What does Judith Jarvis Thomson's looped trolley problem show about Kant?

the bystander does not need or use the one workman to save the five, because the latter’s presence on the track contributes nothing to achieving the agent’s end of saving the five. By contrast, the agent on the footbridge does need to use the one if he is to save the five. Thomson... introduces a modified Bystander scenario called the Loop case.


Looping Trolley: This is the same as Switch, except that the side track with one person on it is actually a circle, which loops back on to the main track. If you were to pull the lever WITHOUT one person on the side track, the trolley would merely loop around back onto the main track and kill the five. However, since there IS someone on the main track, the trolley hits them and is stopped by their body, thus saving the five.


For some reason, I find it easier to will sacrificing the life in the loop version of the bystander case. I think it was designed to be the opposite, however, so I'm not sure what that says.

Perhaps the traditional bystander case usually seems to involve a kind of double think, that just not directly intending any harm is sufficient for morality, and the looped version doesn't

  • I would figure that you don't always use people as a means by sacrificing them, even their life, for "the greater good". obviously there are cases when you do, but not every consequence of an action is a means of it, surely (just as an enforced famine or bombing campaign for the greater good seems more means based than feeding your family first even if they are not most at risk of starvation)
    – user56770
    Nov 16, 2021 at 12:47
  • It is unclear what you want "any way" to accomplish. "That it shows Kantian morality does show that we should throw the lever, whether or not we should do so only in loop cases." What is "it", and when should we "throw the lever" according to your way? In the Bystander case, Loop cases, both? Is that supposed to be matched by Kant's morality? "I cannot will inaction in either of those scenarios". Which scenarios exactly? Bystander and Loop only, Footbridge also?
    – Conifold
    Nov 17, 2021 at 21:29
  • maybe I thought it was a choice between 1 or 6 lives, I can't recall now @Conifold
    – user62133
    Aug 15, 2022 at 18:23

3 Answers 3


I think what it shows is that people don't use philosophy to solve moral problems.

Given responsibility over some set of human lives they are not overly attached to they recognize that sacrifices may have to be made for "the greater good" and will pull a lever and wear a poppy once a year. When asked to murder someone in order to achieve that "greater good" they will balk at it due to emotions and society's dim view of murderers.

  • A funny 😁 and yet intriguing 🤔 remark. Danke!
    – Hudjefa
    Feb 15, 2023 at 6:14

Apparently the conceptual problem is as following:

In the bystander case you decide by pulling or not pulling a switch whether 1 or 5 people will die. In the footbridge case you use a person to block the trolley, thereby actively killing them. And in the looped version you divert the trolley which then uses a person to block the trolley and prevent the death of the other 5.

And now apparently Thompson's argument seems to be that using Kant's idea of "never treating another person as just a mean to an end" would make the 1st permissible the 2nd impermissible and the 3rd ... well weird, as it's practically also using the 1 person as a road block thus being impermissible but it's pretty similar to the bystander example which she deems permissible.

Also in terms of the similarity, this paper offers an even more ridiculous scenario, arguing with a drawbridge after the single worker (before the loop) that is either open or closed. So that whether you'd use the single worker as a means or not would depend on whether the drawbridge is up or down. Indicating that "just a bit of track added doesn't really matter", is not actually a valid assertion.

She's also arguing that the loop case can be evaluated differently depending on the reasoning for choosing to divert. Where she essentially creates two agents Tili and Manuela who either want to save the 5 no matter what or act upon "not treating as a means" and so both would divert the bystander, differ on the footbridge and agree on the loop.

And because, according to her understanding of Kant, an action is permissible regardless of whether the reasoning is based in a permissible maxim, this means that because Manuela would be performing a permissible action so would Tili (in the loop case) also perform a permissible action, even though the maxim that she's acting upon would not be permissible.

That being said this only answers the question under the assumption that Thompson's assessment of permissibility of the scenario is correct. And that you can pretend not to use the group that you don't save not as a means to an end to save the other and that "using as a mean" only applies to directly instrumentally using them as a mean.

But you could just as well argue that any decision that you can make in these cases is using one or the other as a mean to the end of saving the other group and that only inaction would truly absolves you from doing that. However one could still argue that you're inaction was motivated by the fact that you wanted things to be how they currently were. Also it's not really a satisfying answer to advocate for inaction when you could have saved people.

But in that case each and every of these cases would have been impermissible and no modification would change anything about them.

  • I don't understand paragraph 5. it makes sense to say an action is impermissible, not a reason for doing something, which I agree on and find helpful, and may correct the idea that not fooling yourself (as in the loop but not the bystander) does not make something moral. but the reasoning in the paragraph is unclear. why does 'not treating as means' involve diverting in the loop scenario? does Thomson explicitly say "because it's indirect use" and if so then what difference does adding Manuela make?
    – user62233
    Aug 18, 2022 at 12:59
  • Her argument is that an impermissible reason doesn't make an action impermissible. So even though Tili's reasoning is impermissible, because she uses the single person as road block rather than treat them as person, the action itself is permissible. As Manuela who would act upon Kantian reasoning could come up with the same action. And she would come up with that same action because it's essentially the same as the bystander example where the single person is simply not part of Manuela's reasoning. Now I find that claim dubious already for the bystander, but Thomson is using that as well.
    – haxor789
    Aug 18, 2022 at 13:07
  • OK so then why do they disagree on the footbridge? and why doesn't Tili overule Manuela rather than vice versa? I may be able to work out why, let me think...
    – user62233
    Aug 18, 2022 at 13:09
  • Because Tili would use the single person merely as a means to an end and Manuela wouldn't. The original answer includes the link to the paper. Also what do you mean by overuse?
    – haxor789
    Aug 18, 2022 at 13:11
  • I just don't understand why Thomson introduces Tili, what difference it makes. I might read the paper.
    – user62233
    Aug 18, 2022 at 13:15

All forms of the trolley problem are deliberately constructed for a deceptive purpose. They are not intended to cause people to think in new and interesting ways about morality.

They are intended to condition people to drop context.

How can I say such an outrageous thing? Follow me.

Consider adding the following context. The reason the five guys can't get out of the way is because they are deliberately sticking their fingers in their ears, covering their eyes, going "La La La La" and ignoring that they could step off the track by taking two steps in almost any direction. The single guy on the other track is a worker who has carefully set the switch to assure no trolley will come while he works on the track.

In this context we would be apt to conclude the five miscreants had created their own fate and the single track worker should not be killed to save them.

Consider instead this alternate context. The reason the five guys can't get off the track is that they have been kidnaped, tied up, and left on the track. The kidnapper is the single guy on the other track.

In this context we would tend to think the kidnapper should get squished instead of his victims.

By adding one of these rather simple contexts we can easily move the trolley problem from one answer to the other. Without the context there is no "right" answer. Yet we are not permitted to add one of these contexts and still be in the trolley problem.

The trolley problem thus leaves us with no choice except to decide on the basis of numbers. Indeed, we are told we must drop the search for context in order to stay in the trolley problem. We are thus conditioned not only to accept dropping context. But to reject it if it is offered. "That's not the same problem!"

The purpose of the trolley problem is the conditioning thus produced.

The trolley problem is a form of "life boat ethics." That is, it is a grotesquely artificial problem involving a contrived crisis. There is no "good" solution where everybody comes to a good end, only sad endings. The same purpose is implicit in life boat problems. Somebody has to get eaten (thrown to the sharks, etc.) or everybody dies. However, all such contrived crisis situations produce flawed answers. The goal in a crisis is to return to normality. The rules in a crisis do not illuminate what should be done in a normal situation.

And yet, the trolley problem and life boat problems are quite popular. Their goal is to condition people to accept numbers instead of context. Five people are hungry while one has "more than he needs." Shouldn't we become cannibals? No thought about how those five became hungry, or how the one got his needs met. Five people want to go to university while one guy has a bank account. Shouldn't we become socialists? Again, no thought of how the five failed to afford school, or how the one got a bank account. The pattern conditions us to "eat the rich" because there are fewer of them.

I reject this pattern entirely. I deny that the trolley problem can teach us anything about morality. It can only condition us to drop context.

You must log in to answer this question.