Agathon, thanks for clarifying. 
This reminds me of dodgeball games, where you have a medic who can pick people up who are hit. Players will throw themselves in front of the medic to ensure that the medic doesn't get hit, and thus, removed from the game. The medic is more important because of his ability to revive downed players.
Real life isn't a dodgeball game, though. In the situation you describe, I would rescue whichever one I felt had the best chance of surviving, typically the one closest to me, or the one least injured. The problem with gauging 'utility' is that it requires us to assume what we do not know.
How do we know that more people will live if we rescue the medic? Perhaps the medic will be killed tomorrow.
Perhaps the soldier will save many lives by defending his country? We simply do not know.

Well it isn't. Imagine that in a war you have the option of saving an injured infantryman or an injured medic. If you save the medic he will be more likely to save other lives than the infantryman will. So if you care about minimizing the number of deaths, the survival of the medic is more important. QED
Real life isn't a dodgeball game, though. In the situation you describe, I would rescue whichever one I felt had the best chance of surviving, typically the one closest to me, or the one least injured. The problem with gauging 'utility' is that it requires us to assume what we do not know.
How do we know that more people will live if we rescue the medic? Perhaps the medic will be killed tomorrow.
Perhaps the soldier will save many lives by defending his country? We simply do not know.
Comment