Ok, so the utility monster is just a means of showing that your stated goals disagree with your intuitions. Philosophers don’t think it’s a critique of utilitarianism, just that it highlights the problems of commonly-proposed definitions of human utility. If the what-do-we-want problem were solved, no one thinks utility monsters would be a real problem, correct?

        • Said Achmiz says:

          If you think you care about happiness alone and you think you care about people equally, considering someone who just feels happiness way more intensely than anyone else could hope to forces you to decide which of these principles is more dear.

          It’s not clear to me why caring about happiness alone necessarily implies having to weight everyone’s happiness “equally”, in the sense that someone who “feels happiness way more intensely” should also count “way more” (?? how much more? the “same” proportion? how do you even measure this…?) in your aggregation function.

          Then again, feeling happiness way more intensely is actually not exactly the premise of the utility monster! The premise is: someone who derives way more utility from a unit of resource than other people. But does feeling happiness way more necessarily translate into deriving way more utility from things that cause you to feel happy? This is not even remotely obvious; if it’s VNM-utility we’re discussing, for instance, it is defined only up to positive affine transformation, so there’s no reason to believe intensity of happiness maps to increased utility — and in fact in such a case that (interpersonal) comparison is not even defined.

          • Hannah says:

            Yes, if a theory has something other than happiness as utility then the relevant utility monster has some other property that makes it derive more utility per resource than other agents.

            Caring about happiness (or desire-satisfaction, or paperclips) doesn’t commit you to caring about equality. It’s completely consistent to bite the bullet and say, “Well, this theory says we should spend all our effort helping the monster, despite all our intuitions to the contrary.”

            But if you don’t want to throw out all those moral intuitions — if you want to build a theory that never tells you to give up everything for the monster’s sake — then this has implications for what definition of individual utility you should choose and how you should go about aggregating it.

        • Paul Torek says:

          “If you’re thinking of assigning moral weight to happiness-moments or desire-fulfillment or suchlike…”

          Hannah, I think you’ve raised a crucial point, which deserves more discussion. What’s a happiness-moment? Do they have quantifiable intensities? (Would that be happiness, or pleasure?) Even if intensities are quantifiable, do we really think that (for example) having one second of 1-million-intensity pleasure is better than having 10,000 seconds of various pleasures of average level 90? (In one’s own life – just to stave off moral complications.) And even if we do think that, how do we measure achievements or knowledge or friendships against that scale? The concept of an overall “utility” seems too fuzzy to fit into the precise mathematics of a von Neumann-Morgenstern, or Savage, or (etc) type formulation. Nor does it seem to help to switch from pleasures to desires. (Nor do I find the VNM axioms, or minor variants on them, particularly compelling.)

          Re-thinking our conception of the well-being of individuals might make many of these “problems” disappear, in the sense of no longer being able to formulate the supposed problem.

      • Nestor says:

        It’s not hipocrisy to profess to love dear leader very, very much. It’s self preservation.

    • DanielLC says:

      It’s a typical paradox. It’s caused by two different thought processes converging on two different solutions of the same problem, where only one can be correct. If your mind was entirely in harmony with itself, you would see no problem, but that’s not how the human brain works.

  18. Anonymous says:

    This is not related to the actual dilemma being discussed but I find it bizarre that Scott chose as an example the ancient cathedrals which weren’t any more wicked than any present day government spending program; it would be much more appropriate to use NASA as an example of the same concept.
    To me this just sounds like a cheap way to attack the church.

    The ancient cathedrals were useful to the masses. People used them and enjoyed them.

    Space flight is pointless.

    Now that I think of it this is one of the reasons that the recently discussed communist economy (and generally big government economy) fails. Even if you are a brilliant optimizer if your goal is to get Gagarin in space (and other similar goals that are typical of the people in power but don’t make the plebs actually flourish) and you invest a gigantic amount of resources into this compared to the size of your economy then you aren’t doing your people a favor.

    • Matthew says:

      The ancient cathedrals were actually useful to the masses. People used them and enjoyed them.

      Most of the masses never got within sight of a cathedral in pre-modern times.

      • drethelin says:

        even if that’s true (I don’t feel like researching it right now) an estimated 20,000 people visit Cologne Cathedral every day now. That’s a lot of people’s pleasure to simply discount!

        • Anonymous says:

          You must discount it by the discount rate. It’s not much from the point of view of the 13th century. Maybe it was a good investment when construction resumed in the 19th century. Presumably it was restarted because it became cheaper, either in absolute terms because of technology, or easier to spread of more people because of population density.

        • DanielLC says:

          If it takes that long to pay off, there are much better investments.

      • Anonymous says:

        It was useful to the locals, who are probably the ones to pay for it (though I’m not sure). I’m pretty sure that 19th century cathedrals were funded by the locals.

    • Using cathedrals as the example wasn’t Scott’s idea; that was from the original discussion on Facebook. (I saw it on my own feed, but can’t find it myself either. Facebook needs to stop making stuff disappear or otherwise be hard to find later.)

      Those discussing the question of welfare programs vs. manned spaceflight might be interested in this account of an encounter between Ralph Abernathy (Martin Luther King’s successor as head of the Southern Christian Leadership Conference) and Thomas O. Paine (the head of NASA) in the days leading up to the launch of Apollo 11.

    • Bugmaster says:

      Space flight is pointless.

      Umm, I kinda like my weather prediction, global telecommunications, orbital telescopes, and even satellite TV — just to name a few of the benefits of spaceflight.

    • von Kalifornen says:

      If we are to take the nonexistence of God as a given, then the space program looks pretty nice if we actually pursue it.

      • BenSix says:

        Could you elaborate on that? Interested to know what you mean.

        (Also: the space program/poverty issue is very much a live one on the other side of the world.)

        • von Kalifornen says:

          Serious spaceflight research — with a lot of exploration of reducing cost of space lift and colonization technology — would pay off really heavily IFF we support it long enough to actually get a colony. (And can also help us live on an “uninhabitable” earth.)

    • MugaSofer says:

      >Space flight is pointless.



    • Anonymous says:

      Space flight is pointless.

      You won’t be saying that when you’re EXTINCT, will you?

      • RCF says:

        Extinction applies to species, not people.

        • Vulture says:

          I think anon was being flippant, but the point stands. Space flight is not irrelevant to existential risk.

          • RCF says:

            Anon phrased the issue of existential risk as an issue of self-interest, which it is not. And space flight is largely irrelevant to existential risk, at least at it currently stands. At the current level of technology, most existential risks are either not addressable by space flight, or addressable by massively cheaper options (scared that a supervirus will wipe out humans over the course of a month? Build some quarantined bases). A space base would cost trillions of dollars, and I would be hard pressed to think of an existential risk than a space base would take care of that an Earth-based one wouldn’t.

    • Emile says:

      Space flight is pointless.

      I, for one, would rather see much more of our resources dedicated to space flight, provided it’s optimized for long-term goals (like establishing a permanent base somewhere Up There) rather than short term things with no followup.

      (However, I suspect a lot of progress towards “getting a permanent base Up There” can be made without sending anybody up; for example by having a contest of who can get the most effective self-sustaining base under the sea or on the Antartic)

    • Illuminati Initiate says:

      “Space flight is pointless.”

      No. No it’s not.

      Most of the benefits are long term but you have to start at some point. And we already have GPS and satellite communication.

      IIRC many medieval cathedrals took like 150 years to build, so they were long term too. And unlike spaceflight their sole benefit seems to have been looking pretty. (even if you consider people worshiping to be a good thing, you don’t need a huge extravagant cathedral to worship your eldritch abomination of choice in)

      • MugaSofer says:

        >even if you consider people worshiping to be a good thing

        More people worshipping = more money for charity, for a start. Even ignoring the argument that you’re benefiting a lot of people (by a little bit) over time by making Art.

        If you’re choosing between feeding the hungry and cathedrals, the choosing to invest *some* of that money – rather than go broke feeding some small fraction of the poor – is not necessarily stupid.

    • Ghatanathoah says:

      >Space flight is pointless.

      At this point I think the enjoyment people got from watching the movie “Apollo 13” by itself it probably enough to justify our entire space program. We managed to create a historical event that nearly perfectly mimicked the sort of dramatic structure that human beings prefer in their stories. The fact that we did it by accident does not take away from its utility.

      And that’s just the tip of the iceberg. The entertainment value of the space program has been immense, both in the stories it has generated, and the new bits of scientific data it created for people to learn.

      There’s also the tech spin-offs, but I’ll give you that one and assume that they would have been invented soon enough anyway.

      • suntzuanime says:

        I liked Apollo 13 as much as the next guy, but was it really so much better than the next best movie that could have been made that it justifies the space program’s huge expense? Of course not.

        I think it’s fair to count the tech spin-offs as a legit benefit of the space program. Somebody else might have invented the stuff anyway, but they would have had to have spent money and effort on the project. The space program spared them that expense, so you can discount the expense of the space program by that amount.

    • Ialdabaoth says:

      Space flight is pointless.

      If Elua is the God of Humanity – the God of everything it is to be human, then space flight would be one of his most potent sacraments.

      A man can see the island,
      they say,
      through the mists,
      when the sun’s rays
      part them.
      No man can reach the island,
      they say,
      for Mother Sea
      is treacherous
      and deep.
      he ventures.
      His canoe is battered.
      His body is hammered.
      He stands alone
      on unfamiliar shores.
      He dies alone
      The next to come will be luckier.

      By twos they come now,
      and tens,
      and slowly
      so slowly
      The land is filled.
      The shore opens
      to a whole new place
      With new fruits
      and new animals
      and new dangers.
      He is not afraid.
      The other side
      of the Great Jungle
      can be seen,
      they say,
      if you stand atop
      the highest hill.
      No man can cross the jungle,
      they say,
      for Mother Earth
      protects her own
      with wild things,
      and poisons.
      Alone, he ventures.
      He is hungry.
      He is hurt.
      He is frightened.
      He stands alone
      before the greatest desert
      any man has ever seen.
      He dies alone
      The next to come will be luckier.

      They cross the deserts, now,
      in twos and threes
      On camels, wagons and tents.
      They reach far oases,
      build cities,
      erect temples
      and statues
      to those who came before.
      A thousand corpses litter the desert,
      ten thousand fill the jungles,
      a million at the bottom of the sea.
      The next to come will be luckier.

      The first ships
      to carry explorers,
      seekers of fortune
      and freedom…
      Not all who started the journey,
      finished it.
      The next to come will be luckier.

      It is for them that we do this thing.

  19. drethelin says:

    What if building seemingly selfish and wasteful cathedrals is extremely profitable in terms of positive externalities, and feeding the starving poor is beset with negative externalities? How do you even begin to measure the benefit accrued to millions of people over centuries from a cathedral? Isn’t that just privileging people alive now as opposed to people who will be alive in the future to enjoy the cathedral? What if the architectural advances that come from allowing kings to have Louvres and popes to have Vaticans generalize and lead to better structures for everyone?

    What if what we think we’re doing when we try and fill in a bottomless pit is like throwing wood into an ocean, when what we could be doing is building a beautiful galleon to cross it?

    • Anonymous says:

      Also – feeding the masses is dysgenic.

      What we should be doing with the poor is discourage them from reproducing, only then we should think of feeding them.

      • Multiheaded says:

        Yet reactionaries always freak out about emancipating women and sending them to college, which is seen to be a big thing in reducing fertility in developing countries; look at Iran’s demographic transition, for example.

        (But please substitute “overpopulation” for “dysgenics”, I do not wish to legitimate the latter.)

        • Anonymous says:

          Neoreactionaries freak out about emancipating women and sending them to college because the women who go to college are precisely the type of woman whose genes, we as a society want, and therefore the reduction in their fertility is a bad thing. College attendance filters women strongly for intelligence, conscientiousness, rule-following and low time preference etc all of which are valuable traits in the modern world and so in the long run, decreasing the proportion of these traits in the gene pool does not seem to be a good idea.

          As for overpopulation, it very much depends on who is doing the populating.

          • nydwracu says:

            I’m surprised Multiheaded hasn’t come up with a right-Communist elevator pitch based on the relevant West vs. East Germany data.

          • coffeespoons says:

            Reactionaries don’t seem to support policies (e.g. extended maternity leave, flexible working) that might make it easier for career women to have children.

          • Multiheaded says:

            Yep, state socialist countries generally had a rather wide range of measures to control demographics, even though they wanted nothing to do with hereditarianism.

            coffeespoons: and paternity leave is right out because ew, how feminizing!

          • Anonymous says:

            I agree about the maternity leave and support for working mothers. I think the reason neoreactionaries oppose these is because they come mainly from libertarian circles and despite the professed love for monarchy etc, they have an instinctive aversion to big government.

          • Hedonic Treader says:

            Neoreactionaries freak out about emancipating women and sending them to college because the women who go to college are precisely the type of woman whose genes, we as a society want, and therefore the reduction in their fertility is a bad thing.

            Laughably simplistic. Neoreaction is an intellectual joke.

          • nydwracu says:

            Reactionaries don’t seem to support policies (e.g. extended maternity leave, flexible working) that might make it easier for career women to have children.

            Those are probably preferable to the status quo; they’re also weak compromises. If both parents work full-time, the task of raising children will have to be outsourced, and, unless we in the Anglosphere overcome a thousand years of inertia and bring back the extended family, that most likely means outsourcing it to the Cathedral.

            Sending women to college is sending women to the Cathedral, and I don’t see what’s so great about people taking on a massive sunk cost to gain admission to a place where they’re cut off from their entire previous social environment, placed in a controlled one designed to propagate certain ideals, and exposed to rampant drug use and casual sex to simulate, Hassan i Sabbah-style, the liberated paradise to come.

          • Hedonic Treader says:

            Sending women to college is sending women to the Cathedral

            Richard Dawkins would probably say, “I don’t dress women. They dress themselves.”

          • Chris H. says:

            There is something very funny about a reactionary ranting about “The Cathedral” in a thread that is partially about actual cathedrals. This is not any sort of argument, but merely an observation.

        • MugaSofer says:

          “please substitute “overpopulation” for “dysgenics”, I do not wish to legitimate the latter”

          … whereas you you *do* wish to legitimate the former? I am genuinely confused by this statement.

          Overpopulation simply isn’t an issue. It flatly does not exist. Surely you don’t wish to associate with such a concept?

          • Multiheaded says:

            Local overpopulation, which economists would probably call something different. As in, too many people in a village, then some move to a city slum with too many people – not, like, straining the resources of a whole nation, just excess/unproductive labour power, etc.

          • MugaSofer says:


            Ah, I see. Yes, that’s a legitimate concern; sorry I wasn’t charitable enough to think of it.

          • Vulture says:

            Before this discussion goes any further, I have a critical point to make: Guys, the verb is “legitimize“!

          • Jack Crassus says:

            Thanks for all the effective altruism, guys. Why do smart utilitarians care so much about allowing the most miserable parts of the world to increase population exponentially? It gives you a challenging project to work on?

      • Alexander Stanislaw says:

        It is not dysgenic. Preventing children from starving to death causes their parents to have fewer children.

        • RCF says:

          Saving children increases their share of the gene pool. That is mathematically certain. There may be local violation of that rule, but it is not possible for there to be global violations.

          Note: I’m not using local/global in the physical location sense.

          • Chris H. says:

            Saving children increases their individual share of the gene pool. But saving children from demographic group X may still reduce demographic group X’s share of the gene pool if it proportionally reduces the number of children an average member of group X who lived through childhood has more than it increases the fraction of children in group X who will live through childhood.

            Which is not only empirically supported, but totally plausible, given that individuals may want a reasonable guarantee that some of their children will survive childhood, rather than a certain expected number. Conservative example: Suppose I want at least a 99% chance that at least one of my kids will live to adulthood. If the cumulative mortality rate from birth through 15 is, say, 10%, then 2 kids is enough, and the average number of surviving children that I have is 1.8. But if the cumulative mortality rate through 15 is 25%, then I need 4 — and the average number that survive is now 3!

        • Anonymous says:

          “Preventing children from starving to death causes their parents to have fewer children.”

          Correlation is not causation.

      • Illuminati Initiate says:

        First off, can someone explain this claim to me. It seems that improving living standards (for example, by making sure the poor have a steady supply of food) REDUCES reproduction rate, so how is it dysgenic?

        Second, correct me if I’m wrong but aren’t dysgenic/eugenic effects generally rather long term things? If it takes 100 years for them to show up it’s probably not an issue, because at that point society could probably just subsidize designer babies with +inteligence. (assuming it hasn’t had to restrict population growth already because of anti-aging technology. Edit: which actually solves the problem on its own also).

        • nydwracu says:

          Yes, let’s base all our plans around the assumption that the Second Coming will come real soon now. That can’t go wrong.

          • Illuminati Initiate says:

            Low level genetic engineering in 100 years is hardly the second coming.

            Edit: And I’m not really even against eugenics per se. I just don’t see it really being worth it. Well, except that I AM against letting people who carry things like Huntington’s reproduce unless they are willing to undergo embryo screening.

        • MugaSofer says:

          >It seems that improving living standards (for example, by making sure the poor have a steady supply of food) REDUCES reproduction rate, so how is it dysgenic?

          It’s dysgenic compared to, e.g., killing them. Not compared to making them sad.

          Now that you say it, this is probably the wrong comparison to make … except perhaps for the fact that the elites will always be less sad and thus reproduce less?

          • Chris H. says:

            It might dysgenic compared to systematically killing them, but that’s not what “let them starve” generally amounts to in practice. In practice, “let them starve” usually means “let a large fraction of them die, more or less at random, but not so large a fraction that it renders them unable to (over)compensate by producing more children in order to up the odds of at least one two making it through.”

          • MugaSofer says:

            Yeah, I think that your comparison is the right one. What matters is probably the effects of being awful to people, not the effects of killing them.

            Of course, it still probably doesn’t matter very much either way.

          • Anonymous says:

            I’m the one who first suggested that feeding people is dysgenic.

            I wasn’t necessarily thinking about starvation killing people, although this used to be an important mechanism through which the population number adapted itself to the food supply. I was rather thinking that I have much anectodical evidence of people making only as many children as they can afford to support financially, and then stop. Or people being motivated to make children by newly found prosperity.

            And I hear that many poor people make many children precisely because they want the extra welfare money that comes with more children.

            Also, for men, receiving welfare means higher social status (even if it’s just in the sense that your status isn’t as low as it would otherwise be) and thus more success with women and greater likelyhood of conceiving children genetically outcompeting other men.

          • houseboatonstyx says:

            That improving living standard reduces reproduction rates is not believable to me.

            That would depend on where the demographic is at start.

            In the worst conditions, where women have no better way to spend their time, some might use the welfare to choose more babies (or relax their contraception).

            A middle class family who wants the comfort of two incomes, will choose to limit new babies. At the higher income/comfort end, a family that prefers many children will add more.

        • Anonymous says:

          That improving living standard reduces reproduction rates is not believable to me. As far as I can tell the argument for this is that more developed countries have lower natality. However, correlation is not causation, and if you simply improve the standard of living of third world countries by giving them stuff, you aren’t going to magically bestow upon them all the other features of the developed world.

          I believe it’s the other way around – whether it’s welfare towards the poor within the first world or welfare towards the third world, it increases birth rates. I provided some mechanisms for this in a different anonymous post around here.

          – My anecdotical observation is that increased prosperity inspires individuals to have more children, and decreased prosperity inspires them to have less.
          – Poor women who make many children just so they can get more welfare. This is an example of welfare being dysgenic.
          – Welfare raises one’s social status. In first world welfare receiving males, this means greater reproductive success.
          – Third world famines that do not happen.

          • Nornagest says:

            if you simply improve the standard of living of third world countries by giving them stuff

            Simply giving stuff to third-world countries tends not to materially improve their standard of living in the long run. Serious foreign aid tends to fall into two categories: emergency response (e.g. food and shelter for refugees or disaster victims), or infrastructure and development assistance. The former is short-term by design. The latter works precisely by building in advantages that more developed nations already enjoy; this is, in fact, the implicit framework behind the words “developed” and “developing”.

            If you’re looking at birth rates specifically, though, I believe the usual claim isn’t about wealth in general but healthcare access and infant mortality mitigation in particular. Since this varies substantially relative to GDP per capita (Cuba for example is a poor nation with a reasonably good healthcare system), it shouldn’t be too hard to analyze as a natural experiment.

          • Anonymous says:

            When we look at a class such as first world countries, developed countries, rich countries, call them whatever you like, it’s risky to identify traits (in this case what you call “development”) and deem them the very essence of being first world. They could be just a symptom of other underlying traits. So the same still applies – I’m not convinced that you can turn third world into first world by giving them stuff such as infrastructure. Development from without simply is not the same as development from within.

            Cuba is more than Haiti plus hospitals.

            I’m interested in seeing statistics about help from the first world actually reducing third world natality. Until then I’ll be skeptical. Not that creating stuff in the third world is necessarily the wrong thing to do. But it would be wise to help the world’s poor while also trying to actively reduce their birth rates in more direct ways.

            Anyhow when I first brought up welfare being dysgenic I was thinking primarily of welfare within the first world.

    • Ken Arromdee says:

      Isn’t that just privileging people alive now as opposed to people who will be alive in the future to enjoy the cathedral?

      Scott also suggests believing a Robin Hanson line of reasoniong which concludes that preferences of past individuals especially matter.

      I personally think that all these arguments about what altruists should logically do, that conclude that everybody should do things that nobody would actually do, are really just arguments against altruism rather than arguments that we should do as many of those things as we can. One man’s modus ponens is another man’s modus tollens, you know.

  20. Richard says:

    Prisons aren’t that bad. No, really.
    If you prefer video this is amusing, or Michael Moore is always worth a laugh.

    From the outside, the problem with the US penal system seems to be that it is optimised for revenge, not for minimising crime. To me, this seems so utterly stupid that I can’t even begin to wrap my head around it. Why would any sane society optimise the penal system for anything other than reducing crime?

    • Fazathra says:

      Because we aren’t a sane society; we are a society where people vote for politicians who are tough on crime and the people running the prison system, judiciary etc lobby ceaselessly for the expansion of their bureaucratic empires.

      • Carinthium says:

        Actually, if deontology or SOME variants of virtue ethics are assumed this can make perfect sense. A Kantian, for example, considers it imperative to kill murderers regardless of pragmatic benefit. Similar sorts of deontologies could easily justify the prison system as it exists.

        Really it’s because people are more deontological than you realise. If you can’t reasonably expect them to see the problems with deontology (debatable but likely), you can’t reasonably expect them to see the problems with the prison system.

        • peterdjones says:

          I do wish you would.nt keep bringing in Kant.The deontologyin question has a lot more to do with the Old Testament than with anything that liberal enlightenment philosopher wrote.

          • Carinthium says:

            Kant typifies modern deontology far better than the Old Testament does. Since God clearly doesn’t exist, Kant also typifies the deontology with the highest degree of plausibility (an intuitionist system would be virtue ethical unless you want it to be deontological by definition).

            In this case in particular, Kant made a prime example. The Old Testament has references to justice, but it’s idea of justice is far less absolute and therefore far less deontological as God can break it.

          • peterdjones says:

            How can Kant tpify the average Texas voter who has never heard of him…but has spent many hours in Church?

          • Carinthium says:

            What’s more relevant than that is that Kant makes the best possible case for a deontological posistion that can be made, so I was using him as a defence.

            I mentioned that he typifies modern deontology because the man on the street is inconsistent even if in this case they are deontological.

          • Nornagest says:

            How can Kant tpify the average Texas voter who has never heard of him…but has spent many hours in Church?

            It’s entirely plausible to me that Immanuel Kant has more to do with the average Texas voter’s practical ethics than St. Peter, even if the average Texas voter has never heard of the former and hears of the latter every Sunday (or, more plausibly, on Easter and Christmas and one Sunday a month).

            I mean, I’m not sure I actually believe that, but the reasons I doubt it have more do to with skepticism of philosophy’s impact than faith in religion’s. Occasional kerfuffles over evolution aside, there is not much that’s specifically Christian about modern Western — even Texan — ethics.

        • peterdjones says:

          “The person” can see the problems with retributi’ve systems in some countries, so you are attributing too much to psychology, and not enough to sociology.

    • RCF says:

      Evolution has provided us with an extremely advanced System 1 capability for deterring defection. It’s hardly perfect, but it’s very powerful and readily accessible. Trying to appeal to System 2 makes people feel like you’re trying to get away with something. Moreover, trying to come up with a rational system for deterrence is very tricky. You have several different order of effects, and a perfect optimization system would have to deal with people gaming the system, and results that people would feel unfair.

    • DanielLC says:

      Signalling, for one. Someone who’s likely to go to prison has a reason to not want to be tough on crime, so everyone will try to be tougher on crime than everyone else so it looks like they’re not a criminal.

      Also, he wasn’t referring to “the world’s most humane prison”.

  21. Ken Arromdee says:

    Incidentally, I don’t think Robin Hanson’s argument applies here He argues that we should respect past people’s preferences about what should happen today, not past people’s preferences about what should have happened back then.

    I also don’t think it makes much sense. When considering past people’s preferences about what should happen today, you have to deal with compound discounting, not compound interest; a caveman’s preference that something happened today is discounted to almost nothing, not increased to 10^163.

    • RCF says:

      I think that Hanson pulled a fast one, conscious or not: he points out that the resources that we spend on doing what past people would have wanted should be discounted, but fails to note that the utility from doing it should also be discounted, which of course makes for a very lopsided case for doing what they would have wanted.

      And that’s on top of his idea that we can give someone utility by doing something that doesn’t affect them in any way.

      • Ken Arromdee says:

        Okay, that explains it. He was discounting only the resources, and I was discounting only the utility.

        That being said, we are deciding whether to use our resources to benefit cavemen, in which case we should use the loss in utility to us of using up the resources, compared to the gain in utility to the cavemen. The cavemen’s gain in utility is discounted; the loss to us is not. He seems to have this backwards.

  22. nydwracu says:

    The key word here is ‘investment’.

    Say you have one cathedral worth of money, and you can either donate it to the poor or build a cathedral. If you donate it to the poor, the poor get one cathedral worth of money. If you build a cathedral, you can use that cathedral to make money somehow: you can use them to impress church members into donating more, or inspire people to attend church more frequently and (in the current time) become Christian, and so on. (Given the collapse of Protestantism, which isn’t as inclined to glorifying God in that particular way, the Catholics are probably on to something.) And maybe the cathedral earns more than a cathedral worth of money in the end.

    As for the developing world: resources donated to the developing world can’t be used for investment to create new resources that can better aid the developing world. (There are also arguments that foreign aid is actively counterproductive, but I don’t know anything about them other than that they exist.)

    The pit god analogy fails on two counts: first, the developing world is not an inverse utility monster, so the utilitarian calculus doesn’t break in the ways that bottomless pits break it, and second, you can’t kill a god, so investment is futile.

    • von Kalifornen says:

      This is pretty much the explanation.

      Foreign aid often seems to 1. Be a bunch of Stuff We Don’t Need donated out of charitable ardor but without much consideration (esp. the case with disasters and clothing) and 2. can screw up economies, to point that leaders from poor countries will ask it to stop. If a bunch of food is put on the market, then prices can be very, very wrong. See “Please stop the aid”.

      I suspect that loans for investment, schools, and the like are very good. Space programs would be good if we actually pursued them — a new world would be extremely valuable.

    • Deiseach says:

      Why pick on past cathedrals? How about modern examples: should there be tax breaks and tax incentives for donations of art works to galleries? Should there be public art programmes at all? Why not seize the works in Museums of Modern Art, sell them, and spend the proceeds on welfare programmes instead, and tear down the museums to build affordable housing on the sites instead?

    • RCF says:

      SA isn’t concerned with “money”, he is concerned with utility. The Church doesn’t “make” money, it just causes people to transfer money to it. Unless the Church has a large advantage is turning money into utility (a position I would not expect SA to endorse), SA would not find spending resources to get people to donate money to the Church to be increasing total utility.

      • Nornagest says:

        the Church has a large advantage is turning money into utility (a position I would not expect SA to endorse)

        Now? No. Seven hundred years ago when most of those cathedrals were being built, though, you could make a better case for it.

        It wasn’t very effective in the modern sense, but the Church at the time was about the only organization that was into systematized altruism; in fact, it wouldn’t be too far wrong to say that the medieval church functioned primarily as the social-services arm of what passed for a government back then. In its social context, I’d say that’s a serious advantage in turning money into utility.

        • Jack Crassus says:

          Medieval European economic productivity was already far ahead of sub-Saharan Africa. To be “effective” in creating utility, surely medieval Christians should have staged pilgrimages to Africa to give out any excess wealth that they had instead of hoarding it and using it for things like Cathedrals and inventing the modern world.

          • Nornagest says:

            I don’t know a lot about premodern Africa, but pilgrimages were expensive in medieval Europe. Expensive enough that going on them was a status symbol well into the Renaissance — and I’m not talking going to the Holy Land, I’m talking going to a church a hundred miles away that claimed to enshrine the toenail clippings of some saint.

            Staging a pilgrimage to give away wealth, from a medieval perspective, would be kind of like delivering emergency food supplies by gold-plated limousine.

  23. DonGS says:

    You forgot one of the biggest ones: animal suffering in the food system. I readily admit to throwing a tarp over that one as I go about my meat-eating life, but I ain’t proud of it.

    It’s something I never really took seriously until reading about what egg-laying hens go through in factory farms — it’s horrific.

    • Vanzetti says:

      What wheat goes through on the field is also horrific… so I’m switching to nutrients 100% synthesized from inorganic matter.

      *dies of starvation*

      • JohannesD says:

        All food is 100% synthesized from inorganic matter.

      • James says:

        Are you just making a silly joke, or are you implying that it’s inconsistent to care about the suffering of animals and not the “suffering” of plants?

        For the record, I have no issue with silly jokes.

        • Vanzetti says:

          >Are you just making a silly joke, or are you implying that it’s inconsistent to care about the suffering of animals and not the “suffering” of plants?


          Seriously though, any discussion of what can and can’t suffer end with the oh-so-original realization that “suffering” is a qualia and cannot be measured.

          The real problem is not suffering but “OMG I can’t eat kittens they are too cute”. Well, that’s why you hire someone else to cook them. Or just eat the ugly, uncute kittens.

          • James says:

            Suffering can’t be measured accurately, but that doesn’t mean one can’t make educated guesses. I think we can do this by observing behaviours – certain behaviours (such as, say, cannabalistically pecking at one’s peers) seem to correlate convincingly with chronic stress/suffering. The other way we can do it is by analogy with our own experience – essentially the same leap that we make when we empathize with other humans: “I would suffer in situation X, and this entity is sufficiently similar to me in the relevant ways that I can presume it would suffer in situation X, too.”

            Clearly, a chicken or a cow is more analogous to you or I than an ear of wheat.

          • Vanzetti says:

            > “I would suffer in situation X, and this entity is sufficiently similar to me in the relevant ways that I can presume it would suffer in situation X, too.”

            OK, but what is sufficient similarity? Where do you draw the line? When does the fetus become a human being?

            This is a very old discussion. 🙂

          • Daniel says:

            Why do you assume a line is being drawn? I would put probabilities on certain things being able to suffer, and how much they can suffer. For instance, I think it’s very likely that cows and chickens suffer immensely in farms, but very unlikely that plants suffer at all.

          • Vegan says:

            Vanzetti says:
            “Seriously though, any discussion of what can and can’t suffer end with the oh-so-original realization that “suffering” is a qualia and cannot be measured.”
            “The real problem is not suffering but “OMG I can’t eat kittens they are too cute”. Well, that’s why you hire someone else to cook them. Or just eat the ugly, uncute kittens.”

            Vanzetti, the very same argument you’re making also implies:

            – that all forms of animals cruelty are prefectly okay, down to tasering animals for fun
            – that human suffering can’t be measured either and therefore the above also extends to fellow human beings which leads to complete sociopathy/psychopathy.

            Do you agree with those statements?

            “OK, but what is sufficient similarity? Where do you draw the line? When does the fetus become a human being?”

            Where do you draw the line, Vanzetti? You do draw a line, right? Or do you think that the only reason we don’t eat babies is that they are cute and therefore we should hire someone else to cook them? And what if someone used your own line-busting arguments (“sufference is a qualia and can’t be measured”) against your own line?

            Personally, speaking as an ethical vegan, I wouldn’t put the ethical problem in terms of suffering. After all the reason we don’t kill human beings isn’t simply that they would suffer. We abstain from killing human beings even if it’s painless. We don’t kill people swiftly in their sleep.
            The same applies to animals. The life of mammals and birds has a certain intrinsic value, much smaller but analogous to the value of human life. It is possible to kill them if there is a powerful reason that overwhelms the value of their life. I don’t deem eating meat to be a sufficiently powerful reason.

            Even if you don’t agree with the above, keep in mind that meat animals are killed in ways that make them suffer which is also an ethical problem, and by itself it does justify ethical veganism.

            In both cases, it’s about how much of something is there. The sufference of wheat is smaller than the sufference of mammals, and the value of wheat life is smaller than the value of mammal life. This isn’t rocket ethics; it’s intuitive. If we eat wheat, therefore, we cause less sufference than if we eat mammals, and we destroy “life” to a lesser degree. (and besides, all the calories and protein and other nutrients in animals ultimately come from plants that have been eaten, so there’s no way you can spare the plants). So, you see, you don’t even have to draw a “line” at all.

          • anon1 says:

            Animals are probably capable of suffering, so we should make some effort to avoid causing them to suffer. That doesn’t mean there’s anything wrong with killing one, unless some person is attached to it, and then the wrong is that you’ve harmed that person. My view on babies too young to speak is the same, except that people get attached to them so easily that after one is born, killing it would almost certainly actually harm a sentient person. So it’s usually wrong to painlessly kill a baby, and we don’t lose a lot by making a universal rule against killing babies.

          • Anonymous says:

            > “suffering” is a qualia and cannot be measured.

            I don’t suffer more when I get kicked in the face than when I relax in bed?

          • MugaSofer says:

            >The real problem is not suffering but “OMG I can’t eat kittens they are too cute”. Well, that’s why you hire someone else to cook them. Or just eat the ugly, uncute kittens.

            I know at least one vegetarian who has killed their own meat pre-vegetarianism. He’s ex-army; he has also mentioned in related discussions that he found it disturbing how easy it would be (ethics aside) to kill a human. (Part of the reason he left the army.) I doubt he became a vegetarian because of squeamishness.

            As a vegetarian myself (who currently lives in a rural area,) I’m pretty sure I *could* kill my own meat with minimal squick problems. Kittens or not.

            And most people – myself included – can simply buy factory-farmed meat if they want, which is even more horrific but much less squicky. I really don’t think your explanation is plausible, even taking the outside view.

          • Illuminati Initiate says:

            Your argument could easily be used to justify not caring about humans either.

            Also your model of people who don’t want to eat certain kinds of meat for ethical reasons does not at all account for how many of them actually act. I certainly don’t care how ugly the animal is, and neither do many other people in that category.

          • Ken Arromdee says:

            I would put probabilities on certain things being able to suffer, and how much they can suffer.

            If you do that, you end up saying that we should preferably do things that prevent the suffering of insects, worms, and other creatures that we normally don’t care about. The probability that they can suffer is so low, but there are a huge number of them.

            Also, I wonder why more animal rights activists don’t insist that we should prevent wild animals from being born in order to prevent their suffering in the same way that not raising cows means that cows don’t suffer.

          • Vegan says:

            To Ken Arromdee:

            “If you do that, you end up saying that we should preferably do things that prevent the suffering of insects, worms, and other creatures that we normally don’t care about. The probability that they can suffer is so low, but there are a huge number of them.”

            Even if an individual insect is able to suffer, its sufference remains much less important than that of an animal closer to us. Consider the difference in the number of neurons among many things.

            “Also, I wonder why more animal rights activists don’t insist that we should prevent wild animals from being born in order to prevent their suffering in the same way that not raising cows means that cows don’t suffer.”

            1 –
            2 – Who says that among wild animals sufference outweights happiness?
            3 – Animal rights activist are the wrong group for this. As the words imply they believe animals have “rights” which a sterilization of the wilderness would clearly violate.
            4 – Some value the environment.

          • DanielLC says:

            One simple test of suffering is if something starts trying to avoid whatever happened at the time of suffering. Since wheat can’t learn to avoid anything, it clearly cannot suffer. Or, at the very least, we have no evidence that something is making it suffer as opposed to feeling pleasure.

      • Deiseach says:

        I’m switching to nutrients 100% synthesized from inorganic matter.

        From Chapter One, “The Napoleon of Notting Hill”, G.K. Chesterton:

        Then there was the opposite school. There was Mr. Edward Carpenter, who thought we should in a very short time return to Nature, and live simply and slowly as the animals do. And Edward Carpenter was followed by James Pickie, D.D. (of Pocohontas College), who said that men were immensely improved by grazing, or taking their food slowly and continuously, after the manner of cows. And he said that he had, with the most encouraging results, turned city men out on all fours in a field covered with veal cutlets. Then Tolstoy and the Humanitarians said that the world was growing more merciful, and therefore no one would ever desire to kill. And Mr. Mick not only became a vegetarian, but at length declared vegetarianism doomed (“shedding,” as he called it finely, “the green blood of the silent animals”), and predicted that men in a better age would live on nothing but salt. And then came the pamphlet from Oregon (where the thing was tried), the pamphlet called “Why should Salt suffer?” and there was more trouble.

    • Daniel says:

      Why not just stop eating animals? Going vegetarian/vegan doesn’t have to be incredibly hard, and you can still be healthy and eat tasty food – for a guide on how to do so, see here.

      • MugaSofer says:

        Vegetarian, can confirm. I think the horror stories about how hard vegetarianism is are a part of that tarp.

        Then again, I once accidentally bought non-free-range eggs and felt physically unable to eat them, even though it didn’t do any good. Not sure how much of that is cause vs effect. I’m inclined to think it’s a virtue-ethical thing, building up the right habits … but maybe I’m just more motivated so it doesn’t seem as hard?

        • Vegan says:

          The strangest thing that I keep hearing is that vegan food is supposed to be very expensive, when in fact it’s very cheap – the most expensive foods are meat and dairy, if you do without them and focus instead on legumes-grains-veggies you save lots of money.

          • suntzuanime says:

            They’re referring to prepared/packaged vegan food. A box of Hotpockets runs like a dollar less than a box of vegan Hotpocket-equivalents. Sure, if you cook for yourself, like some sort of cavewoman, it might be cheaper, but if you want to reap the benefits of an industrial society?

          • Vegan says:

            I thought that the benefits of an industrial society was that everybody can afford a well equipped kitchen in their apartment with refrigerator and running water and dishwasher so you can enjoy cooking.

            The ancient Roman common folk did not have kitchen in their tiny homes and had to buy already cooked food and eat outside.

          • MugaSofer says:

            And the modern lower class do likewise, I’m afraid. (It’s supposedly a significant contributor to obesity.)

            If your budget is so tight you’re largely subsisting off fast food, that could be a real problem.

          • Vegan says:

            Industrial society even put in my kitchen magical flames that appear and disappear on command! The things you take for granted.

      • Anonymous says:

        Correct me if I’m wrong — and stop me if this is too derailing or off-topic — but aren’t things like animal fats very important, and aren’t non-animal sources of protein not great?

        • James says:

          There’s some truth in the idea that plant protein sources are less useful to the body (have to be broken down and rebuilt further, iirc) than animal proteins. It doesn’t mean they’re no good, though. I think it just makes it more important to make sure you get enough and a good enough range.

        • MugaSofer says:

          I don’t know about animal fats – I thought most people got too *much* of those, but I’m no nutritionist.

          Protein, however, does mean you have to make a point of eating more beans, nuts, and other protein-containing foods than a non-vegetarian. (And eggs, if you’re not vegan.) At least, I assume so; I’ve never run across a vegetarian with protein deficiency in person.

          Well, or you can cheat and eat that Quorn stuff. I’ve seen people mistake it for real meat while in the middle of eating it, and I’m assured it contains all the protein but less fat. It’s more expensive than meat, though, so that wouldn’t work if you’re on a tight budget.

          • BenSix says:

            I don’t know about animal fats – I thought most people got too *much* of those, but I’m no nutritionist.

            I can’t judge whether or not they are dangerous either but they are not essential. Even if one needs sat fats they can be obtained from coconut and dark chocolate.

        • Chris H. says:

          As I understand things:

          Getting enough omega-3 fatty acids is clearly quite important, but can be done entirely with plant sources.

          Getting enough long-chain omega-3 fatty acids might well be important, and is not realistically possible with plant sources at this time. But land meat generally doesn’t provide much of those anyway, and 2 modest servings of fatty fish a week is probably plenty.

          Regarding protein: Animal-source protein typically contains essential amino acids in ratios that are closer to those a human being needs than plant-source protein (although, IIRC, there are a few exceptions). This just means you need to eat more than one (type of) source of protein if you’re going to eat vegan (or near-vegan), and possibly to eat slightly more total protein than you would have to otherwise. If you eat respectable amounts of both legumes and grains, any legumes and grains, you should be fine.

          There was at one time a semi-popular belief that it was necessary to eat “complete protein” within a given meal, which would add to the hassle a bit, but the evidence is rather unambiguous that this is not actually true. Eating exclusively grain protein for weeks and then switching to legume protein is not viable, but essentially any reasonable eating pattern is fine.

        • Vegan says:

          “Protein” is not the issue, individual aminoacids are.

          You have a daily requirement for every one of the essential aminoacid. However, this is very easy to meet whether you eat vegan or not. The requirementes are actually surprisingly low and easy to meet. Other people in this thread who said that on a vegan diet it’s harder to get protein and you need to be very conscious of the protein you eat are exaggerating. In reality it’s actually difficult to fail to get enough of all essential aminoacids. It is more of a problem (whether you’re vegan or not) to get enough minerals and vitamins.

          Animal fats, there’s a lot of talk by groups who advocate certain diets (paleo, low-carb, weston price) about the importance of animal fats as a source of certain vitamins (A, E, K, D). In reality those vitamins, except for the vitamin D, are even more abundant in vegetables. Vitamin D is best obtained from either supplements or sunlight, and most people get most of their D from those sources anyways – very few people get their vitamin D from animal fats.

          There is some controversy as for whether the omega-3 fatty acids contained in plants (requiring conversion into long chain) are as useful for the human body as those in fish (already long chain). I believe that the experiments that suggest that plant omega-3 is not useful are flawed because they were performed on people who were not deficient in omega-3 to begin with, and you can’t expect the body to necessarily take the trouble to process chemicals it doesn’t need. It’s the same with beta-carotene – it doesn’t get converted by the body to retinol if you don’t need retinol. We can infer that plant omega-3’s are sufficient because long term vegan don’t get symptoms or diseases associated with lack of omega-3, and the milk of vegan moms is not deficient in omega-3. Having said that, if you are in doubt, you can always take supplements of the long chain type of omega-3 (vegan ones are made from seaweed).

          The REAL issue – the thing you might be missing from a vegan diet – is vitamin B12. When eating vegan, it’s a necessity to supplement with vitamin B12. (the supplements are not made out of animals).

          • anon1 says:

            > In reality those vitamins, except for the vitamin D, are even more abundant in vegetables. Vitamin D is best obtained from either supplements or sunlight…

            Mushrooms become an excellent source of vitamin D if they’re exposed to sunlight, and they continue producing it after they’re picked. Outdoor-grown, wild, or sun-dried mushrooms are all good for this.

          • Daniel says:

            Re Vitamin B12: A delicious solution is to eat a fair bit of nutritional yeast, which is tasty, cheesy, and tends to be fortified with B12.

          • Anonymous says:

            What about people who need significantly more protein than the average person? For example, I have enough muscle mass that I am nearly considered overweight, yet I have very low bf. I do competitive bodybuilding. Is plant protein sufficient? I ask because I have never seen a vegetarian or vegan bodybuilder, weightlifter, or powerlifter who was both 1) good at what they did (looking good, weightlifting, powerlifting) and 2) not on PEDs.

          • J. Quinton says:

            For example, I have enough muscle mass that I am nearly considered overweight, yet I have very low bf. I do competitive bodybuilding. Is plant protein sufficient? I ask because I have never seen a vegetarian or vegan bodybuilder, weightlifter, or powerlifter who was both 1) good at what they did (looking good, weightlifting, powerlifting) and 2) not on PEDs.

            My BMI is 29, which is one… unit of BMI… from being technically obese. Yet my bodyfat percentage is about 13%. I’m not a professional bodybuilder, I just like doing it.

            And yes, most other bodybuilders I’ve met or whose writings I’ve read treat being vegan and being a (natural) bodybuilder as some sort of sick joke.

          • Vegan says:

            There are vegan protein powders, such as this one:


            I have no idea if it’s the best – it’s just the only one I’ve tried – but I’m pretty sure that with that kind of stuff, it’s easy to get any amount of protein as a vegan.

            How much protein do you need anyways? Plenty of vegan foods have 25% or 30% of calories from protein and it isn’t hard to compose a food regime full of those foods.

            I believe that the reason that bodybuilders despise veganism and there aren’t many vegan bodybuilders is simply that bodybuilding and veganism attract opposite character types.

            I’m not into bodybuilding so I can’t tell whether someone is on PED’s, but I know there are strong vegans.


          • Nita says:

            What about people who need significantly more protein than the average person?

            Such people tend to supplement their diet with protein powder, which is either lacto-vegetarian (whey) or vegan (soy).

          • Vegan says:

            There are vegan protein powders not made out of soy. The one I linked to in my previous post is from rice protein.

        • DanielLC says:

          How much effort do you currently go through to have a good diet? It might sound okay to say that you care more about your own health than you do about the suffering of wild animals. But if you’re like most Americans, you care more about tasty food than you do about your own health, so what you’d be saying is that you care more about tasty food than you do about the suffering of wild animals.

          More to the point, from what I understand, there’s a list of amino acids you need that you can’t produce yourself. There’s one or two that’s particularly common among animals, beans, and nuts. Just make sure you replace your meat with beans and/or nuts, and you’re find as far as protean goes.

          I’m told you need vitamin B-12, which you can only naturally get from animals and mushrooms, but it’s easy enough to farm the bacteria that actually produce it in animals so buying vitamin B-12 is cheap, and it’s water soluble so it’s basically impossible to overdose.

  24. von Kalifornen says:

    Here’s my interpretation: The problem is that many pits are places where you can end up throwing resources into indefinitely for only linear returns. On the other hand, a lot of other projects (FAI, if you have hope for it, is best example, and so are most types of research and development) have the potential to give much greater benefits.

    In particular, one risks entering a “Last Men” situation where a country is so dedicated to providing for everyone and saving people from pits that it *never* advances. I think that some of the best-to-live-in modern nations are like that, and that they may either be crushed by Moloch or simply overthrown if a collapse comes.

    Consider the modern world. I think that generic feed-the-poor type charity or welfare is fine in developed countries because the pit isn’t bottomless and industrial civilization is strong. But doing this globally, in addition to any sorts of malthusian or perverse incentives issues, can just consume a ton of excess resources for only saving (resources / cost per person).

    Some things can really help though. Promoting birth control in impoverished lands will shrink the pit, as will demographic transitions that cause wealthier countries to have less population growth and live below the malthusian limits. I’d suspect that medical stuff to prevent disease from stopping progress also helps, as does providing capital (that might be the biggest one.)

    OTOH, many things we build are valuable for a seemingly indefinite period, like cathedrals. A space program, if bravely pursued, can give us a whole new world. So they can seem preferrable to the Linear Charity Waste.

    One other issue: I think that cathedrals and the like are also a bit complicated because building such things today would be vastly less expensive and take much less fraction of our capacity. And it’s hard to tell how much we would be able to take up our capacity.

  25. Ken Arromdee says:

    I think that some of the best-to-live-in modern nations are like that, and that they may either be crushed by Moloch or simply overthrown if a collapse comes.

    Or they just mooch off of the countries that don’t do it so much.

    It doesn’t matter that there’s no innovation because of your insanely high tax rate, if there’s innovation in another country and you get to read their scientific papers.

    • peterdjones says:

      “Mooching” off patents happens in developing countries that don’t respect, and can’t afford generous welfare systems…and are rapidly growing.

      Otoh, the Sweden style countries do respect patent law. Some are innovative , too.

      Germany and Japan both have extensive welfare systems, and both are highly innovative as judged by patents filed,

      • Vanzetti says:

        Erm, the entire world mooches off the drug development in the USA. Arguably, the western world also mooches off USA defense spending.

  26. Vanzetti says:

    Utilitarianist philosophy, as it is “practiced”, seems to start with a vision of an ideal world where there is One True Utility Function, and then break down and cry uncontrollably when encountering the endless paradoxes this vision entails.

    Maybe it should, instead, start from the real world, where each agent has his own utility function, which also changes over time, and work from there…

    • MugaSofer says:

      In my experience, everyone has their own utility function, but those utility functions are altruistic – they include terms for “satisfy everyone else’s utility function too”.

      Not by coincidence, this makes it much easier to cooperate if we just all cooperate to maximize the collective utility function. (Usually this happens in small clumps, but that isn’t an inherent limitation.)

    • JME says:

      I already always see utility functions being assigned to individual agents. (In fact, I’m not even sure a “one true function for everything” function is even considered a utility function. It might be considered a social welfare function instead — although that distinction seems to be more in economics (or perhaps political economy) than philosophy.) What do you have in mind when you say Utilitarianism starts with a vision of an ideal world where there is One True Utility Function?”

      • anon1 says:

        The One True Utility function is that you must maximize the total, or average, or some function anyway, of all the individual utility functions. And therefore actually caring about your own utility is morally abhorrent.

        • JME says:

          I’m not saying that assignment of individual utility functions to individual agents implies that “every individual agent maximize her own utility function” is the correct ethical principle. You could have each agent have an individual utility function, and still think that morally, we should consider all utility functions. I’m just saying that, in practice, I have generally seen utility functions being assigned to individuals, so I’m not sure what the basis is for Vanzetti’s claim that utilitarians want to have one big utility function rather each agent having his own utility function.

          • Vanzetti says:

            > I’m just saying that, in practice, I have generally seen utility functions being assigned to individuals, so I’m not sure what the basis is for Vanzetti’s claim that utilitarians want to have one big utility function rather each agent having his own utility function.

            Well, my observation of utilitarianism in “practice” consists mostly of reading lesswrong. And there it is always about “humanity”.

        • RCF says:

          But my utility function is one of the utilities that goes into the One True Utility. So why should I not care about it?

          • anon1 says:

            You should care about it, but no more than you care about any one of the other billions of people in the world, so it rounds down to nothing. The only real way to justify taking care of yourself or trying to be happy, if you really believe in standard utilitarianism, is if you have determined that it makes you able to earn more money to give away. If you’re doing better than skating on the edge of poverty and burnout in order to save the people in the pit, you’re probably doing a terribly wrong thing.

            Some people seem able to cope with that, but I can’t.

  27. Nestor says:

    Aren’t cathedrals (Pyramids, misc “prestige projects”) a form of indirect social welfare? All those workmen are getting paid while gainfully employed. The distinction between “giving the people money” and “giving the people money to build a nice building” is somewhat meaningless.

    • Anonymous says:

      Except that at the end of the former you get nothing while at the end of the latter you get a nice building.

      • Nestor says:

        At the end of both you get live people with full bellies, in the latter case probably is an easier sell to the authority ponying up the money as they end up with a positional good to reflect their status. Whether cathedrals/pyramids/etc are actually worth anything in themselves besides that is a matter of opinion/perspective

        • Anonymous says:

          “Whether cathedrals/pyramids/etc are actually worth anything in themselves besides that is a matter of opinion/perspective”

          True, but so is having people with full bellies.

    • RCF says:

      I hardly find the difference between people working all day on a cathedral, and being able to spend the time doing something else, to be meaningless. And there were other resources that were consumed.

      • Nestor says:

        The workers may feel fulfilled makign the Cathedral too, if they are believers or using their specialized skills. Also the money trickles down into the community, the persons feeding the workers, the support infrastructure, the town pub.

        • RCF says:

          That’s a variant of the broken window fallacy. Just because more money is moving around doesn’t mean things are getting better.

    • DanielLC says:

      If you give away money, you can help the poor stretch the money further that they get from what work they can find. If you hire them, they can’t work anywhere else. You have to give them enough to live off of.

  28. Martin says:

    A few years ago, I thought of a bottomless pit-type problem with utilitarianism. I started with average utilitarianism, on the grounds that the basic moral entities are people — in the sense that people, not societies, are the ones who feel happy or sad; a thousand slightly happy people may collectively amass more happiness than a hundred really happy people, but there’s no-one to feel the aggregate of that happiness.

    Thus, it would seem to be a good idea to reduce the population of Earth, even if people are quite happy on average, and replace all sentient life with a few genetically engineered super-happy beings — the average happiness level would increase tremendously.

    However, if there’s a planet on the other side of the universe populated by billions of miserable creatures, then our planetary increase in average happiness will be at the cost of a universe-wide decrease in average happiness. It seems weird that a course of action can be moral or not depending on the existence of people who you can never be sure exist and whom you can’t affect in any way.

    • suntzuanime says:

      Take it one step further: it’s real people, not potential people, who feel happy or sad. A thousand slightly happy actual people amass more happiness than a hundred really happy hypothetical people, because they’re real, and fuck you if you want to kill them for the sake of fake people.

  29. Jaskologist says:

    FWIW, this debate is as old as Christianity itself (or Judaism, for that matter).

    St Basil (late 300AD) gave the following sermon inveighing against pretty much every luxury:

    He who strips a man of his clothes is to be called a thief. Is not he who, when he is able, fails to clothe the naked, worthy of no other title? The bread which you do not use is the bread of the hungry; the garment hanging in your wardrobe is the garment of him who is naked; the shoes that you do not wear are the shoes of the one who is barefoot; the money that you keep locked away is the money of the poor; the acts of charity that you do not perform are so many injustices that you commit.

    • Jaskologist says:

      I think the answer probably lies in Martin Luther’s idea of vocation. Different people are called to do different things. A lot of those things are mundane, and not everyone is called to save the world, or preach, or whatever. This view probably requires an understanding of God imbuing everyone with a telos, though. If you take a purely materialistic view of the world, I’m not sure how you avoid viewing humanity as just another machine that needs to be maximally optimized.

      (Given the context of the post, it’s worth further pointing out that Luther, whatever his views on vocation, started the Reformation in response to the practice of selling indulgences, which were financing the building of St. Peter’s Cathedral.)

  30. Johannes says:

    If just another machine, why should it be optimized? If something should be optimized, sa telos is presupposed and even another objective, that it should be optimized with respect to this telos. It is a very curious thing that many people apparently think utilitarianism was somehow “more modest” or “materialist” or “naturalist” compared to other ethics (like Kantian or Aristotelian). It is not.

    • Jaskologist says:

      In theory you’re right, but in practice I find materialists are notoriously bad at following the implications of their thought all the way to the end. Only the telos that they find inconvenient gets discarded, and now they are free to munificently mold humanity to their liking, without being held back by minor concerns for individual humans.

      (Also, people think utilitarianism was more materialist because it literally believed that we could find objective numbers for human happiness and plug them into a formula to solve morality forever, like we do in physics.)

  31. Emile says:

    I think that for the medieval churche, the decision to build cathedrals rather than feed the poor was probably the right one: partly because it was and still is a socially useful building (previously as a refuge, a place for large gatherings, etc.) and partly because they got money for the cathedrals mostly from the status contest between the burghers of various cities who wanted to show how their city was *the best* – the church wouldn’t have been able to collect as much money if it’s only use for it was “give it to the poor” (this is a gross simplification and a good deal of speculation, I’m not aware of the details of medieval church finances).

    HOWEVER, that money may have been better used invested in reducing poverty in the long term, for example by inquiring into the causes of poverty (instead of taking it’s existence as granted), experimenting with various programs and seeing which ones tended to work, advising rulers, and maybe even engaging in Bene Gesserit style selective breeding to reduce the prevalence of some causes of pain and misery (heck, the church already had a way of removing people from the gene pool – offer them a somewhat comfortable(ish) life as a monk).

    That would have made for a good alternate history…

  32. Shmi Nux says:

    Just wanted to note that most of the actual bottomless pit-looking examples you give can be made shallow-bottom by adding some transparency. If the suffering in nursing homes and prisons is plainly visible, things will change pretty quickly without major expense.

  33. houseboatonstyx says:

    One point I have not seen covered in this discussion so far. An individual with the virtue and/or deon bias, plus a strong local-utility ethic, is more likely to look for an effective solution for stopping or mitigating that Omelas practice, than someone who simply stays or simply walks away.

    Walking away in hope of finding an effective rebel group may find none. Or if found, it is not likely that this one walker will be important to the rebel group’s success.

    Staying and looking for ways to subvert or reform the system, is likely to turn up some useful information.

  34. suntzuanime says:

    I like this blog because it seems perfectly normal to have a post titled “Bottomless Pits of Suffering” right after a post titled “Cuddle Culture”.

  35. Helping the poor versus cathedrals is Easy Mode for geeks. Hard Mode is helping the poor versus science (although geeks tend to be ignorant of the extent to which the Church supported the precursors to modern science).

    • Alejandro says:

      Even though an atheist, I actually find the dilemma in principle harder with cathedrals. If the science has practical applications such that an expected utility computation finds it helps more the poor in the long term, then making the choice to support it is easy; and if it is pure theoretical knowledge, then eliminating poverty is a higher priority and the discoveries will be done eventually (the universe is not going anywhere).

      Creating something beautiful, that expresses our deepest values, and that can potentially last for millennia, is a much more incommensurable thing to weigh against helping the poor.

      • Eli says:

        Creating something beautiful, that expresses our deepest values, and that can potentially last for millennia, is a much more incommensurable thing to weigh against helping the poor.

        It’s really not. Your mind just trips up when it encounters words like “millennia” or “deepest values” or “eternity”.

  36. I mean, this is kind of the standard view of history. Except that in the standard view, conservatives tack on “But really, the bottomless pit wasn’t so bad, and the sulfurous flames gave you a nice, warm feeling inside.” And leftists tack on “but in the end, everyone including the people in the nice town benefitted from the increased understanding and diversity this created, so really history was just this series of obvious win-win propositions that everyone was just too stupid to figure out, until now.”

    With respect to the human rights abuses (in first-world societies) perpetrated on women, blacks, gays, and other who were punished by law, I think what actually happened is that the ‘nice townspeople’ had a deep (but not bottomless) pit of suffering that they maintained just outside town, and threw members of these groups into. The simplest way of fixing the problem is to let the people inside out, stop throwing more in, and collapse the damn pit (i.e., change the laws that force individuals of these groups to suffer).

    The conservative and leftist narratives are both wrong; it’s wrong to throw people into the pit, and it’s wrong to take the money of the townsfolk (and their descendants to the Nth generation, who are horrified by pits and would never throw anyone into them, and never have) and give it to pit-people (defined as anyone who would have been thrown into the pit once upon a time), and force them to live next to pit-people they don’t want to have anything to do with.

    • MugaSofer says:

      >it’s wrong to take the money of the townsfolk (and their descendants to the Nth generation, who are horrified by pits and would never throw anyone into them, and never have) and give it to pit-people (defined as anyone who would have been thrown into the pit once upon a time)

      I don’t think this is what’s happening. Although I’ve encountered a few SJWs who wish it was.

      • jaimeastorga2000 says:

        Food stamps, affirmative action, tuition grants, etc… are basically this, in practice if not in theory.

        • Illuminati Initiate says:

          Affirmative action can act like that sometimes*, but the other things you listed are given based on actual disadvantages, not disadvantages of ancestors.

          *Affirmative action is intended to compensate for current discrimination, but it does sometimes benefit people solely for having pit ancestors.

  37. grendelkhan says:

    It’s true in the Least Convenient Possible World, but it isn’t in this one: prison, especially longer prison terms, is one of the most expensive, inefficient ways to reduce crime. (For instance, people generally commit most of their crime during a certain age, and don’t have a very well-developed time preference at that point. Doubling the sentence costs twice as much, but doesn’t cut crime twice as far.)

  38. Robert White says:

    It’s a false dichotomy. The problem isn’t “help the poor vs support the arts”, it’s “support the people” versus “attack, coerce, and politic all the non-pious”.

    It’s not like “the church” ran out of money doing the one or the other, they had the money to do both. And they largely did what the understood at the time to be of value regarding the endless tide of the poor all while building those cathedrals and they _still_ ended up being the largest property holder and richest central bank in three continents.

    The problem is the horrific waste of manpower, not money, involved in doing things like trying to re-take The Holy Land™ (compare to trying to drive The Gays out of Disneyland for a contemporary act of religious futility).

    If all those people had been studying and experimenting on how to improve crop yields or bring education to the masses; if they’d been working to democratize the feudalistic system; if they’d been less than an Enormous Bag Of Dicks to people who _didn’t_ follow their dogma; then things would have been better.

    Some few were doing that sort of work developing thought and practise, and we remember them fondly. But far too many were doing quite the opposite.

    Unfortunately the underlying premise of religious thought, the idea that the faithful posess an insight that no others posess and that forcing others to accept that insight is the only way to prevent eternal torment or other equally egregious suffering, is anathema to progressive and helpful thinking. After all, if you think you have The Final Answer™, and you think every moment of life is being optimized by an omnipotent overseer, then the intermediate questions of mortal welfare become immaterial distractions at best.

    Of course that sort of thinking is _impossible_ for the church to avoid. Its core message is “we have all the answers” so it almost universally excludes any attempt to _really_ question an enormously large number of things taken as axiom. The poor had “their role, as determined by god” and the bible contained All The Truth™. There were exceptions to this mental enslavement, but far too few, and many of them came to terrible ends.

    For instance the Jews knew to isolate the sick and drain the boubles during the plague. They and the Muslims beleived in cleaning the body in a way that metropolitan christians did not (because christian cities didn’t really have clean water etc). The Jews had seen the plage before and their traditions retained the memory. Since fewer Jews were dying of the plague the were villified as in league with the devil or as the bringers of the plague, and the church _refused_ to take the advice brought by the Jews. That’s some fine thinking there Lou. But it’s enevitable thinking because of the underlying dogma.

    Fewer cathedrals and more poor houses, or vice versa, would have had zero impact on the outcome of the age. More tolerance and less dogma would have been a godsend.

    • Susebron says:

      It’s a false dichotomy. The problem isn’t “help the poor vs support the arts”, it’s “support the people” versus “attack, coerce, and politic all the non-pious”.

      Regardless of the reality, that isn’t the point of the example. The question is actually “poor vs. arts”, because that was the underlying intention.

    • suntzuanime says:

      The Crusades to retake The Holy Land™ were the original “fight them over there so we don’t have to fight them over here”, with The Holy Land™ acting as a Schelling Point to coordinate the armies of diverse Christian kingdoms against Islamic imperialism. They didn’t manage to save Greece or Anatolia, but they saved Iberia, Italy, and the Balkans, so I wouldn’t call it “futile” at all.

      • Matthew says:

        The Crusades were more like a)a clever method for winning the struggle to be the legitimate pope and b)a useful outlet for younger sons of the landed class in a culture of primogenitor. I think you’re taking the naive view here.

        • Chris H. says:

          They can’t have been both? The high water mark of Islamic expansion into Eastern Christendom was in 1529, with a reasonably close reprise in 1683.

          • Carinthium says:

            On a side point, historically speaking the postulation that the Crusades were doomed to failure in the Holy Land doesn’t make that much sense.

            Admittedly they were a long shot from a long-term perspective. But there are plenty of times things could have gone the other way- Barbarossa’s thwarted invasion force, the Christian invasion of Egypt (dumb though it was, if it had suceeded things would have gotten better), Saladin’s existence (if for some reason he doesn’t come to power things are much better).

      • Multiheaded says:

        Did Sister Y teach you how to troll like this?

        • Chris H. says:

          You don’t think that Alexios I Komnenos’s plea for help against the invading Turks had anything to do with the launching of the First Crusade? Or the continuing warfare in Muslim-conquered Iberia? Or fears that the rapid Muslim expansion in the East would not stop until it was checked, as Charles Martel had checked it in the West?

          • Multiheaded says:

            You don’t think that Alexios I Komnenos’s plea for help against the invading Turks had anything to do with the launching of the First Crusade?

            If the Crusaders gave a nonzero amount of fucks about him, they probably wouldn’t have sacked Constantinople.

            Or the continuing warfare in Muslim-conquered Iberia?

            So why didn’t they, y’know, go there instead?

          • Chris H. says:

            The First Crusade was a response to Alexios I Komnenos’s pleas for help (as well as several other factors), and took place in 1096-1099. The Fourth Crusade sacked Constantinople in 1202, 84 years after Alexios I’s death, and was an amazing clusterfuck on a whole host of levels. Also, the Crusaders initially retook Constantinople for the Byzantine claimant they supported without great destruction, did serious damage after the populace began rioting against them and killing their co-ethnics, and sacked the place after the populace revolted, killed the Emperor they were backing, and installed another, who was unwilling to give them what they needed for the Crusade to remain functional, let alone what they had previously been promised.

            None of this is to excuse the Fourth Crusaders’ atrocities or stupidities. It is merely to point out that the outcome of the Fourth Crusade was not what was planned even at its own launch, let alone at the launch of the First Crusade.

            And Christian kings were fighting Muslims in Iberia — and making progress. Meanwhile, half the Byzantine Empire had been swallowed up in the space of a few decades. Which front more desperately needed reinforcement?

      • ADifferentAnonymous says:

        A professor I had claimed the Crusades were at least partly a scheme by the Vatican to get the European nations to stop fighting each other so much.

        • Chris H. says:

          Really, the Crusades were an impressive convergence of several different goals held by different people and having (at least in principle) very little to do with each other.

          • Vegan says:

            Any explanation for “the Crusades” must begin by noting that the when they took place, the church had been exhorting Christianity to stop fighting each other and go fight the common enemies for centuries, and continued to do so for centuries afterwards. The crusades were just a subset of church sponsored wars.

        • Vegan says:

          That was a recurring theme in the middle ages. It was not a conspiratorial scheme, it was explicit: fellow Christians, stop fighting each other and go fight the infidels.

    • MugaSofer says:

      >the underlying premise of religious thought, the idea that the faithful posess an insight that no others posess and that forcing others to accept that insight is the only way to prevent eternal torment

      bursts out laughing

      You … really aren’t familiar with a wide range of demographics, are you?

      I assure you, this is not a law of nature, it’s a custom of your tribe.

      • Chris H. says:

        That’s perhaps a bit strong; I’d say there are a number of basic patterns that religion can naturally develop given human psychology, but which ones it does varies from religion to religion for a variety of reasons. To identify the need to save the heathen through violence as a distinctly Western notion is to ignore its prominence in Japanese Buddhism at various times, particularly during the the lead-up to World War II and the war itself. (On a vaguely related note, the remarkable similarity of key aspects of Jodo Shinshu and Calvinism probably says something both interesting and unpleasant about human nature.)

  39. Zanzard says:

    I’m sure someone around here must have proposed this… but what if, instead of the “nice town”, which already is willing to make small sacrifices, having to make huge sacrifices for the people in the bottomless pit to leave the pit; what would they say about making small sacrifices so as to make the bottomless pit a “very deep pit of despair”, and then a “deep pit”, then a “pit”, and then eventually just a “area that in general has a lower elevation”?

    I have lived all my life in what was called, when I was a child, a “Third World Country”. Nowadays its a “developing nation”. Here we have my “nice town”, which is quite nice indeed, and to truly find pits of despair we have to look further and further way. Some of these pits have already been demoted to the “lower elevation” status.

    The trade-off that the nice town VS bottomless pit dillema offers doesn’t factor time. This is OK for thought experiments, but time can be a valuable resource in real world decisions…. even though I seldom see it being factored in any kind of decision really…

    A final point: my country has always had both “nice towns” and “bottomless pits of despair”. The ratio has fortunately been incresing in favor of the “nice towns”. But anyway there are places in the world where maybe there’s nothing but bottomless pits of despair, and I don’t know if time can possibly favor places like these.

    • Ialdabaoth says:

      I can even see utilitarian reasons to play things this way, with a few pockets of niceness amid the pits of despair (to be elevated later). That way, we get to experiment with different ways of not being pits of despair before elevating the rest of the world and risking premature optimization failures.

      • Multiheaded says:

        IMO all the biggest problems are with the scalability anyway, while on the other hand every small pocket of niceness usually looks good enough while you’re both throwing resources at it and micromanaging it properly. (Ask the Eastern Bloc!)

  40. Eli says:

    Don’t be silly, Scott. The big problem with these utilitarian “dilemmas” is that they deny the basic fact of history: there is no nice, prosperous town that was not once a “bottomless” pit of suffering. The real world does not optimize itself to our liking: we pour in extensive amounts of effort to make a tiny circle of firelight in the dark. Mind, this is why the very concept of “everyone in the First World gives all their money to the Third World RIGHT NOW IN HUGE CHARITABLE DONATIONS TO GIVEDIRECTLY” is completely ill-founded: wealth transfer can solve problems of inequality (maldistribution of resources), but we need to both massively increase the productivity of poor areas and ensure that they retain access to the fruits of their own labors (rather than those fruits being shipped off to the West).

    And yes, slowly, by steady effort, we are widening that circle of firelight. As of today, the territory over which we need widen it is finite: there is only so much suffering in the world.

    But then again, I’m a post-hoc consequentialist (as you once called it), so I define my morality in relation to the real world rather than in relation to a Solomonoff-weighted ensemble of all possible worlds (as the utilitarians do).

    • MugaSofer says:

      I’m pretty sure that the existence of the Third World is a result of maldistribution of resources.

      However, even if we couldn’t “solve” the Third World simply by providing free food and medical care, or money – which seems unlikely, but maybe there’s something wrong with their societies that will trap them in poverty even if they aren’t poor to start with –

      – well, then the optimal way to expand the Circle of Firelight (that is, the First World, right?) is pretty clearly to invade.

      • Multiheaded says:

        I’d trust a competent!Soviet invasion with actually doing good to a much greater degree than, say, an EU one. The precedents are there.

        • Geirr says:

          Why? I can see why you might believe that, i.e. Afghanistan but a willingness to use internment camps would afaics erase the differences. I was going to add a willingness to expropriate/redistribute property to get local buy in but anyplace with a government functional enough for that to matter wouldn’t be in danger of humanitarian invasion. Poor rechtstaat grow very fast.

          • Multiheaded says:

            Central Asia in the 1920s-1930s; backing actually popular and non-corrupt leaders like Patrice Lumumba looks like he didn’t actually get much Soviet support, but it seems like Soviet foreign policy was less actively destructive…

          • Geirr says:

            @Multiheaded Central Asia is a wonderful example of successful Soviet cultural imperialism and making things less shitty on many levels. But the stans were integral parts of the USSR for 2, maybe 3 political generations and received large subsidies the entire time. If any EU government was actually able to credibly commit to “We’re never leaving” levels of occupation and internment for the inconvenient they could do as well. Look at Bosnia, the only reason the Republika Srpska hasn’t joined Serbia is the occupying force won’t allow it. Bosnia is an EU colony with many of the trappings of a state, not anything with any real independence. The assurance that they’re never, ever leaving is lacking and they still haven’t fucked it up.

            I must say that the absolute certainty that the occupying power would prefer you all dead to withdrawal does make pacifying the occupied easier but I don’t think the difference between a Soviet and an EU annexation force would be much if they were willing to use internment and it was clear this was annexation. Hell, anyplace worth doing a humanitarian invasion of just allowing free travel to the EU would probably depopulate it.

      • Eli says:

        – well, then the optimal way to expand the Circle of Firelight (that is, the First World, right?) is pretty clearly to invade.

        Except that we can’t trust our First World governments to develop the Third World instead of exploiting it. Otherwise I’d bite this bullet: Third World people really are being unjustly deprived of our finer lifestyles and greater civil rights by the endemic willful ignorance of their entire societies.

        • Geirr says:

          Why not? Invade, annex, give citizenship, allow free travel to the metropole, boom, anyplace worth a humanitarian invasion is depopulated by emigration. The value of any natural resources will be dwarfed by the extra costs of social welfare.

  41. Ghatanathoah says:

    I’m not sure our intuition about the Bottomless Pit of suffering are what makes Pascal’s Mugging so counterintuitive. You could easily rewrite Pascal’s mugging so it isn’t about strangers (i.e. I will make you and everyone in your nice town immortal and torture you forever if you don’t give me $5) and it still seems counterintuitive to me.

    • Susebron says:

      One could view potential future selves as strangers. You can’t see them from your current perspective, certainly.

  42. DavidS says:

    Really interesting (the speculation about both progressives and conservatives being ‘right’ in that way was totally new to me).

    I wonder though whether this is entirely an issue of the bottomless pit being external and it being possible to ignore it, or whether a lot of it is just the ‘bottomless’ bit.

    I find that in my own life (trying to get good at something, trying to achieve something at work, trying to lose weight, whatever), having a big task where immediate effort doesn’t achieve a significant proportion of the overall result can be demoralising – it has to be broken down so you can feel like the achievement means something.

    If this is a large chunk of it, it might make sense of how charities tend to zoom in on more resolveable things – whether ‘save this particular child who we’re sending you a picture of’ or ‘contribute to curing this particular disease/problem’.

    Many people would rather spend £100m defeating a disease that kills 10 people a year rather than cutting by 1% a disease that kills 10,000 a year. But many people would ALSO rather use £1k to completely pay off a £1k overdraft with 5% interest than pay £1k off a £20,000 credit card debt at 15% interest.

  43. Pingback: Our future older selves | N=1

  44. I found it! I was going through my activity log on Facebook and I finally found the Facebook conversation that started it all!

    Here it is!

  45. Pingback: Poor Countries in Space | Ordinary Times