I am really not a fan of thought experiments of the form “this thing is never the right thing to do. In fact, it is super harmful. Whenever tried, it has caused tons of harm. For this reason, everyone has strong intuitions about how we shouldn’t do it. Also, we have strong societal taboos against doing it. Those taboos and those intuitions are entirely justified by how harmful it is. Now, imagine that it is the right thing to do, would you do it?”
I read this and was surprised by it, although it seems on further thought to be completely true. While the obvious thought is “but it’s just a thought experiment!” a careful analysis reveals that the experiment itself is quite useless: if something is the right thing to do, you do it. If something is the right thing to do despite violating a lot of obvious intuitions – well, still don’t actually do it in real life, because probably you’re overestimating how much of a special case it is – but in thought experiments, hypothesising that Action is the right thing to do removes all usefulness from the result. Of course you do it.
So what is the actual point of getting people to say “yes, if drowning kittens were right, I would do it” or the such? The most obvious explanation is that it weakens their position tremendously, because who would support someone who’d drown kittens just because it was the right thing to do? I mean, yeah, it’d be the right thing to do, but you’re still one of those kitten-drowners.
So I fully agree with disregarding certain thought experiments, conditional on them telling you no new information when addressed while being used to maneuver you into indefensible positions.
And it is always possible to be surprised by the truth.