If I understand the orthogonality thesis properly then it is possible to have any utility function. So if we were to try to satisfy the nonpersonal preferences of nonexistant people we would be paralyzed with indecision, because for every nonexistant creature with a utility function saying “Do X” there would be another nonexistant creature with a utility function saying “Don’t do X.”
This also means that your suggestion to “get some reasonable measure across the preferences of never-existent people, and see if there’s anything that sticks out from the mass” probably wouldn’t work. For everything that stuck out there would be another thing that stuck out saying to do the opposite.
Now Robin Hanson, who I think was responsible for starting this whole line of thought, replied to similar objections by suggesting that maybe not all non-existant people’s preferences should count morally. But if I recall, one of his main arguments for taking nonexistant people’s preferences into account in the first place was that if you started ignoring people’s preferences, where would you stop?
So overall, I think that regarding the preferences of people who don’t exist, and never will exist, as morally relevant, isn’t a good idea.
If I understand the orthogonality thesis properly then it is possible to have any utility function. So if we were to try to satisfy the nonpersonal preferences of nonexistant people we would be paralyzed with indecision, because for every nonexistant creature with a utility function saying “Do X” there would be another nonexistant creature with a utility function saying “Don’t do X.”
This also means that your suggestion to “get some reasonable measure across the preferences of never-existent people, and see if there’s anything that sticks out from the mass” probably wouldn’t work. For everything that stuck out there would be another thing that stuck out saying to do the opposite.
Now Robin Hanson, who I think was responsible for starting this whole line of thought, replied to similar objections by suggesting that maybe not all non-existant people’s preferences should count morally. But if I recall, one of his main arguments for taking nonexistant people’s preferences into account in the first place was that if you started ignoring people’s preferences, where would you stop?
So overall, I think that regarding the preferences of people who don’t exist, and never will exist, as morally relevant, isn’t a good idea.