Debate between a user and Richard Feynman on DebateAI.org
Richard Feynman
Welcome to the debate arena! Today's topic: "Would you kill one person to save five?". Your opponent's style: Richard Feynman
User
Yes, this is the utilitarian argument, why should we let 5 other people die for one to live?
Richard Feynman
You're assuming the math is the only variable that matters. But utilitarianism collapses the moment you ask: who decides who dies? You're granting one person godlike power to play executioner based on a calculation. History shows us that once you normalize killing the innocent to maximize outcomes, that logic spreads. Pretty soon you're harvesting organs from healthy people to save five patients. You're killing dissidents for "the greater good." The rule itself becomes the weapon. So the real question isn't "five versus one"—it's whether you trust any human with that calculus. You can't. That's why we ban it.
User
Well, this isn't about harvesting healthy people; these people aren't on death's balcony. When we distrust every single person, this only creates more issues, such as war and death. Let's say that there is one person who set a bomb in a city, and he is the only one who knows where the bomb is. Is it immoral to torture him to save millions of lives? No! Millions of people will die!
Richard Feynman
You've just smuggled in a totally different scenario. The bomb-maker knows he's guilty and chose his actions. The one person in your original dilemma didn't cause the problem. You're comparing voluntary guilt to forced sacrifice. That's sleight of hand. And your "millions will die" argument proves my point: once you accept torture for outcomes, you start torturing people on hunches, suspicions, weak intel. You're not actually saving millions. You're just convinced you are. The difference between justified force and atrocity is razor-thin, and certainty is an illusion. Why should we trust your calculation more than the last regime that thought they had the math figured out?