A and B are standing by the edge of a cliff. The cliff edge under A is crumbling and she will surely fall to her death if B does not grab her hand. If B grabs her hand, she will almost certainly save her, but there is a small but unknown probability that B will either fall off the cliff with A or will be injured in the process of saving A. You are a third party, C, who is watching. You cannot help A or B, but you can legally require B to help A, or even physically coerce B into rending assistance. Is it moral for you to do so?or
X volunteers to take part in an ethical experiment. He thinks it will be some simple classroom exercise, but instead he finds himself in a large room with only one exit. Blocking the exit is another person, Y who is tied to a chair sitting on a trap door. X can only leave he room by pressing a button which will cause Y to fall through the trap door into a meat grinder and be gruesomely killed. X is told that there is a 1 in 1000 chance that the roof will fall in and kill both of them before the experiment is over. If he pushes the button, he can go free but Y will be killed. If X waits till the experiment is over and the roof does not fall in, they will both go free. Is it right for X to push the button in order to be sure that he will be safe?Yes, I wrote one of these. (Actually, I kind of touched up both, but I think my modified version of the one I didn't originate is still true to the original's point.) The only excuse I can make for my behavior is:
MrsDarwin was listening to me discuss the course of argument and remarked, "Of course, you're not going to change any minds with that argument. It's too abstract." She's right, of course.
It strikes me that these sorts of "thought experiments" tend to be more extended rhetorical questions than actual experiments. It's really a way to say "I think that the moral situation we're discussing is like this." As such, the response from the person challenged with such a hypothetical is usually not to answer the hypothetical but to challenge the thought experiment or propose an alternate one. There's some utility to this, so long as you're just trying to understand how you think about something differently rather than reach any kind of agreement. However, it seems to me that it's seriously problematic to discuss moral questions in a context other than the way in which people actually face moral questions.