I found this poll on BBC News and found the contrast in response between questions 2 and 3 interesting. I’ll quote the relevant sections here as they are brief:
The interesting point is that these are essentially the same dilemma- kill one innocent to save five- but the responses are markedly different.
How would you explain this difference?
My take would be that there is a difference in the level of human interaction between the two choices. One involves pushing a button - very depersonalized. Many people could take this choice and live with it because they could disconnect in their mind the cause and effect.
The other involves being directly responsible at a human-to-human level for the death of another person. While the former is semantically the same thing, we can trick our brains into thinking it’s not. “I just pushed a button” is a lot easier to digest than “I just pushed a man to his death”
I think a similar behavior mechanism is at play with road rage, too. When we’re behind the wheel, safe in our metal boxes it’s easy to forget there are actual people behind the wheels of the cars around us, so we tend to depersonalize.
Firstly, there aren’t many trolley cars any more, they’re mostly light rail vehicles. Secondly, these have bells and horns to warn persons that they are coming. Thirdly, any track switches in publicly-accessible areas are normally locked so that unauthorized persons can’t mess with them. Fourthly, what are all these mooks doing standing around on active railroad tracks? Lastly, I don’t care how fat they other guy is, there’s no way his body is bulky enough to stop a runaway rail vehicle that probably weighs 50 tons or more.
Sorry. Can you tell I used to work for a railroad?
Anyway, maybe the difference is a) throwing (not flipping) the switch is less immediate an act of murder than pushing a guy to his death; b) maybe most of the people who answered the second question found the situation absurd and were responding to that.
One difference is that in the second example, you’re definitely going to kill the guy (and probably not stop the trolley), but in the first, the potential victim has a shot at jumping out of the way.
So I’m strong enough to push someone who is, in turn, heavy enough to stop a rail car cold?
Why can’t I stop the rail car with those same arms?
Now I just need to find a phone booth to change real fast…
Sailboat
I think perhaps the perceived difference is that in the first case, you aren’t deliberately killing the one guy, you are simply diverting the car away from the track where it’s going to kill several people, and accepting the death of the other person. In the second case, you are deliberately killing the one (fat) guy, in order to save several other people.
Mind you, this is a pretty fine distinction to make, logically speaking, but emotionally I think that’s what’s going on here.
I think that there is also the case that poushing the big man off the bridge won’t guarantee saving the 5 others. Not in the way that switching the train tracks would guarantee saving the five others in the first scenario. Furthemore as has been allready stated the one man in the first scenario has the same chance of escape as any of the five. In the second scenario the one man has no chance of escape at all.
As with many of these hypotheticals it is usually best just to wait for the giant squid to eat the trolly car and save all six of the people standing on the tracks.
Some of you are reading way too much into the scenarios. #1 clearly states:
“five people who will definitely be killed unless you, a bystander, flip a switch”
“where it will kill one person.”
No “maybe”, no “might get out of the way”, no “jump to safety”. It’s a binary decision. Do nothing, 5 die; do something, one dies. There is no ambiguity in scenario 2, either. It says:
“five people on the track will be saved.”
“Next to you is a fat man. He would certainly block the trolley”
“although he’d undoubtedly die”
“A small nudge and he’d fall right onto the track below.”
No “maybe”, no “super-human strength” needed, no “he might not stop trolley”. Just the same binary decision - do nothing, 5 die; do something, 1 dies.
The only difference is that the “something” in scenario 1 is “push a button”. In scenario 2 it’s “push a person”. Everything else is identical. I think XJETGIRLX nailed it - it’s less personal, and therfor easier, for us to push a button than to push a fellow human being even if the resut is the same.
In the fat man scenario, the good result (five lives are saved) is a direct causal result of the bad result (pushing the single man in front of the train).
In the switch scenario, the single man dies as an unintended consequence of changing the train’s track (principle of double effect).
I think MEB is on the right track (sorry!) with this. In the first instance, you are saving 5 lives. In the second instance, you are taking a life.
On preview, what Sanders said.
In answering this question, I think most people would make the assumption that these decisions are split-second decisions, and that the answers may be different if the actor is given some time (say, a week) to decide on what course of action they take. A split-second decision that directly causes the death of another person could be seen as indicative of an immoral person, and would thus be a less moral action.
Trolley cars/streetcars/trams/light rail (the difference is often hard to agree upon), are actually relatively easy to derail. They have a much different wheel/rail profile and interface to heavy rail vehicles. Importantly the flange is generally much shallower. A human body would be enough to stop a trolley car, but only through the secondary method of derailing the thing, not by acting as a brake. Of course, the most LRVs are designed that a human body wouldn’t be swept under the front anyway.
Sorry. Can you tell I’m a rail nut? 
Now returning you to your thread…
My answer to the thread: if it’s split second, I have no idea what I’d do until I was in that situation
This may be how people interpret scenario 1, but it is absolutely not how it reads. The death of the one is NOT unintended. The scenario clearly states that “it will kill one person”. There is no difference between the decision in either scenario. Either 5 people die or one person dies, and you are the sole arbiter. Only the mechanism changes.
DJ: the point is that people do not read or handle the scenario like a computer reading a program.
They translate it into human terms, apply some “this is plausible, this is not” discount, then their emotions arrive at a decision and their logical brain comes up with a justification that the consciousness reports as “my idea on what to do in this situation.” I can’t give a cite, but AFAIK that is the prevailing mainstream view on how cognition works in most cases.
Given that that is how most people think, and certainly how most poeple would think in a real situation with seconds (or fractions thereof) to decide, its not surprising that the results are so different.
back to the OP: Since the yes & no answers total 100%, we don’t get a true picture of the decision-making because the question & available answers didn’t have enough options
I’d like to see choices of 1) I couldn’t decide before time ran out, 2) I decided not to get involved. 3) I decided to get involved, and my decision was to NOT throw the switch, 4) I decided to get involved, and my decision was TO throw the swtitch.
While choices 1-3 have the same outcome for the victims, they are very VERY different in terms of what they reveal about the person and the problem.
A single action can have many effects. That an effect is forseen does not mean it is intended.
The biggest difference is that, in the case of pushing the man in front of tracks, it is because of his death that the others are saved. In the case of switching the tracks, the five are safe, regardless of whether or not there’s a man standing on the other tracks.
But people will never respond to these questions EXACTLY as they are written… they will put their own judgement in to see if the scenarios described are believable, and if not, will let their own judgements for how something might not be exactly as it is presented modify their decision.
Even if the question says the fat guy will ‘certainly’ block the trolley, I don’t think most people would ever take that statement as a 100% assurance, nor should they. (In fact, something about the way that element is added to the scenario seems so patronizing that I distrust it immediately.) So ‘but what if he DOESN’T stop it?’ is a reasonable response for people to have, even if you’ve tried to steer them away from it with strongest possible phrasing in the question.
To a certain extent, that’s a skill that a lot of people have learned in their real-life decision making skill that just automatically carries over. ‘What is there in this situation that might not be as it first seems, and how could that affect the consequences of my actions?’ is a good question to ask most of the time. Even in this sort of opinion poll – if people are going to use the answers you give to their questions for anything important, and they might be unconsciously distorting or biasing the facts in the questions they ask you (hypothetical or not,) it makes sense to try to filter out that bias before you answer.