11 ms·

The tyranny of the marginal user reminds me of population ethics' The Repugnant Conclusion.[0] This is the conclusion of utilitarianism, where if you have N peo

by bdcs 3y ago

The tyranny of the marginal user reminds me of population ethics' The Repugnant Conclusion.[0] This is the conclusion of utilitarianism, where if you have N people each with 10 happiness, well then, it would be better to have 10N people with 1.1 happiness, or 100N people with 0.111 happiness, until you have infinite people with barely any happiness. Substitute profit for happiness, and you get the tyranny of the marginal user. Perhaps the resolutions to the Repugnant Conclusion (Section 2, "Eight Ways of Dealing with the Repugnant Conclusion") can also be applied to the tyranny of the marginal user. Though to be honest, I find none of the resolutions wholly compelling. [0] https://plato.stanford.edu/ARCHIVES/WIN2009/entries/repugnant-conclusion/ https://plato.stanford.edu/ARCHIVES/WIN2009/entries/repugnan...

PaulDavisThe1st 3y ago
Many versions of utilitarianism never specified the function to compute the sum for the many. Your example assumes that the function is simple addition, but others have been proposed that reflect some of the complexities of the human condition a little more explicitly (e.g. sad neighbors make neighbors sad).
- onlyrealcuzzo 3y ago
  Yeah, utilitarianism means you want to act in a way that's beneficial to most people. There's many ways you can interpret that, though. But I think if you say, before we had 1 apple per person, and now we have 2x as many apples, but they're all owned by one person - that's hard to argue it's utilitarian. If before you had 100 apples, and everyone who wanted one had one, and now you have 10,000 apples distributed to people at random, but only 1 in 100 people who wants one has one - that also seems hard to argue as utilitarian. Businesses are value maximization functions. They'll only be utilitarian if that happens to maximize value. In the case of software - if you go from 1m users to 10m users - that doesn't imply utilitarianism. It implies that was good for gaming some metric - which more often than not these days is growth, not profit.
- tasty_freeze 3y ago
  Reinforcing your point, Peter Singer, philosopher and noted utilitarian, has explicitly said that he weights misery far more than happiness in his own framework. From a personal level, he said he'd give up the 10 best days of his life to remove the one worst day of his life (or something like that). All of his work with effective altruism is aimed at reducing suffering of those worst off in the world and spends no time with how to make the well off even happier.
  - frereubu 3y ago
    I hadn’t heard that about Singer’s philosophy (unsurprisingly as I’ve read very little of his work). It’s interesting for me in that it lines up with Kahnemann & Tversky’s “losses loom larger than gains” heuristic in psychology.
    - tasty_freeze 3y ago
      Singer publishes his book, "The Life You Can Save" for free -- it was revised for the 10th anniversary. It is available in various formats, including an audio book. It is a short, easy read. It is also one of the most impactful books I've ever read. https://www.thelifeyoucansave.org/ https://www.thelifeyoucansave.org/ Along the header there is an item "free book" if you want to get the book. In short, there are a few theses in the book that combine in a compelling way. This isn't his summary, this is the summary I got out of reading it. (1) the suffering of someone you don't know is just as real as the suffering of someone you do know. (2) there are many worthy causes, but we aren't allocating enough resources to address them all, so it is best to allocate them such that they do the most good for the worst off (3) if you are reading hacker news, you are likely in the 1% (worldwide) and your primary needs are already being met and you can make meaningful contributions to help those most in need without affecting your lifestyle much. Each chapter tackles a topic. For instance, part of it brings up many the counter arguments about why someone might not want to donate and then debunks them. Eg, in a poll, most US citizens believe that 10-20% of their taxes are going to foreign, and say about half that, say 5%, would be reasonable. In fact it is well under 1%. Or people will say there have always been poor people and there will always be poor people (counter argument: to the people who are helped, aid makes a huge difference to them personally, and there have actually been great strides in the past 30 years at reducing global poverty). Another chapter talks about the typical feel good news item about someone in the community who has gone blind and so the community pitches in to pay for a seeing eye dog for the person (it costs up to $50K to train and vet such dogs). Yet that same $50K could have prevented river blindness for thousands of children, or been enough to perform cataract surgery on thousands of blind adults. People often say, why give to a charity in Africa? It will just end up in someone's pocket before it helps anyway. The book talks about the effective altruism movement and describes how charities are vetted and monitored. The website above has links to many such vetted charities. Another chapter talks about where to draw the line? Sure, I can afford $700 to pay for corrective surgery for a woman suffering from fistula and feel good about myself. But in reality I could afford another $700, then another $700, etc. Do I need to keep donating to those worse off until I am one of the worse off people? (spoiler: no)
- jancsika 3y ago
  > (e.g. sad neighbors make neighbors sad) I much prefer, "I'd rather have a bottle in front of me than a frontal lobotomy." At least in that case nobody will confuse a trucker hat slogan for a viable system of ethics.
- fouronnes3 3y ago
  Assuming linearity of utility either in individuals or in aggregation is a very common straw man of utilitarianism.
  - polygamous_bat 3y ago
    Doesn't have to be linear, ANY strictly increasing function for aggregating the utility leads to the same conclusion.
- tshaddox 3y ago
  Which conceivable method of summing is the least problematic? Depending on the summing method you might find yourself advocating creating as many people as possible with positive utility, or eliminate everyone with below-average utility, etc.
  - Karrot_Kream 3y ago
    Utility is very complicated and summing might not even be possible. Folks have argued for completely different utility systems, such as cardinal utility where utility is modeled purely as relations instead of something that is isomorphic to a real. Even going by the mainstream view of ordinal utility, utility tends to be a convex function (simplistically, having 1 food is much better than having no food, but having 1000 food isn't that much better than having 500 food.) Modeling utility as something purely isomorphic to reals gives it all the fun paradoxes that we know the reals have and can be used to create some really wacky results. The "repugnant conclusion" is a direct consequence of that.
wilg 3y ago
I've never understood this problem. To me, it seems that since you've defined a minimum "worth living" amount of happiness and unbounded population, it makes complete sense that the answer would be that it is better to have lots of people whose life is worth living rather than fewer. Is it not tautological? Like it seems like you have to take "worth living" seriously, since that is the element that is doing all the work. If it's worth living, you've factored in everything that matters already.
- mhb 3y ago
  If you pack the whole problem into a definition of "worth living", then you're right. But the premise is that there is a range from extreme misery through neutral through extremely happy. The repugnant conclusion is that it is better to have many people in a state that is barely above neutral.
  - wilg 3y ago
    I'm not the one packing it, the setup of the problem does it. "Barely above neutral" means you've picked an acceptable state. And then we are supposed to consider that acceptable state "repugnant"?
    - mhb 3y ago
      There's a comparison. If the scale goes from -100 to +100, the conclusion is that if we have 8 billion people in the world with average happiness of +10, it is better to immiserate them in order to have 80 billion with average happiness +1.01. It's not that the acceptable state of 1.01 is repugnant, it's that the conclusion seems counterintuitive and ethically problematic to many people, as it suggests that we should prefer creating a massive population of people who are barely happy over a smaller population of people who are very happy.
      - wilg 3y ago
        I guess I just don't understand how if your axioms are 1) X is an acceptable level of happiness and 2) more people are better than fewer it is in any way surprising or problematic to end up with infinite people at happiness X. Perhaps people don't see that (2) is a part of the premise?
hammock 3y ago
Tyranny of the marginal user is a riff on the Nassim Taleb classic "The Most Intolerant Wins: The Dictatorship of the Small Minority": https://medium.com/incerto/the-most-intolerant-wins-the-dictatorship-of-the-small-minority-3f1f83ce4e15 https://medium.com/incerto/the-most-intolerant-wins-the-dict...
mvdtnz 3y ago
Why would anyone think that a large overall pool of happiness is somehow better than a high per capita happiness? This seems like the kind of thing that's incredibly obvious to everyone but the academic philosopher.
- saint_fiasco 3y ago
  In this particular case, it's because the success of an ad-funded service depends on the amount of users it has. If you don't like the repugnant conclusion you have to change something in the conditions of the environment so that you make it not be true. Arguing against it and calling your refutation obvious doesn't do anything.
  - mvdtnz 3y ago
    That is an incredibly long bow to draw. Corporations are optimising for their own profits, not anyone's happiness.
    - saint_fiasco 3y ago
      I agree. The math that applies to corporate profits is not the same that should apply for human happiness. But we have to acknowledge that the weird philosophical thought experiment that can't possibly convince anyone except weird philosophers turned out to be convincing to other entities after all. Compare the trolley problem, a famous thought experiment that people used to laugh at, up until a couple of years when suddenly important people began to ask important questions like "should we relax the safety standards for potentially life-saving vaccines" and "how much larger than Y does X need to be so that preventing X functionally illiterate children are worth the price of Y dead children"
- wilg 3y ago
  It's just a question of if you value other people existing or not. If you don't, focus on per-capita happiness, if you do then you focus on meeting a minimum threshold of happiness for everyone. I don't see how you couldn't value other people existing – I think they have just as much of a right to experience the universe as I do.
tyre 3y ago
The Repugnant Conclusion is one of those silly problems in philosophy that don’t make much sense outside of academics. Utilitarianism ought to be about maximizing the happiness (total and distribution) of an existing population. Merging it with natalism isn’t realistic or meaningful, so we end up with these population morality debates. The happiness of a unconceived possible human is null (not the same as zero!) Compare to Rawls’s Original Position, which uses an unborn person to make the hypothetical work but is ultimately about optimizing for happiness in an existing population. We really shouldn’t get ourselves tied into knots about the possibility of pumping out trillions of humans because an algorithm says they’ll be marginally net content. That’s not the end goal of any reasonable, practical, or sane system of ethics.
- caturopath 3y ago
  I think you might be missing a big part of what this sort of philosophy is really about. > Utilitarianism ought to be about maximizing the happiness (total and distribution) of an existing population For those who accept your claim above, lots of stuff follows, but your claim is a bold assertion that isn't accepted by everyone involved, or even many people involved. The repugnant conclusion is a thought experiment where one starts with certain stripped-down claims not including yours here and follow it to its logical conclusion. This is worth doing because many people find it plausible that those axioms define a good ethical system, but the fact they require the repugnant conclusion causes people to say "Something in here seems to be wrong or incomplete." People have proposed many alternate axioms, and your take is just one which isn't popular. I suspect part of the reason yours isn't popular is - People seek axiological answers from their ethical systems, so they wish to be able to answer "Are these two unlike worlds better?" -- even if they aren't asking "What action should I take?" Many people want to know "What is better?" so they explore questions of what are better, period, and something they want is to always to have such questions be answerable. Some folks have explored a concept along the lines of yours, where sometimes there just isn't a comparison available, but giving up on being able to compare every pair isn't popular. - We actually make decisions or imagine the ability to make future real decisions that result in there being more or fewer persons. Is it right to have kids? Is it right to subsidize childbearing? Is it right to attempt to make a ton of virtual persons? > The happiness of a unconceived possible human is null (not the same as zero!) Okay, if you say "Total utilitarianism (and all similar things) are wrong", then of course you don't reach the repugnant conclusion via Parfit's argument. "A, B, C implies D", "Well, not B" is not a very interesting argument here. Your null posing also doesn't really answer how we _should_ handle questions of what to do that result in persons being created or destroyed. > We really shouldn’t get ourselves tied into knots about the possibility of pumping out trillions of humans because an algorithm says they’ll be marginally net content. That’s not the end goal of any reasonable, practical, or sane system of ethics. Okay, what is the end goal? If you'll enlighten us, then we can all know. Until then, folks are going to keep trying to figure it out. Parfit explored a system that many people might have thought sounded good on its premises, but proved it led to the repugnant conclusion. The normal reaction is, "Okay, that wasn't the right recipe. Let's keep looking. I want to find a better recipe so I know what to do in hard, real cases." Since such folks rejected the ethical system because it led to the repugnant conclusion, they could be less confident in its prescriptions in more practical situations -- they know that the premises of the system don't reflect what they want to adopt as their ethical system.
feoren 3y ago
That conclusion is not repugnant at all, it's just that its phrasing is so simplistic as to be nearly a straw-man. It's a poisoned intuition pump, because it makes you imagine a situation that doesn't follow at all from utilitarianism. First of all, you're imagining dividing happiness among more people, but imagining them all with the same amount of suffering. You're picturing a drudging life where people work all day and have barely any source of happiness. But if you can magically divide up some total amount of happiness, why not the same with suffering? This is the entire source of the word "repugnant", because it sounds like you get infinite suffering with finite happiness. That does not follow from anything utilitarianism stipulates; you've simply created an awful world and falsely called it utilitarianism. Try to imagine all these people living a nearly completely neutral life, erring a bit on the happier side, and it suddenly doesn't sound so bad. Secondly, you're ignoring the fact that people can create happiness for others. What fixed finite "happiness" resource are we divvying up here? Surely a world with 10 billion people has more great works of art for all to enjoy than a world with 10 people, not to mention far less loneliness. It's crazy to think the total amount of happiness to distribute is independent of the world population. There are many more reasonable objections to even the existence of that so-called "conclusion" without even starting on the many ways of dealing with it.
- julianeon 3y ago
  Here's a simpler way to phrase the problem. The current world population is about 8 billion. By this argument, and also by your argument, it should actually be 999 billion. Or a number even higher than that. The conclusion boils down to: 1. Find maximum population number earth can support. 2. Hit that number. I do think that, when put this way, it seems simplistic.
  - curiousllama 3y ago
    To be fair, boiling something down to a simple statement does indeed tend to produce simplistic statements
  - olddustytrail 3y ago
    Here's an even simpler way to phrase the problem. The current world population is about 8 billion. By my argument it should be 2 billion. Your argument is therefore rather foolish.
coldtea 3y ago
AKA the Repugnantly Ignorant in the Human-Ways Nerd's Idea of Ethics conclusion!
scythe 3y ago
The problem is that the "repugnant conclusion" is a matter of definitions. A moral theory is (basically) freely chosen: you can change the definitions whenever you like. Not so for B2C SaaS. The utilities are always measured in dollars and they always aggregate by simple addition. You can't simply redefine the problem away by changing the economic assumptions, because they exist in physical space and not in theory space.
crabbone 3y ago
One way to deal with this problem is to ask why do we use the arithmetic sum to calculate the total happiness?. There are plenty of ways this can go. Say, if you believe that two very happy people are better than four half as happy people, then you can define this sum function as sum(happiness_per_person) / number_of_people. Of course, this isn't the only way. Utilitarianism opens a lot of questions about comparability of utility (or happiness) of different people as well as summation. Is it a totally ordered set? Is it a partially ordered set? Perhaps utility is incomparable (that'd be sad and kind of defeat the whole doctrine, but still). Also, can unhappiness be compensated by happiness? We unthinkingly rush to treat unhappiness as we would negative numbers and try to sum that with happiness, but what if it doesn't work? What if the person who has no happiness or unhappiness isn't in the same place as the person who is equally happy and unhappy (their dog died, but they found a million $ on the same day)? A more typical classroom question would be about chopping up a healthy person for organs to fix X unhealthy people -- is there a number of unhealthy people which would justify killing a healthy person for spare parts?
Ensorceled 3y ago
Every time I've engaged in debate over this, it always comes down to believing that the world is zero sum and there is a limited amount of "happiness" that can be distributed. That may be true for some things, but for many decisions it is not true. There is enough food to feed everyone if we choose to distribute it properly. There is enough housing to house everyone. etc. etc. There may not be enough cardiologists or Dali originals ...
didibus 3y ago
I really don't see the issue with your happiness split. You have 10 people, and they're are equally unhappy. This is perfect, because now they are all equally incentivized to do something about it. They're motivated to work together and collaborate for change. If you do any other split where some people will be very happy and others very unhappy, you've now created certain category of people who are incentivized to maintain the current system and repress any desire for change from the unhappy people.
Vt71fcAqt7 3y ago
There is a minimum happiness threshold mH. We can increase population P until happiness H reaches mH, give or take some depending on how close you want to get to mH.
Terr_ 3y ago
> infinite people with barely any happiness That reminds me of the SMBC "Existifier" comic, which satirizes the idea that merely helping something exist is morally positive. https://www.smbc-comics.com/comic/existence https://www.smbc-comics.com/comic/existence
mercenario 3y ago
> This is the conclusion of utilitarianism, where if you have N people each with 10 happiness, well then, it would be better to have 10N people with 1.1 happiness, or 100N people with 0.111 happiness, until you have infinite people with barely any happiness 1) Population isn't infinite, you can't continue this for too long 2) Your assumption completely depends on how costly it is to increase +1 happiness and to increase +1 user, you don't even mention it. And these costs are not fixed, it increases, so even if it is cheaper to add +1 user in the beginning, it will not continue to be cheaper indefinitely So, nothing is preventing you from increasing happiness at the same time you increase users.

by bdcs 3y ago

PaulDavisThe1st 3y ago

Many versions of utilitarianism never specified the function to compute the sum for the many. Your example assumes that the function is simple addition, but others have been proposed that reflect some of the complexities of the human condition a little more explicitly (e.g. sad neighbors make neighbors sad).

onlyrealcuzzo 3y ago
Yeah, utilitarianism means you want to act in a way that's beneficial to most people. There's many ways you can interpret that, though. But I think if you say, before we had 1 apple per person, and now we have 2x as many apples, but they're all owned by one person - that's hard to argue it's utilitarian. If before you had 100 apples, and everyone who wanted one had one, and now you have 10,000 apples distributed to people at random, but only 1 in 100 people who wants one has one - that also seems hard to argue as utilitarian. Businesses are value maximization functions. They'll only be utilitarian if that happens to maximize value. In the case of software - if you go from 1m users to 10m users - that doesn't imply utilitarianism. It implies that was good for gaming some metric - which more often than not these days is growth, not profit.
tasty_freeze 3y ago
Reinforcing your point, Peter Singer, philosopher and noted utilitarian, has explicitly said that he weights misery far more than happiness in his own framework. From a personal level, he said he'd give up the 10 best days of his life to remove the one worst day of his life (or something like that). All of his work with effective altruism is aimed at reducing suffering of those worst off in the world and spends no time with how to make the well off even happier.
- frereubu 3y ago
  I hadn’t heard that about Singer’s philosophy (unsurprisingly as I’ve read very little of his work). It’s interesting for me in that it lines up with Kahnemann & Tversky’s “losses loom larger than gains” heuristic in psychology.
  - tasty_freeze 3y ago
    Singer publishes his book, "The Life You Can Save" for free -- it was revised for the 10th anniversary. It is available in various formats, including an audio book. It is a short, easy read. It is also one of the most impactful books I've ever read. https://www.thelifeyoucansave.org/ https://www.thelifeyoucansave.org/ Along the header there is an item "free book" if you want to get the book. In short, there are a few theses in the book that combine in a compelling way. This isn't his summary, this is the summary I got out of reading it. (1) the suffering of someone you don't know is just as real as the suffering of someone you do know. (2) there are many worthy causes, but we aren't allocating enough resources to address them all, so it is best to allocate them such that they do the most good for the worst off (3) if you are reading hacker news, you are likely in the 1% (worldwide) and your primary needs are already being met and you can make meaningful contributions to help those most in need without affecting your lifestyle much. Each chapter tackles a topic. For instance, part of it brings up many the counter arguments about why someone might not want to donate and then debunks them. Eg, in a poll, most US citizens believe that 10-20% of their taxes are going to foreign, and say about half that, say 5%, would be reasonable. In fact it is well under 1%. Or people will say there have always been poor people and there will always be poor people (counter argument: to the people who are helped, aid makes a huge difference to them personally, and there have actually been great strides in the past 30 years at reducing global poverty). Another chapter talks about the typical feel good news item about someone in the community who has gone blind and so the community pitches in to pay for a seeing eye dog for the person (it costs up to $50K to train and vet such dogs). Yet that same $50K could have prevented river blindness for thousands of children, or been enough to perform cataract surgery on thousands of blind adults. People often say, why give to a charity in Africa? It will just end up in someone's pocket before it helps anyway. The book talks about the effective altruism movement and describes how charities are vetted and monitored. The website above has links to many such vetted charities. Another chapter talks about where to draw the line? Sure, I can afford $700 to pay for corrective surgery for a woman suffering from fistula and feel good about myself. But in reality I could afford another $700, then another $700, etc. Do I need to keep donating to those worse off until I am one of the worse off people? (spoiler: no)
jancsika 3y ago
> (e.g. sad neighbors make neighbors sad) I much prefer, "I'd rather have a bottle in front of me than a frontal lobotomy." At least in that case nobody will confuse a trucker hat slogan for a viable system of ethics.
fouronnes3 3y ago
Assuming linearity of utility either in individuals or in aggregation is a very common straw man of utilitarianism.
- polygamous_bat 3y ago
  Doesn't have to be linear, ANY strictly increasing function for aggregating the utility leads to the same conclusion.
tshaddox 3y ago
Which conceivable method of summing is the least problematic? Depending on the summing method you might find yourself advocating creating as many people as possible with positive utility, or eliminate everyone with below-average utility, etc.
- Karrot_Kream 3y ago
  Utility is very complicated and summing might not even be possible. Folks have argued for completely different utility systems, such as cardinal utility where utility is modeled purely as relations instead of something that is isomorphic to a real. Even going by the mainstream view of ordinal utility, utility tends to be a convex function (simplistically, having 1 food is much better than having no food, but having 1000 food isn't that much better than having 500 food.) Modeling utility as something purely isomorphic to reals gives it all the fun paradoxes that we know the reals have and can be used to create some really wacky results. The "repugnant conclusion" is a direct consequence of that.

wilg 3y ago

I've never understood this problem. To me, it seems that since you've defined a minimum "worth living" amount of happiness and unbounded population, it makes complete sense that the answer would be that it is better to have lots of people whose life is worth living rather than fewer. Is it not tautological? Like it seems like you have to take "worth living" seriously, since that is the element that is doing all the work. If it's worth living, you've factored in everything that matters already.

mhb 3y ago
If you pack the whole problem into a definition of "worth living", then you're right. But the premise is that there is a range from extreme misery through neutral through extremely happy. The repugnant conclusion is that it is better to have many people in a state that is barely above neutral.
- wilg 3y ago
  I'm not the one packing it, the setup of the problem does it. "Barely above neutral" means you've picked an acceptable state. And then we are supposed to consider that acceptable state "repugnant"?
  - mhb 3y ago
    There's a comparison. If the scale goes from -100 to +100, the conclusion is that if we have 8 billion people in the world with average happiness of +10, it is better to immiserate them in order to have 80 billion with average happiness +1.01. It's not that the acceptable state of 1.01 is repugnant, it's that the conclusion seems counterintuitive and ethically problematic to many people, as it suggests that we should prefer creating a massive population of people who are barely happy over a smaller population of people who are very happy.
    - wilg 3y ago
      I guess I just don't understand how if your axioms are 1) X is an acceptable level of happiness and 2) more people are better than fewer it is in any way surprising or problematic to end up with infinite people at happiness X. Perhaps people don't see that (2) is a part of the premise?

hammock 3y ago

Tyranny of the marginal user is a riff on the Nassim Taleb classic "The Most Intolerant Wins: The Dictatorship of the Small Minority": https://medium.com/incerto/the-most-intolerant-wins-the-dictatorship-of-the-small-minority-3f1f83ce4e15 https://medium.com/incerto/the-most-intolerant-wins-the-dict...

mvdtnz 3y ago

Why would anyone think that a large overall pool of happiness is somehow better than a high per capita happiness? This seems like the kind of thing that's incredibly obvious to everyone but the academic philosopher.

saint_fiasco 3y ago
In this particular case, it's because the success of an ad-funded service depends on the amount of users it has. If you don't like the repugnant conclusion you have to change something in the conditions of the environment so that you make it not be true. Arguing against it and calling your refutation obvious doesn't do anything.
- mvdtnz 3y ago
  That is an incredibly long bow to draw. Corporations are optimising for their own profits, not anyone's happiness.
  - saint_fiasco 3y ago
    I agree. The math that applies to corporate profits is not the same that should apply for human happiness. But we have to acknowledge that the weird philosophical thought experiment that can't possibly convince anyone except weird philosophers turned out to be convincing to other entities after all. Compare the trolley problem, a famous thought experiment that people used to laugh at, up until a couple of years when suddenly important people began to ask important questions like "should we relax the safety standards for potentially life-saving vaccines" and "how much larger than Y does X need to be so that preventing X functionally illiterate children are worth the price of Y dead children"
wilg 3y ago
It's just a question of if you value other people existing or not. If you don't, focus on per-capita happiness, if you do then you focus on meeting a minimum threshold of happiness for everyone. I don't see how you couldn't value other people existing – I think they have just as much of a right to experience the universe as I do.

tyre 3y ago

The Repugnant Conclusion is one of those silly problems in philosophy that don’t make much sense outside of academics. Utilitarianism ought to be about maximizing the happiness (total and distribution) of an existing population. Merging it with natalism isn’t realistic or meaningful, so we end up with these population morality debates. The happiness of a unconceived possible human is null (not the same as zero!) Compare to Rawls’s Original Position, which uses an unborn person to make the hypothetical work but is ultimately about optimizing for happiness in an existing population. We really shouldn’t get ourselves tied into knots about the possibility of pumping out trillions of humans because an algorithm says they’ll be marginally net content. That’s not the end goal of any reasonable, practical, or sane system of ethics.

caturopath 3y ago
I think you might be missing a big part of what this sort of philosophy is really about. > Utilitarianism ought to be about maximizing the happiness (total and distribution) of an existing population For those who accept your claim above, lots of stuff follows, but your claim is a bold assertion that isn't accepted by everyone involved, or even many people involved. The repugnant conclusion is a thought experiment where one starts with certain stripped-down claims not including yours here and follow it to its logical conclusion. This is worth doing because many people find it plausible that those axioms define a good ethical system, but the fact they require the repugnant conclusion causes people to say "Something in here seems to be wrong or incomplete." People have proposed many alternate axioms, and your take is just one which isn't popular. I suspect part of the reason yours isn't popular is - People seek axiological answers from their ethical systems, so they wish to be able to answer "Are these two unlike worlds better?" -- even if they aren't asking "What action should I take?" Many people want to know "What is better?" so they explore questions of what are better, period, and something they want is to always to have such questions be answerable. Some folks have explored a concept along the lines of yours, where sometimes there just isn't a comparison available, but giving up on being able to compare every pair isn't popular. - We actually make decisions or imagine the ability to make future real decisions that result in there being more or fewer persons. Is it right to have kids? Is it right to subsidize childbearing? Is it right to attempt to make a ton of virtual persons? > The happiness of a unconceived possible human is null (not the same as zero!) Okay, if you say "Total utilitarianism (and all similar things) are wrong", then of course you don't reach the repugnant conclusion via Parfit's argument. "A, B, C implies D", "Well, not B" is not a very interesting argument here. Your null posing also doesn't really answer how we _should_ handle questions of what to do that result in persons being created or destroyed. > We really shouldn’t get ourselves tied into knots about the possibility of pumping out trillions of humans because an algorithm says they’ll be marginally net content. That’s not the end goal of any reasonable, practical, or sane system of ethics. Okay, what is the end goal? If you'll enlighten us, then we can all know. Until then, folks are going to keep trying to figure it out. Parfit explored a system that many people might have thought sounded good on its premises, but proved it led to the repugnant conclusion. The normal reaction is, "Okay, that wasn't the right recipe. Let's keep looking. I want to find a better recipe so I know what to do in hard, real cases." Since such folks rejected the ethical system because it led to the repugnant conclusion, they could be less confident in its prescriptions in more practical situations -- they know that the premises of the system don't reflect what they want to adopt as their ethical system.

feoren 3y ago

That conclusion is not repugnant at all, it's just that its phrasing is so simplistic as to be nearly a straw-man. It's a poisoned intuition pump, because it makes you imagine a situation that doesn't follow at all from utilitarianism. First of all, you're imagining dividing happiness among more people, but imagining them all with the same amount of suffering. You're picturing a drudging life where people work all day and have barely any source of happiness. But if you can magically divide up some total amount of happiness, why not the same with suffering? This is the entire source of the word "repugnant", because it sounds like you get infinite suffering with finite happiness. That does not follow from anything utilitarianism stipulates; you've simply created an awful world and falsely called it utilitarianism. Try to imagine all these people living a nearly completely neutral life, erring a bit on the happier side, and it suddenly doesn't sound so bad. Secondly, you're ignoring the fact that people can create happiness for others. What fixed finite "happiness" resource are we divvying up here? Surely a world with 10 billion people has more great works of art for all to enjoy than a world with 10 people, not to mention far less loneliness. It's crazy to think the total amount of happiness to distribute is independent of the world population. There are many more reasonable objections to even the existence of that so-called "conclusion" without even starting on the many ways of dealing with it.

julianeon 3y ago
Here's a simpler way to phrase the problem. The current world population is about 8 billion. By this argument, and also by your argument, it should actually be 999 billion. Or a number even higher than that. The conclusion boils down to: 1. Find maximum population number earth can support. 2. Hit that number. I do think that, when put this way, it seems simplistic.
- curiousllama 3y ago
  To be fair, boiling something down to a simple statement does indeed tend to produce simplistic statements
- olddustytrail 3y ago
  Here's an even simpler way to phrase the problem. The current world population is about 8 billion. By my argument it should be 2 billion. Your argument is therefore rather foolish.

coldtea 3y ago

AKA the Repugnantly Ignorant in the Human-Ways Nerd's Idea of Ethics conclusion!

scythe 3y ago

The problem is that the "repugnant conclusion" is a matter of definitions. A moral theory is (basically) freely chosen: you can change the definitions whenever you like. Not so for B2C SaaS. The utilities are always measured in dollars and they always aggregate by simple addition. You can't simply redefine the problem away by changing the economic assumptions, because they exist in physical space and not in theory space.

crabbone 3y ago

One way to deal with this problem is to ask why do we use the arithmetic sum to calculate the total happiness?. There are plenty of ways this can go. Say, if you believe that two very happy people are better than four half as happy people, then you can define this sum function as sum(happiness_per_person) / number_of_people. Of course, this isn't the only way. Utilitarianism opens a lot of questions about comparability of utility (or happiness) of different people as well as summation. Is it a totally ordered set? Is it a partially ordered set? Perhaps utility is incomparable (that'd be sad and kind of defeat the whole doctrine, but still). Also, can unhappiness be compensated by happiness? We unthinkingly rush to treat unhappiness as we would negative numbers and try to sum that with happiness, but what if it doesn't work? What if the person who has no happiness or unhappiness isn't in the same place as the person who is equally happy and unhappy (their dog died, but they found a million $ on the same day)? A more typical classroom question would be about chopping up a healthy person for organs to fix X unhealthy people -- is there a number of unhealthy people which would justify killing a healthy person for spare parts?

Ensorceled 3y ago

Every time I've engaged in debate over this, it always comes down to believing that the world is zero sum and there is a limited amount of "happiness" that can be distributed. That may be true for some things, but for many decisions it is not true. There is enough food to feed everyone if we choose to distribute it properly. There is enough housing to house everyone. etc. etc. There may not be enough cardiologists or Dali originals ...

didibus 3y ago

I really don't see the issue with your happiness split. You have 10 people, and they're are equally unhappy. This is perfect, because now they are all equally incentivized to do something about it. They're motivated to work together and collaborate for change. If you do any other split where some people will be very happy and others very unhappy, you've now created certain category of people who are incentivized to maintain the current system and repress any desire for change from the unhappy people.

Vt71fcAqt7 3y ago

There is a minimum happiness threshold mH. We can increase population P until happiness H reaches mH, give or take some depending on how close you want to get to mH.

Terr_ 3y ago

> infinite people with barely any happiness That reminds me of the SMBC "Existifier" comic, which satirizes the idea that merely helping something exist is morally positive. https://www.smbc-comics.com/comic/existence https://www.smbc-comics.com/comic/existence

mercenario 3y ago

> This is the conclusion of utilitarianism, where if you have N people each with 10 happiness, well then, it would be better to have 10N people with 1.1 happiness, or 100N people with 0.111 happiness, until you have infinite people with barely any happiness 1) Population isn't infinite, you can't continue this for too long 2) Your assumption completely depends on how costly it is to increase +1 happiness and to increase +1 user, you don't even mention it. And these costs are not fixed, it increases, so even if it is cheaper to add +1 user in the beginning, it will not continue to be cheaper indefinitely So, nothing is preventing you from increasing happiness at the same time you increase users.