4 ms·

I don't think that's a flaw; that's addressed in the very next essay in the sequence, titled "Purchase Fuzzies and Utilons Separately."[0] The idea is that if y

by YellOh 3y ago

I don't think that's a flaw; that's addressed in the very next essay in the sequence, titled "Purchase Fuzzies and Utilons Separately."[0] The idea is that if you're a lawyer that's trying to do maximum good, food kitchen work is a very poor trade for you. But if you're trying to get a warm fuzzy feeling, working in a soup kitchen may be pretty effective. Purchasing warm fuzzies is sometimes efficient (from an emotional or long-term mental health perspective), even though it's very inefficient at doing good directly. But ideally you should be able to be honest with yourself about what it is you're actually buying with that hour. [0] https://www.lesswrong.com/posts/3p3CYauiX8oLjmwRF/purchase-fuzzies-and-utilons-separately https://www.lesswrong.com/posts/3p3CYauiX8oLjmwRF/purchase-f...

vorpalhex 3y ago
The problem is that whether or the lawyer is attempting to buy warm fuzzies or understand other people is reliant on internal state of the lawyer. As external observers, we can only hypothesize based clues. Sometimes I do my own drywall work at my house. Not because it's more efficient (it'd be better to hire someone) but because I enjoy it in this sort of zen way. That's a really weird abstract sort of reasoning that utility calculations can't capture well. If the lawyer works in the soup kitchen because the rest of their time is spent around rich people and they want to better understand the whole of human experience, that also can't be factored into the utility calculus.
- YellOh 3y ago
  The reason it's a puzzle is because most people, usually including the lawyer themself, assumes soup kitchen volunteers are volunteering for the purpose of helping others. If the primary purpose of the hour volunteering is not to bring maximum good to others but instead to enjoy the hour or socialize, then the hour may be spent efficiently for the lawyer, but it is not pursuing the primary purpose of "caring for others' good." Any (functional) utility calculation does take into account internal state & knowledge gained. Pretty much by definition, utilons are internal-only. The lawyer's enjoyment of the hour is factored in; however, if the point of the hour is to maximize total utility across all humans, the lawyer would be better to spend an extra hour at work (unless soup kitchen work provides a large enough multiplier to mental health, or they wouldn't be motivated to do lawyer work without the hour at the soup kitchen). Even in cases where soup kitchen work is better in the long-haul, those hours in the soup kitchen are spent buying fuzzies, not utilons directly. This only works from a utilitarian perspective. If the lawyer getting maximal enjoyment from their good is more important than the amount of good produced, then soup kitchening will ~always be more effective, but you're no longer engaging with the core philosophy of the essay.
  - vorpalhex 3y ago
    I disagree that the benefit is either for the lawyer or others and that the utility calculus can correctly capture this. I have a good friend who is a lawyer who does a mixture of family law and immigration law. He is wealthy, and spends most of his time around other wealthy people. Sometimes he volunteers to do (non-lawyer) work with immigrants. He doesn't do it for fuzzies. He might explain it as attempting to better understand his client pool. He has to understand, to a good enough degree, their lives and mental state to best work with them. There is a calculus here. Some omniscient entity could work out a grand formula of "How much does this time spent with immigrants help him become a better lawyer to help immigrants". That is a knowable variable.. But it can't be known by us. It's too much of an internal state. Even if we had high quality assessments for "skill as an immigration lawyer" and tested him before and after his volunteering, we wouldn't see a linear correlation. He is hoping for "aha" moments of understanding. Thus the problem is a utilitarian problem. There is a calculus of how much he improves the lives of others. Yet the utilitarian calculus can't capture the value of these volunteering events yet they do feed into his ability. Utility calculus always has this flaw. Humans are not cold logic utility machines. We require a mix of soft (culture, religion, emotion) inputs and hard inputs (training, logic) to do our best. Even though we can have mathmatical conversations about the soft inputs, we can't actually run the utility calculus on them.
    - YellOh 3y ago
      Yes, nobody (including utilitarians!) thinks you're actually able to calculate the exact number of utilons without omniscience. There aren't probes in your brain to determine what fraction of a utilon you get from looking at a cute puppy. Lots of utilitarians try to determine aggregate effects on utility given the data we have available, and then use that as input to make best-guess shorthand rules for maximizing utility. And generally, efficient charities > inefficient charities, and donating [from well-employed people] > fuzzy-maximizing hours. YMMV and it doesn't apply to every possible lawyer or situation, only the majority of them. (Critical thinking still required.) Hence why so many utilitarians use contractualism/virtue ethics/etc for day-to-day decisions; utility is the "reason" that informs big-picture decisions, but it's too impractical to start crunching numbers at every possible crossroads. See also: the many, many warnings against "naive act utilitarianism" even from within big utilitarian groups (LW, EA, 80k, etc.). Pretty sure we've veered off from the actual problem the original post is getting at, but if you want to talk more about the practicalities of utilitarianism, I'd be very happy to (or to link to SEP/LW/EA essays that dig more into this issue). :-) One great book that addresses this issue specifically is Against Empathy, on the limits of reasoning by what "feels" best (while still respecting that it's impossible to perfectly calculate utility irl); would highly recommend if you're interested.

by YellOh 3y ago

vorpalhex 3y ago

The problem is that whether or the lawyer is attempting to buy warm fuzzies or understand other people is reliant on internal state of the lawyer. As external observers, we can only hypothesize based clues. Sometimes I do my own drywall work at my house. Not because it's more efficient (it'd be better to hire someone) but because I enjoy it in this sort of zen way. That's a really weird abstract sort of reasoning that utility calculations can't capture well. If the lawyer works in the soup kitchen because the rest of their time is spent around rich people and they want to better understand the whole of human experience, that also can't be factored into the utility calculus.

YellOh 3y ago
The reason it's a puzzle is because most people, usually including the lawyer themself, assumes soup kitchen volunteers are volunteering for the purpose of helping others. If the primary purpose of the hour volunteering is not to bring maximum good to others but instead to enjoy the hour or socialize, then the hour may be spent efficiently for the lawyer, but it is not pursuing the primary purpose of "caring for others' good." Any (functional) utility calculation does take into account internal state & knowledge gained. Pretty much by definition, utilons are internal-only. The lawyer's enjoyment of the hour is factored in; however, if the point of the hour is to maximize total utility across all humans, the lawyer would be better to spend an extra hour at work (unless soup kitchen work provides a large enough multiplier to mental health, or they wouldn't be motivated to do lawyer work without the hour at the soup kitchen). Even in cases where soup kitchen work is better in the long-haul, those hours in the soup kitchen are spent buying fuzzies, not utilons directly. This only works from a utilitarian perspective. If the lawyer getting maximal enjoyment from their good is more important than the amount of good produced, then soup kitchening will ~always be more effective, but you're no longer engaging with the core philosophy of the essay.
- vorpalhex 3y ago
  I disagree that the benefit is either for the lawyer or others and that the utility calculus can correctly capture this. I have a good friend who is a lawyer who does a mixture of family law and immigration law. He is wealthy, and spends most of his time around other wealthy people. Sometimes he volunteers to do (non-lawyer) work with immigrants. He doesn't do it for fuzzies. He might explain it as attempting to better understand his client pool. He has to understand, to a good enough degree, their lives and mental state to best work with them. There is a calculus here. Some omniscient entity could work out a grand formula of "How much does this time spent with immigrants help him become a better lawyer to help immigrants". That is a knowable variable.. But it can't be known by us. It's too much of an internal state. Even if we had high quality assessments for "skill as an immigration lawyer" and tested him before and after his volunteering, we wouldn't see a linear correlation. He is hoping for "aha" moments of understanding. Thus the problem is a utilitarian problem. There is a calculus of how much he improves the lives of others. Yet the utilitarian calculus can't capture the value of these volunteering events yet they do feed into his ability. Utility calculus always has this flaw. Humans are not cold logic utility machines. We require a mix of soft (culture, religion, emotion) inputs and hard inputs (training, logic) to do our best. Even though we can have mathmatical conversations about the soft inputs, we can't actually run the utility calculus on them.
  - YellOh 3y ago
    Yes, nobody (including utilitarians!) thinks you're actually able to calculate the exact number of utilons without omniscience. There aren't probes in your brain to determine what fraction of a utilon you get from looking at a cute puppy. Lots of utilitarians try to determine aggregate effects on utility given the data we have available, and then use that as input to make best-guess shorthand rules for maximizing utility. And generally, efficient charities > inefficient charities, and donating [from well-employed people] > fuzzy-maximizing hours. YMMV and it doesn't apply to every possible lawyer or situation, only the majority of them. (Critical thinking still required.) Hence why so many utilitarians use contractualism/virtue ethics/etc for day-to-day decisions; utility is the "reason" that informs big-picture decisions, but it's too impractical to start crunching numbers at every possible crossroads. See also: the many, many warnings against "naive act utilitarianism" even from within big utilitarian groups (LW, EA, 80k, etc.). Pretty sure we've veered off from the actual problem the original post is getting at, but if you want to talk more about the practicalities of utilitarianism, I'd be very happy to (or to link to SEP/LW/EA essays that dig more into this issue). :-) One great book that addresses this issue specifically is Against Empathy, on the limits of reasoning by what "feels" best (while still respecting that it's impossible to perfectly calculate utility irl); would highly recommend if you're interested.