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 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 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. If you try to solve it, you’ll release prisoners, and at least a few of them are going to commit new crimes which you will have to see and think about. Or you can just whistle, pretend not to notice, and continue to enjoy your nice low-crime society.
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.