Trolley problems, would-you-rathers, drowning children, the superpower to create superpowers, and the fact that everyone's a goddamn politician when it comes to morality
I don't dislike hypotheticals, but I can easily see from your two examples why people would dislike them. Singer makes the inference that you value children's lives more than $3000, even though you only said you value it more than a suit. However, there is no suit that I value $3000, and in real life, I won't own such a suit - I'll use $3000 to pay for rent - and I value not being homeless way more than $3000. These two scenarios are not equivalent. I can understand that people would be suspicious of hypotheticals which are then switched to a non-equivalent scenario which makes them look bad (same goes for chicken hypothetical).
I think hypotheticals are great, but trying to directly apply the outcome to messy real life is like trying to design wiring assuming zero wire resistance, like in high school. It's generally a bad idea. Details matter, and poorly designed thought experiment is about as valuable as poorly designed lab experiment.
Maybe I also should write something about why it's OK (to an extent) to be a bundle of contradictions.
I think that identifying in the moral weight between thought experiments and real life like you just did, acknowledging how $3000 might matter more to you than a stranger you can’t see or hear, is most of the way to “engaging with thought experiment.” Now you’re engaged in a different tradeoff between rent and another person, and you can grapple with your moral reasons for choosing one or the other.
Refusing on principle to use this structured logic, avoiding trying to examine the moral differences between hypotheticals and real life, is how you would end up going against whatever moral interests you have.
Thought experiments pin down one vector of morality so you can examine it against the others you hold dear. They’re how we prioritize.
Just giving a shout out to contradiction here. The hypotheticals are annoying because they presume there's much value in interrogating the consistency of our ethics in the first place. There's not. Right and wrong are too complex for a consistent ethical system to handle. Good and evil are too intertwined to isolate in a lab.
What's important is learning to see all the tradeoffs involved and interrogate your immediate values in actual complex situations, which is pretty much the opposite of isolating a choice in a contrived hypothetical circumstance. Those people dodging your trolley problem, looking for a way to have their cake and eat it, too--theyre doing that because that's the proper and healthy response to a difficult tradeoff!
Yes, eventually you might have to flip that trolly switch or watch even more people die, and whether or not you do in practice might boil down to whether or not you ate breakfast that day. But you should be spending every second you can spare looking for a way around the tough decision, or at least surveying and introspecting for particulars that might sway you in the actual circumstance. And if you are stuck making the trolly problem choice, guess what, both answers were evil! No virtue points for anyone, life just sucks sometimes!
People respond to your hypothetical by trying to dodge it because that is the ethical thing to do. God save us from the bros who are like, “Oh I already know what I'm gonna do we covered this class of problem in ethics class!” They tend to have more self-righteousness than actual knowledge of good and evil.
“However … in real life …” is not a valid response to any hypothetical. The whole point is to isolate decisions in a way that rarely occurs in real life. The premise is that you are wearing a $3000 suit, just as that there is a drowning child.
“In real life I would never walk along a riverbank” is just misunderstanding what you’re being asked, or being a jackass.
And I didn't respond that way! Singer brought up real life when he moved from the hypothetical to claiming that answering "yes" means you should donate $3000 right now. Since it's not longer a hypothetical, real life constraints now apply.
Right. Well, you did mention a “real life” objection to the original premise. Singer is perhaps misusing the structure, or at least skipping some steps in his advocacy. To me, the key to that hypo is not the exact price of the suit (“expensive” would suffice), but that no one else is in a position to do anything. Once you’re talking about IRL donations, there are lots of other people’s decisions involved, questions about priorities and effectiveness, and no particular urgency for the now-abstracted beneficiaries (because there will be someone in equal need tomorrow).
These are also good differences. For me, ultimately, what matters is that it's just a completely and obviously different situation. And I don't find it surprising that people would get angry about someone posing an easy hypothetical question, then smuggling much more complex situation while claiming they are the same. People can feel the slight of hand - and if you do it often enough, which unfortunately has already happened, they would just become hostile to all hypotheticals.
Most of the times when it may look like I don’t know how to answer a moral thought experiment, it’s because it’s yet another tedious deontological argument against consequentialism that relies upon the same sleazy error as good old Transplant, namely drawing on an instinctive revulsion that’s caused by the inability of our brains to ignore real-world negative side effects explicitly ruled out of the thought experiment. Very often I just feel like saying, “look, dude, I think a better world is better” and letting them dot the “i”s and cross the “t”s, instead of going, “Yes, if replacing every pediatrician with immortal vampire Jimmy Saville somehow made the world a better place overall, it would be a good thing to do.”
Oh yeah, in a bad faith context *acting* like a politician isn’t a bad idea. When I say “it’s a bad idea to let your enemies pin you down with evil questions for quotes” I mean it! You don’t want to be known as “the guy who chose to do that horrible thing in that hypothetical” because it was weighed against a staggering amount of abstract suffering that sounds less bad in vibes-world. But you DO need to grapple with the question in your head.
But I desperately hope people in their head know what the right answer is, and are trying to grapple their morality with what their answer to the hypothetical is. I worry that people really are just leaving their morality unexamined, and this post is definitely against doing that.
“Superpower” is super hard to define, so I wouldn’t fault people for not being able to answer that question. Basically, how god-like can you make yourself and it still be fairly called “a” superpower? Also that’s actually a kind of moral dilemma in itself. Your superpower of choice is absolutely sick and I think definitely counts as a superpower, but could some form of psychic healing at a distance count? Because if something like that would count, it would be really hard not to choose that (I’d feel morally obligated to) even though it would probably make life very tedious as I would feel a strong duty to use it as much as possible and to figure out which people it would be best to heal. But maybe that’s already not interpreting the question in good faith. Maybe that question is supposed to be answered selfishly: “Which superpower would you WANT the most, not which one would you choose?” Anyway, great post—thank you!
Now this is a fun point! And an interesting example of taking the thought experiment to its natural conclusion vs adversarially changing the terms.
Yeah, if I could actually, really choose a superpower, I would want something that allowed me to figure out the actual moral nature of good and evil, and the ability to make the future good. Something like omniscience probably would work there. BUT that’s kinda not in line with the spirit of the question, as you say! Follow up questions could be asked: “okay, so this is what I would WANT, but the most FUN superpower would be…” or “this is the superpower I instinctively crave the most”
Acknowledging the pros and cons and differences between thought experiments are good! Refusing to answer them or failing to examine any of the reasons behind what you pick what you pick is bad.
Enjoyed this piece! I do think "I find that hypothetical unpleasant to think about" is a potentially valid and honest answer, if you're willing to acknowledge that you feel you have a right to not think about deeply unpleasant things that you're not forced to by the universe. And before you say that's a bad policy (which it might be), you either need to acknowledge that everyone does this all the time---in fact, we almost never don't do it---or persuade me otherwise, which I think will be hard to do
It is precisely the hypothetical nature of the thought experiments that makes them unanswerable. I wouldn't throw a fat man on to the tracks because I wouldn't think it would stop a train and I couldn't lift him and if push came to shove he would probably throw me. Pretending that the many such issues dont exist make the question meaningless. And then if I do give an answer to keep you happy some eejit will draw random wild conclusions about my beliefs.
Giving $3000 to save a child is a completely different thing than giving $3000 to a group of strangers who supposedly do what they say with it, outside your line of sight, that hypothetically saves a child. Also, why are you trying to get $3000 from me instead of from someone with 3000x more than me, of whom there are many?
Questions like "would you rather have snakes for arms or snakes for legs?" can be answered to full satisfaction with "that's a stupid question and I don't want to answer".
Many of the hypotheticals I see on TPOS are of the "too ridiculous to answer" variety. Moral choices made in the heat of the moment, like pulling a trolley lever or jumping into a pond, are far more dependent on constraints like courage and strength and quickness of thought than they are on passive moral law reasoning. I think the reluctance to engage with hypotheticals like the ones you name comes not just from moral weakness, but from a rejection of the framework that morality is just a math problem with feelings as the variables.
What's more, something like "pressing a button to inflict pain on a kid or animal" is the kind of moral choice that approximately nobody makes at any point in their lives, and I tend to think that moral reasoning from such a starting point will bear few insights. Many hypothetical questions can be readily rejected for simply being too silly to be worth the discussion.
That is a power though. It also has moral implications or whatever you were trying to get out of the question.
“But half the comments are somehow people who object to the idea of asking questions to find your morality. It’s vital to accept and analyze what makes different cases different”
As people have already said most people asking these questions in EA are trying to manipulate you. There is a 0% change a typical EA poster hears an answer and goes, “Hm, maybe you’re right and I’m wrong and EA is a waste of my time”. They are crafted specifically to generate outcomes they want and render the questioner immune to scrutiny.
The questions, especially in regards to finding moral answers, wildly misgauge how much people know and understand when they’re making moral choices.
Broadly I agree. You should be willing and able to answer such questions in your head, and out loud in good faith contexts where the question is philosophically meaningful.
A lot of the time the thought experiment is formatted to beg the question though. "Here is a politically point where I think the right answer is A but lots of people disagree. Here is a thought experiment which has been deliberately designed and ridiculously exaggerated so that only an evil person would pick any option other than A'. Here is a shaky and tenuous chain of logic trying to connect A to A' and convince you to support political option A."
Often A' is so obvious that people don't even bother to address that they would pick it. Everyone would choose A' except really strongly principled non-utilitarians. The entirety of the argument is that A' does not imply A because the thought experiment is badly constructed, so often the argument just starts there and stays there.
I don't dislike hypotheticals, but I can easily see from your two examples why people would dislike them. Singer makes the inference that you value children's lives more than $3000, even though you only said you value it more than a suit. However, there is no suit that I value $3000, and in real life, I won't own such a suit - I'll use $3000 to pay for rent - and I value not being homeless way more than $3000. These two scenarios are not equivalent. I can understand that people would be suspicious of hypotheticals which are then switched to a non-equivalent scenario which makes them look bad (same goes for chicken hypothetical).
I think hypotheticals are great, but trying to directly apply the outcome to messy real life is like trying to design wiring assuming zero wire resistance, like in high school. It's generally a bad idea. Details matter, and poorly designed thought experiment is about as valuable as poorly designed lab experiment.
Maybe I also should write something about why it's OK (to an extent) to be a bundle of contradictions.
I think we potentially agree.
I think that identifying in the moral weight between thought experiments and real life like you just did, acknowledging how $3000 might matter more to you than a stranger you can’t see or hear, is most of the way to “engaging with thought experiment.” Now you’re engaged in a different tradeoff between rent and another person, and you can grapple with your moral reasons for choosing one or the other.
Refusing on principle to use this structured logic, avoiding trying to examine the moral differences between hypotheticals and real life, is how you would end up going against whatever moral interests you have.
Thought experiments pin down one vector of morality so you can examine it against the others you hold dear. They’re how we prioritize.
Just giving a shout out to contradiction here. The hypotheticals are annoying because they presume there's much value in interrogating the consistency of our ethics in the first place. There's not. Right and wrong are too complex for a consistent ethical system to handle. Good and evil are too intertwined to isolate in a lab.
What's important is learning to see all the tradeoffs involved and interrogate your immediate values in actual complex situations, which is pretty much the opposite of isolating a choice in a contrived hypothetical circumstance. Those people dodging your trolley problem, looking for a way to have their cake and eat it, too--theyre doing that because that's the proper and healthy response to a difficult tradeoff!
Yes, eventually you might have to flip that trolly switch or watch even more people die, and whether or not you do in practice might boil down to whether or not you ate breakfast that day. But you should be spending every second you can spare looking for a way around the tough decision, or at least surveying and introspecting for particulars that might sway you in the actual circumstance. And if you are stuck making the trolly problem choice, guess what, both answers were evil! No virtue points for anyone, life just sucks sometimes!
People respond to your hypothetical by trying to dodge it because that is the ethical thing to do. God save us from the bros who are like, “Oh I already know what I'm gonna do we covered this class of problem in ethics class!” They tend to have more self-righteousness than actual knowledge of good and evil.
“However … in real life …” is not a valid response to any hypothetical. The whole point is to isolate decisions in a way that rarely occurs in real life. The premise is that you are wearing a $3000 suit, just as that there is a drowning child.
“In real life I would never walk along a riverbank” is just misunderstanding what you’re being asked, or being a jackass.
And I didn't respond that way! Singer brought up real life when he moved from the hypothetical to claiming that answering "yes" means you should donate $3000 right now. Since it's not longer a hypothetical, real life constraints now apply.
Right. Well, you did mention a “real life” objection to the original premise. Singer is perhaps misusing the structure, or at least skipping some steps in his advocacy. To me, the key to that hypo is not the exact price of the suit (“expensive” would suffice), but that no one else is in a position to do anything. Once you’re talking about IRL donations, there are lots of other people’s decisions involved, questions about priorities and effectiveness, and no particular urgency for the now-abstracted beneficiaries (because there will be someone in equal need tomorrow).
These are also good differences. For me, ultimately, what matters is that it's just a completely and obviously different situation. And I don't find it surprising that people would get angry about someone posing an easy hypothetical question, then smuggling much more complex situation while claiming they are the same. People can feel the slight of hand - and if you do it often enough, which unfortunately has already happened, they would just become hostile to all hypotheticals.
Most of the times when it may look like I don’t know how to answer a moral thought experiment, it’s because it’s yet another tedious deontological argument against consequentialism that relies upon the same sleazy error as good old Transplant, namely drawing on an instinctive revulsion that’s caused by the inability of our brains to ignore real-world negative side effects explicitly ruled out of the thought experiment. Very often I just feel like saying, “look, dude, I think a better world is better” and letting them dot the “i”s and cross the “t”s, instead of going, “Yes, if replacing every pediatrician with immortal vampire Jimmy Saville somehow made the world a better place overall, it would be a good thing to do.”
Oh yeah, in a bad faith context *acting* like a politician isn’t a bad idea. When I say “it’s a bad idea to let your enemies pin you down with evil questions for quotes” I mean it! You don’t want to be known as “the guy who chose to do that horrible thing in that hypothetical” because it was weighed against a staggering amount of abstract suffering that sounds less bad in vibes-world. But you DO need to grapple with the question in your head.
But I desperately hope people in their head know what the right answer is, and are trying to grapple their morality with what their answer to the hypothetical is. I worry that people really are just leaving their morality unexamined, and this post is definitely against doing that.
“Superpower” is super hard to define, so I wouldn’t fault people for not being able to answer that question. Basically, how god-like can you make yourself and it still be fairly called “a” superpower? Also that’s actually a kind of moral dilemma in itself. Your superpower of choice is absolutely sick and I think definitely counts as a superpower, but could some form of psychic healing at a distance count? Because if something like that would count, it would be really hard not to choose that (I’d feel morally obligated to) even though it would probably make life very tedious as I would feel a strong duty to use it as much as possible and to figure out which people it would be best to heal. But maybe that’s already not interpreting the question in good faith. Maybe that question is supposed to be answered selfishly: “Which superpower would you WANT the most, not which one would you choose?” Anyway, great post—thank you!
Now this is a fun point! And an interesting example of taking the thought experiment to its natural conclusion vs adversarially changing the terms.
Yeah, if I could actually, really choose a superpower, I would want something that allowed me to figure out the actual moral nature of good and evil, and the ability to make the future good. Something like omniscience probably would work there. BUT that’s kinda not in line with the spirit of the question, as you say! Follow up questions could be asked: “okay, so this is what I would WANT, but the most FUN superpower would be…” or “this is the superpower I instinctively crave the most”
Acknowledging the pros and cons and differences between thought experiments are good! Refusing to answer them or failing to examine any of the reasons behind what you pick what you pick is bad.
Great comment.
Enjoyed this piece! I do think "I find that hypothetical unpleasant to think about" is a potentially valid and honest answer, if you're willing to acknowledge that you feel you have a right to not think about deeply unpleasant things that you're not forced to by the universe. And before you say that's a bad policy (which it might be), you either need to acknowledge that everyone does this all the time---in fact, we almost never don't do it---or persuade me otherwise, which I think will be hard to do
Great post! Could you link the Jessie Ewesmont post please? Both links just lead to her profile atm
Oops, I’ll edit that when I’m back at my computer
https://open.substack.com/pub/jessieewesmont/p/taking-thought-experiments-seriously?r=2bgctn&utm_medium=ios
No.
It is precisely the hypothetical nature of the thought experiments that makes them unanswerable. I wouldn't throw a fat man on to the tracks because I wouldn't think it would stop a train and I couldn't lift him and if push came to shove he would probably throw me. Pretending that the many such issues dont exist make the question meaningless. And then if I do give an answer to keep you happy some eejit will draw random wild conclusions about my beliefs.
Giving $3000 to save a child is a completely different thing than giving $3000 to a group of strangers who supposedly do what they say with it, outside your line of sight, that hypothetically saves a child. Also, why are you trying to get $3000 from me instead of from someone with 3000x more than me, of whom there are many?
What if the hypothetical is disingenuous?
Have you stopped beating your father?
Killing yourself is always an option, so I believe you should at least be able to choose suicide in response to any hypothetical choice put to you.
Questions like "would you rather have snakes for arms or snakes for legs?" can be answered to full satisfaction with "that's a stupid question and I don't want to answer".
Many of the hypotheticals I see on TPOS are of the "too ridiculous to answer" variety. Moral choices made in the heat of the moment, like pulling a trolley lever or jumping into a pond, are far more dependent on constraints like courage and strength and quickness of thought than they are on passive moral law reasoning. I think the reluctance to engage with hypotheticals like the ones you name comes not just from moral weakness, but from a rejection of the framework that morality is just a math problem with feelings as the variables.
What's more, something like "pressing a button to inflict pain on a kid or animal" is the kind of moral choice that approximately nobody makes at any point in their lives, and I tend to think that moral reasoning from such a starting point will bear few insights. Many hypothetical questions can be readily rejected for simply being too silly to be worth the discussion.
"the superpower to create superpowers"
That is a power though. It also has moral implications or whatever you were trying to get out of the question.
“But half the comments are somehow people who object to the idea of asking questions to find your morality. It’s vital to accept and analyze what makes different cases different”
As people have already said most people asking these questions in EA are trying to manipulate you. There is a 0% change a typical EA poster hears an answer and goes, “Hm, maybe you’re right and I’m wrong and EA is a waste of my time”. They are crafted specifically to generate outcomes they want and render the questioner immune to scrutiny.
The questions, especially in regards to finding moral answers, wildly misgauge how much people know and understand when they’re making moral choices.
Voice reveal?
Broadly I agree. You should be willing and able to answer such questions in your head, and out loud in good faith contexts where the question is philosophically meaningful.
A lot of the time the thought experiment is formatted to beg the question though. "Here is a politically point where I think the right answer is A but lots of people disagree. Here is a thought experiment which has been deliberately designed and ridiculously exaggerated so that only an evil person would pick any option other than A'. Here is a shaky and tenuous chain of logic trying to connect A to A' and convince you to support political option A."
Often A' is so obvious that people don't even bother to address that they would pick it. Everyone would choose A' except really strongly principled non-utilitarians. The entirety of the argument is that A' does not imply A because the thought experiment is badly constructed, so often the argument just starts there and stays there.
No.
I don’t have to answer anything.
If your question is annoying, stupid, or otherwise ridiculous, my answer will be a fart.
And you will deserve it.
I don’t have to do anything except pay taxes and die.