They share a single Audum.

That is another interest tangent to explore.

What level of 'active involvement' would change the views.

So for people like Inso0 who think the answer is clear and easy and the person is wrong/complicit for not acting solely based on the math what about if the choice to save the 5 by sacrificing the 4 involved more active participation, as you say.

Instead of just throwing a switch from a distance, you had to go into a room and pick from a group of people the 4 people you select to die to save the 5?

Or, next level, you choose the 4 and must be the one to kill them for the 5 to survive.


In a perfect math, pure logic, driven decision neither of those added factors should matter. 4 lives lost to save 5 should be the only calculus. But I think we all recognize it would not play out that way.

So to David's question ...Why not?

No. your answer isn't "why not". It's "it will make me feel badly and that is more important than reducing the number of random (to me) deaths."

I recall polling data that says most people flip the switch to reduce deaths in the original trolly problem, but most do NOT flip the switch in the parallel scenario that is more active vs passive. For instance, pressing the button that authorizes a team to go and kill a healthy person to harvest five organs that saves the lives of five people only sick because of their organ failures.

This would only apply to the original question if the five you save would have also been beaten by a hammer.

THE ONLY ASPECT OF THIS QUESTION THAT IS SUPPOSED TO COME INTO PLAY IS THAT YOU ANONYMOUSLY HAVE CHANGED WHO WILL DIE (ALL BY THE SAME METHOD) TO DIFFERENT BUT A LESSOR NUMBER OF PEOPLE WHO YOU HAVE ZERO INFORMATION ABOUT.

Different people have died because of you. Only you even know about it. That's it.

I am not disagreeing but to be clear are you saying that even if you had to go in a room, select the 4 to die to save the 5, and kill them yourself, you are saying the answer should be purely math driven and easy or evident?
(Edit I see your last post now)

Those feelings matter of course but it's beside the main point.

To hopefully make this clear let's suppose we're programming a very advanced AI to make these decisions. There will be times when, for example we want it to minimise the deaths of a crashing vehical but also times when for example, we dont want it to harvest organs to minimse deaths. We're not going to program every possible scenario, were going to teach it what matters when making the decision so it deals with all scenarios as best as possible. The AI has no feelings and lets even give it perfect enough knowledge of the potential death counts.

What concept might we teach the AI?

- Minimize death (loss) is first imperative
-

so organ harvesting then

That's one option. Is it really what you/society wants?

In programming AI i assume the other bullets would potentially moderate or create exceptions to the prior stated principles.

So while you teach the AI the first and highest priority to rank is "minimize death', that would not prevent you giving it rules to create a very high bar of prohibition, such as 'organ harvesting'.

Edit :

Think of it like the Star Trek Prime Directive. "Non-Interference Directive, or the principle of non-interference." Just because it was the top and highest thing one must consider does not mean there were not a number of lesser considerations you could list and program that moderate it. But you want it known it is the top or highest principle to aspire towards.

Big fan of the default approach although i'm not sure about it here (no doubt it can be made to work).

R1: Minimise deaths iff not minimize deaths

But then all the interesting detail is in rules for 'not minimise deaths' so we haven't answered anything yet.

David, you don't have to go ALL CAPS on me. I of course am aware that I am modifying the original trolley problem. And I answered the original question as posed.

My point was that physical or moral revulsion on occasion might cause people not to act in a situation where acting would be the rational choice.

Q1.b. Would you sacrifice your own life to save two random people?

Zero % say 'hell naw'.

Or so I am guessing anyway.

That is all very nice but the question of when a certain number of deaths is preferable to another outcome is not the trolley problem. The problem is whether it is Ok for you to not intervene to cause that preferable outcome, when the only reason not to is because you don't want to be the reason for those deaths.

I dont think you've captured that in the question but if you want to ask it directly then for me it's not a good reason. It may be ok for those who think it is a good reason.

I'm pretty sure once you have the power to pull the switch you're responsible for the outcome.

If you refuse to act you cause an outcome just as if you do act.

Sucks to be you.

Then again, humans are arguably a scourge to the universe and 2 less may the the +EV result.

I think this is a perspective thing. If you do nothing then what happens is no different to if you were removed from the scenario altogether, and that does seem to be something different to you having an active role in changing the scenario. Largely which perspective we take will depend on the context and whether we see the consequences of extending that thought as good or bad.

Most people will agree that inaction can be immoral. Easy example, child neglect. But not all forms of inaction, even with tangible negative outcomes, are going to be seen as immoral.

Real world example: I have five pounds in my pocket. I'm going to spend that five pounds on something trivial, and I'm pretty sure I'll never notice the loss. However, I'm also aware that if I donated that five pounds to a malaria charity to buy a couple of mosquito nets that it would be preventing a serious illness, probably even a death or two. We probably all think we should do more charity than we do, but I don't think we all walk around thinking we're moral monsters for not giving away all our expendable income. Certainly not in the very palpable way we would for child neglect. From that perspective, I think it's far less intuitive that inaction and action are the same.

Do you mean that to be so absolute?

A lot of people, myself included do not accept or believe that.

For instance in the Trolly example lets put it at 4 and 4 with no discernable differences in the people.

4 people will die on the Trolly when it crashes if you do nothing. Corresponding those 4 will live but 4 other people will be killed as a result of you pulling the level.

In that example I think you have no responsibility for the 4 who die on the Trolly if you do nothing but i think you do have responsibility for the 4 others dying if you do.

Some may argue it is morally neutral but that is different than responsibility.

There are 5 people about the burn alive inside a building. You're standing outside listening to the screams. You can simply open a door and let 2 of them out with no risk to yourself, but the other 3 are stuck under some beams and can't be saved. Do you open the door, or do you say, "Not my business what fate has decided for these people" and just keep walking?

That's pretty much what the OP problem boils down to. You aren't responsible for those 5 impending deaths, but you ARE responsible for not saving 2 of them if given the easy option and you decline to do so.

Edit: I get that there's a slightly different dynamic for the OP in that you're pulling a lever that redirects the train to people who otherwise would not have died, but that's also why you ultimately just have to look at it as a numbers game. You've been put in a position to decide who lives or dies regardless. It's better to limit the loss of life.

Rflush: D2 you're a horrible misanthrope
Rflush: Wait so am I.

Obviously we don't see neglecting the needy as evil. If we did there wouldn't be many needy people in the world.

For some reason we see letting two people needlessly die as possibly evil, at least in the context of the scenario. We're sort of wired randomly.

But I get your point. In real life there's a pretty wide gray area.

I agree.

But in the scenario given there are two additional lives at stake.
So you have to take the responsibility for the two/three you can save and the three/five you can allow to die.

That's why it sucks to be you.

We have to consider this from a hedonistic pov rather than simply 'responsibility'

We're trying to maximise our satisfaction so if we can walk and not feel repsonsible while we would feel responsible for the deaths if we acted then we 'should' walk away. And vice verca

As a society we can change what will satisfy individuals with norms and laws. In particular because this problem incorporates a lot of angst, if people know how they are expected to act when faced with these problems then that can dramatically chamge how they will feel about their action and hence they will often act far more in accord with what is expected.

I agree we should do more than what we do but the reductio then typically goes further and say any time if it is the case that me not giving my five pounds to charity is the same as an active role then I should not only give this fiver to malaria, but another fiver, and another, until I hit whatever level of basic subsistence is my minimum for survival. Any time we aren't actively doing good then we're not flipping the switch, after all.

One problem with this objection is that the trolley problem doesn't entail any self-sacrifice where some tipping point comes with charitable donations where it does impact my life. But then, if inaction is really equivalent my whole life is a perpetual sequence of trolley problems and I can't escape that.