The Trolley Problem And Experimental Philosophy

The Trolley Problem, as given by Eric Schwitzgebel (as part of a larger survey):

You are standing by the railroad tracks when you notice an empty boxcar rolling out of control. It is moving so fast that anyone it hits will die. Ahead on the main track are five people. There is one person standing on a side track that doesn’t rejoin the main track. If you do nothing, the boxcar will hit the five people on the main track, but not the one person on the side track. If you flip a switch next to you, it will divert the boxcar to the side track where it will hit the one person, and not hit the five people on the main track.

What would you do? The question is clearly a moral one. Pull the switch and you knowingly kill one person. Don’t pull and you will be responsible for killing five persons.

Is that a harsh way to put it? Perhaps you could rephrase the consequences of the dilemma so that it doesn’t put so much of the burden on you. Thus: pull the switch and unfortunately one person will give his life for five; don’t pull and you leave all to their fates.

Schwitzgebel polled academic philosophers on the dilemma:

Accept or lean toward: switch 635 / 931 (68.2%)

Other 225 / 931 (24.1%)

Accept or lean toward: don’t switch 71 / 931 (7.6%)

These kinds of results are increasingly being used to create the area of experimental philosophy (I don’t say Schwitzgebel does this, but he did create a survey).

Two-thirds of professional, academic philosophers—men and women who are tenured and accustomed to ponder daily the great and fundamental questions of life—have decided, by majority vote, that the right thing to do is to pull the switch. Therefore, the right thing to do is to pull the switch. Philosophical conundrums solved by survey!

Don’t know the right moral answer? Eyes blurring over after wading through tomes of turgid prose? Spent years on a problem only to discover that you are no closer to the answer than all those who came before you? Then just poll enough people so that statistical significance is reached and the best answer will emerge!

It is, of course, a fallacy to say that the answer to a question is true because a group of academic philosophers agree that it is true. Philosophers acknowledge this fallacy, thus nearly all—strict moral relativists being an exception—do not use survey results as proof of their beliefs. So why experiment?

Well, surveys can possibly give information that clarify questions; though it seems that answers are just as likely to distract or mislead. But even if experimentation is of little use in providing information to answer philosophical questions, it does guarantee endless opportunities for publishing “research,” a problem greater in importance to academics than any other.

The survey for the trolley problem exemplifies the difficulties. The first is in the wording of the question and how that wording is interpreted in your mind. No two people are going to think about it the same way. Everybody will imagine the events unfolding in a different way. I cannot even write down all my “premises” in the trolley problem. That is, I would not know what I would do until I was actually forced into the situation.

The experiment is clear: you either pull the switch or not. It is not said, so you must imply that you have time to pull the switch and the knowledge of how to operate the switch. You must believe that it is certain that people will die and that, say, some of all of the five or the one will sense the train’s approach and jump clear in the nick of time. You must assume that you do not favor the one or the five in some known way (age, sex, beauty, friendship, hatred, etc.).

True, the problem says, in effect, “They will die with certainty”, but that does not mean that certainty translates into a fixed premise in your mind. When I imagine the problem, I cannot help but think that somebody will at least hear the train.

Proof of this confusion is provided by Schwitzgebel’s experiment. One-quarter of the respondents said they would do “Other”, but there is no “other” offered! It’s pull the switch or sit still. What is “other”? Shouting out the window? Waving your arms wildly? Who knows?

It is clear that the trolley problem—like most moral problems—cannot be reduced to a Benthamite calculus. Most moral questions are so complex that they must be adjudged on a case-by-case basis.


  1. O/T
    Is it telling that the philosophers polled did or could not suggest alternative solutions to the Trolley Problem? That they did not question why only those two specific options were available? Such a lack of curiosity or initiative bothers me.

    Is it also telling that the dilemma is mislabeled as a Trolley Problem when in fact the conveyance moving down the tracks is a railway boxcar? They are truly not the same.

    Sorry to be a spoil sport.

  2. Thanks for the interesting treatment of this thought experiment. It is usually trotted out by liberals in a feeble attempt to make some point about abortion.

  3. Briggs


    That so? In what way?

  4. Adam H

    Scientists prove that 1 live is less important than 5 lives! Only 5% chance of coincidence!

    A related question: would you push an extra fat person in front of the trolley in order to smooshily grind the trolley to a halt before it hits the 5 people standing in the tracks? The mental picture is just so much more vivid…

  5. DAV

    To solve a problem like this it’s necessary to get on the right track. The answer is simple. First, you must ask why anyone would be insane enough to stand on the tracks. The group of 5 are obviously operating under herd mentality. This is a sad condition but often unavoidable. The lone track wanderer has no one to blame but him(her?)self. Throw the switch.

    On the other hand, how often does one get a chance to reduce the carbon footprint of humanity? Not throwing the switch takes less action than pushing a big red Easy Button in a You Tube skit.

    For those who find all of this too confusing, take Option C: get someone else to decide; order a cup of tea; then sit back to watch the fun while humming:

    “I’d love to change the world
    but I don’t know what to do
    so I’ll leave it up to you”
    — A. Lee

  6. Sander van der Wal

    What about the punishments in both situations? Liabilities? Not for wanting to to be a spoilsport, but if the punishments differ then the local society has already decided what the best moral behavior has to be.

  7. Ken

    “Influence, The Psychology of Persuasion,” by Robert Cialdini is considered a sort of “definitive” summary of how people behave in various situations, with some insight provided into “why” but without any philosophizing to muddle the observations presented (e.g. he did an experiment in which priests or seminary students [or something] were to give a talk on the story of the Good Samaritan…and a needy person was planted along their walking path — the inclination of the ‘priests in waiting’ correlated directly with how late they might or might not be).

    As for switching in this story: sit back & do nothing. THAT is one’s safest action.

    When one takes an ‘affirmative act’ that leads to some outcome that otherwise would not have occured, one becomes legally liable for that outcome. Though how much is difficult to predict, and depending on the jurisdiction may be much more or much less than so-called “common sense” would suggest.

    Unless one has reviewed legal cases on a given topic area, and jurisdiction, one should be profoundly prepared for radically divergent case-law decisions relative to what so-called “common sense” would indicate.

    Do nothing and events unfold per forces having nothing to do with you. Do something…and someone dies as a result and their heirs can, and will if they think they can profit from it, sue you. This is true for administering CPR…though many locales have instituted “Good Samaritan” laws that protect one from less than egregious malpractice. Its hard to argue that getting someone killed that otherwise would have lived is protected under such a legal construct. And, frankly, in my selfish opinion, MY LIFE & financial situation is more important than those five.

  8. ad

    It would be against the law to interfere with railroad property. I would have to do nothing.

  9. Andrew Kennett

    This is always an interesting topic and to make it more so we can use the survey to compare different countries (US v China?), or age groups (over 60 v under 20) or this year v ten years hence or AGW alarmists v AGW deniers etc etc. Maybe before any political or philisophical discussion ones Trolley Problem choice should be clearly stated so we know who is talking (for full disclosure: I agee with our host that until I have to make the choice I can’t be sure what I would chose but I guess I would, given no time to think too much or find out more info or another choice, pull the switch)

  10. Speed

    It is a problem that belongs in the economics department where they would spend exactly three seconds thinking, flip the switch and then have an early lunch.

  11. Kevin

    Well, a skilled railroader would quickly;

    Throw the switch sending the front truck (wheel/axle assembly) of the trolley onto the siding, and then;

    Throw the switch back before the second truck arrived.

    This would derail the trolley right after the switch and save all SIX people.

    Nice hypothetical solution to a hypothetical problem.


  12. Alan D McIntire

    ad raises a good point. At least in the US, by flipping the switch I would be opening myself up to a lawsuit by the heirs of the single individual killed. Their lawyer would argue that the other 5 would have been aware of the oncoming boxcar, and would have gotten out of the way in time. Besides risking a large civil judgment I’d be risking a criminal charge of voluntary manslaughter if I flipped the switch.

  13. Doug M

    Can we assume that we know something about the way that trolley switches opperate. Would I answer the question differently if I worked for the municipal railway.

    Hypotheticals are nice in theory. Should the hypothetical actually play out, I don’t have time to ponder the question. My action, reaction or non-action will come down to instinct or training.

    However, there are people who must do the calculus to endanger one life (or many peoples lives) to possibly save annother. Captain Kirk taught me that sometimes the needs of the few, or the one, outweigh the needs of the many.

  14. StephenPickering

    I think the principle ‘first, do no harm’ (as in the Hippocratic oath) has much to recommend it. After all, the outcome of flipping or not flipping the switch cannot really be known for certain. By artificially insisting that there are only two possible outcomes, the problem as stated removes some of the uncertainty necessary for the unrestricted exercise of free will. It thereby tends to reduce a moral problem to a utilitarian calculation. The reply ‘other’ may reflect a rejection by nearly a quarter of the respondents of the presentation of the problem as a moral problem.

  15. If two thirds of mathematicians agreed that a conjecture was true, it would be regarded as not proven.

  16. Richard

    Half thow the switch. All mecahnical rail switches have a position (quite possibly difficult to maintain) in which both rails are not pointing in the ‘right’ direction. Should cause a failure at the points and thus ‘save them all’ 🙂

  17. “Most moral questions are so complex that they must be adjudged on a case-by-case basis.”

    The trolley question is useless because it is not well defined. What is it about exactly? Most people asking or answering the question will make the huge assumption that the 6 people are unique individuals, but the only information provided is that the people are alive (and standing).

    If we assume the people described are unique individuals, then we have added information that was not part of the original question. And we would add a huge amount of information to the question since to believe the six people are individuals would mean we would have created histories and futures for each of them, even if we acknowledge those histories and futures are inaccurate / imaginary.

    Since the response expected is only yes/no/(other) without asking about what other assumptions or additional information were added by the answerer, it’s hard to see what value the collection of responses would have – other than “if you give people an ambiguous question, they may or may not add additional information or assumptions in order to make it a coherent / understandable / sensible question. And respond accordingly.”

    Contrarily, if no additional qualities are allowed about the six people, then the word “people” could be replaced with cows, dogs, snakes, cockroaches, whales, or petunias – all of which have the same quality of being alive (though the whales would probably have difficulty standing).

    So…given a choice to save 1 cockroach or 5 cockroaches, which would you do?

Leave a Reply

Your email address will not be published. Required fields are marked *