Should You Kill the Backpacker?

I mentioned here that I was working on a new interactive activity at my Philosophy Experiments web site. Well, here it is:

Should You Kill the Backpacker?

It looks at some of the complications arising out of the Trolley Problem. More specifically, it largely relies on Judith Jarvis Thomson’s article, “The Trolley Problem”,  which appeared in The Yale Law Journal.

As yet, the activity hasn’t been subject to public scrutiny, so there are bound to be lacunae and errors in logic. Any feedback, therefore, would be much appreciated.

It’s probably worth mentioning that it’s sort of a f0llow up to another activity on the web site that deals with the same issues – but in a less sophisticated way – so if you haven’t yet seen that one, you might want to work through it first:

Should You Kill the Fat Man?

Leave a comment ?


  1. At the end of the test, it told me “You don’t seem to think there is a moral problem in treating a person solely as a means to an end”.

    That wasn’t the question I was asked earlier. I was asked whether I thought the question of treating a person as a means to an end was relevant to the scenario. I of course think there is a moral problem in treating a person solely as a means to an end. I just didn’t think it was the main factor at the point the question was asked.

  2. Hi – Sure, but that’s (partly) why the word “seems” is in there.

    Also, the question is slightly more subtle than you’re allowing here (for precisely the reason you give). It (also) asks whether the idea that it is wrong to treat a person solely as a means to an end adds weight to the judgement that it would be wrong to kill the backbacker for his organs.

    I think I’d want to argue that if one thinks the Kantian thing is relevant at all then it must “add weight”, even if, as you say, it’s not the main factor.

    But I will see if I can make this a bit clearer. Thanks!

  3. I have a hard time accepting Philippa Foot’s judgment that doing nothing in the trolley case is tantamount to killing, while doing nothing in the medical case is tantamount to letting die. If you do nothing in the trolley case, you are letting them die. Still, I think the distinction between killing and letting die is important to these moral scenarios. So my greatest problem when doing the survey was that I was unsure whether or not tension scores are calculated under the assumption that Foot’s analysis of the trolley example is apt.

  4. Hi Ben – Yes, that’s a fair point. I also find Philippa Foot’s judgement… suspect. Judith Jarvis Thomson sort of endorses it, though…

  5. s. wallerstein (aka amos)

    Doesn’t the fact that it’s a doctor in the backpacker example throw the analogy off a bit?

    First of all, doctors have a specific code, which does not permit cutting up backpackers for transplant organs.

    Second, one of the reasons that doctors have such a code is that it enables us to trust them to preserve our lives insofar as it is possible.

    The rest of us obey no such code, so it’s different for a mere non-medical civilian to sacrifice a life to save five lives.

  6. Hey Aka Amos

    I talk about that possible objection here:

    There might be something to it. But… I’m not sure!

  7. I don’t know if this has been discussed before, but I found the matter of time very important.
    The trolley problem happens suddenly, almost in a blink, there isn’t much one can do in such a short time, so you have to go with what you have.
    But a medical problem seems to offer more possibilities, since there is a bit more time available. One can hope for donated organs (for example from accidents or relatives), medicines and other things, even if the disease doesn’t allow much time compared to other diseases.
    Is there any discussion about this question?

  8. s. wallerstein (amos)

    How about this?

    Would you want to live in a society where doctors routinely use the organs of healthy, non-consenting patients for transplants in order to save a greater number of lives?

    I would say “no”.

    Would you want to live in a society where in the case of trains out of control they are routinely directed towards areas with fewer people in them?

    I would say “yes”.

    Now, it may be true that doctors more often find themselves in a situation where they can use the organs of a healthy patient without their consent for a transplant to save a greater number of lives than normal people or even railway workers find themselves in a situation where a out of control train is heading towards 5 persons.

    However, that doctors find themselves in said situation with more frequency than do railway workers seems an argument against letting them do it.

  9. Jeremy,
    The surgeon thought experiment needs work. It’s assumed the heart patient has at least one of every other of the five organs in good shape. Why not kill him (her) and transfer the good body parts to the other four as needed.
    Or take one of each of the body parts needed, except the heart, from the backpacker,distribute as needed and let the one with the heart problem die.
    Without some provision correcting for these possibilities it appears the surgeon is too dumb to be trusted and philosophers are not swift thinkers either.

  10. The logic in the bystander part is wrong. You’re saying if the person on the track is being used as a means to an end then the person shouldn’t be killed, therefore since the person is not being used as a means to an end it’s okay to kill the person. (If A then B therefore not A then not B.)

  11. Gra


    You’ll have to take up the surgeon thought experiment with Judith Jarvis Thomson! I’m staying true to the original.

    The bystander logic is fine. There is no such claim as you state it!

  12. Amos – That’s an interesting line of argument. I think I could have those intuitions on broadly utilitarian grounds (e.g., people would stop coming into hospitals if they thought there was a chance that their organs were going to be ripped out).

    Trouble is, I don’t think it really dissolves the puzzle. So, for example, if I restated the surgeon scenario in these terms:

    Surgeon can save lives by taking the organs of the backpacker; she is contractually bound only to help her patients, not the patients of any other doctor; we know for certain that this practice won’t become routine; is it morally permissible for her to go ahead?

    I reckon the vast majority of people would still say “no”.

    In other words, the puzzle still remains if one takes out all the consequentialist complication (which is at least in part what is interesting here).

  13. First, it currently claims that there is a tension or contradiction between answering the first two questions, “Yes.” But no such tension or contradiction exists. Someone could answer both questions, “Yes” with complete consistency simply by being a naive and relentless consequentialist who just counts bodies.

    I agree that there is a tension between answering both questions “Yes” and most people’s intuitions, but that isn’t what the set-up is purporting to measure. It purports to give you more red points every time it “detects a contradiction or tension in your responses” – I take that to be an internal contradiction or tension.

    Also, the following question is ambiguous: “Does the idea that it is wrong to treat a person solely as a means to an end add weight to the judgement that it would be wrong to kill the backbacker for his organs?”

    I could be a naive consequentialist and still answer that question “Yes.” The principle, if true (or binding, or whatever), does arguably add weight to the moral judgment that it would be wrong to kill the backpacker. I could think that, and on one interpretation of the question thus answer “Yes” … even though I think the principle is not actually true (or actually binding on me, or whatever status such principles are thought to have).

    Why doesn’t it just ask directly whether or not I accept the principle as true or binding, rather than whether the principle adds weight to a certain judgment that I might not accept?

  14. Thanks Russell.

    Your first point:

    A tension is explicitly not (necessarily) a contradiction or error. The analysis page notes this point, and then specifies the two ways in which the term is used:

    1) To designate a lack of fit between beliefs: so, for example, two beliefs will be identified as being in tension if some (fairly) sophisticated reasoning is required in order to reconcile them with each other.

    2) To describe a situation where a response has an implication that many people would find strange or unpalatable.

    So you’re right – of course – about the first 2 questions. But the activity adds a tension score for the out of kilter with most people’s intuitions bit (although the tensions are weighted, so you get the minimum possible “hit” in that situation).

    Your second point. Yes, you’re right. That needs rephrasing. Thanks. Useful.

    The reason the question isn’t more definitive has (had) to do with thoughts about moral pluralism. I wanted to allow that some people might think that the Kantian idea is important, that it has some force, but that it has to be held up against consequentialist considerations.

    (There is also the more prosaic reason that to make it definitive would make the programming much more complex, because it would require lining up the question with the way that people had previously answered. So, for example, if somebody had already responded that it was okay to harvest the backpacker’s organs, they’re not going to be impressed if the activity then asks them whether they think this is definitively ruled out on Kantian grounds – because obviously they don’t).

  15. Jeremy,
    The surgeon experiment: As stated it makes no practical sense. I just was pointing out using it in the form given is sloppy, though rather funny.

    As to the other:
    Am I mistaken in my reading of it, and I fully acknowledge it a possibility?
    Let A = the person on the tracks is being used as a means to an end.
    Let B = the switch should be thrown
    Did I misunderstand the next part?
    A is not true so it’s okay to throw the switch.
    If I got it wrong, I apologize.
    If not, then the logic being used is:
    If A then not B
    Not A therefore B.
    Logically this is incorrect.
    The closest that comes to this is the contra positive which is
    If A then B
    Not B then not A

  16. Hi Gra

    “Did I misunderstand the next part?
    A is not true so it’s okay to throw the switch.”

    No, that doesn’t follow. There may be other reasons why it isn’t okay to throw the switch.

    The activity never assumes that because a particular moral claim is denied (for example, that it is wrong to treat a person solely as a means to an end) that it follows that it would then be okay to throw the switch.

  17. What doesn’t seem to be taken into account is that the trolley is already moving, and (I assume) cannot be stopped. So action must be taken, or the five men will die.

    The backpacker has a right to continue using his organs until he’s finished with them, regardless of the needs of the other patients.

    The issues are more, and more subtle, than the test recognises.

  18. Okay, Jeremy, thanks for that explanation. I guess I have a further worry, then, that the word “tension” is being used to cover two different things that can come apart. One is a sort of prima facie internal inconsistency, while the other is inconsistency with common intuitions.

    When we teach trolley problems to undergrads, often one of the first things that we want to bring out is how easily these do, in fact, come apart. E.g. a “Yes” “Yes” answer to the first two questions seems, prima facie, quite consistent, but the second “Yes” is counterintuitive. A “Yes” “No” answer looks at least superficially inconsistent (some explanation is needed as to why it is not “really” so…), though consistent with most people’s intuitions.

    And once we get to this point, I’m not sure what the “right” way is to quantify and weight those two things so as to place them on a single scale. Or whether there can be a right way.

  19. I don’t think the fat man on the rail-road and the backpacker is the same scenario.
    The fat man is already on the tracks and therefore his life is already implicated in the same way as the other men in the scenario.

    In the backpacker situation, you have to implicate the backpacker’s life in a choice which otherwise he has nothing to do with.

    I think the equivalent would be if the fat man were standing next to the track and you push him in front of the trolly to stop it and save the five men.

    The trolly situation is a straight either or, but the organ one isn’t.

    You could change the organ one into either or by having the 5 dying patients that need all different organs and a dying backpacker that needs all new organs and a donor coming up with all the organs needed. Then the choice is, either save the 5 people with the different organs or just the backpackers life with all of them.

  20. @Russell – Thanks. That’s an interesting point. I could actually pull the two things apart for this activity.

    There’s always something of a trade-off with these activities. On the one hand, there’s the populist impulse to have a single headline score or “diagnosis” or something. On the other hand, there’s the desire to remain at least somewhat true to the philosophy. (Plus for purely pragmatic reasons it’s necessary to keep the decision-trees, etc., under some sort of control, otherwise the programming becomes nightmarishly complicated.)

    But in this instance there are – as you say – good in principle reasons for keeping them separate (since part of the point of the Trolley Problem is precisely that one can quite consistently come to highly counterintuitive conclusions). I’ll have a ponder.

    @George – I think I’m right in saying that the issue of whether people are somehow already implicated in the situation has been debated in the literature. I’ve got to say I’m not entirely convinced it makes a moral difference.

  21. It is never ‘right’ to kill another innocent human being – except to save your own life (otherwise we’re judging people’s lives and the only one we can place a value on is our own), unless a person has given prior consent (explicit or implicit).

    This allows for a government to make policy decisions which will cause harm and/or deaths for certain people over others (e.g. healthcare) but does not allow for the average person to do so, except to stop someone actively causing harm.

    If we accept this (tortured libertarian) position then the standard moral intuitions can be followed: the worker on the line knew and accepted the risks and is gladly(-ish) accepting that the majority be saved; the fat man has made no such agreement; neither has the backpacker.

    This may simply be me retrofitting my intuitions into some form of consistent ‘logic’ since I fear that the backpacker and worker are effectively the same scenario, but am struggling to imagine that I wouldn’t throw the switch, while knowing I wouldn’t kill the backpacker (unless I was needing an organ myself).

  22. Jeremy, one thing I really dislike about the question is that it states: Is it morally permissible…?

    I personally think that it is morally permissible for Pacey to turn the trolley AND to NOT turn the trolley. This makes answering the question that little bit more difficult than it should be – i.e. it doesn’t ask what I think I would actually do, merely which of Pacey’s options I think are morally permissible when I think both of them are.

    Also, if I answer in the perfect libertarian mode (cause no direct harm to anyone) then it adds tension to my score when there is no tension because I have been perfectly consistent and answered No to every question, grrr…

  23. @Keddaw – The “Is it morally permissible…” thing is taken straight from Jarvis Thomson’s article. The idea is that it allows people to think that although it’s okay to turn the trolley, it’s not obligatory.

    We’re definitely not interested in what people would actually do here, since that would allow in extraneous factors (i.e., those that have nothing to do with whether an action is right or wrong). For example, they might not turn the trolley because they’re squeamish, paralysed with indecision, etc.

    You might want to check out:

    In the Face of Death

    This looks at a couple of real-world examples to explore in what sort of circumstances we might think it right to take a life.

  24. A key part that is missing is the degree of certainty.

    In the classic trolley situation, there is a binary choice, with no vagaries involved. Either the group of 5 all die, or the single person dies. There are no other options, and one of the choices must be taken.

    In the case of the backpacker, there are a number of uncertainties involved. First – will the operation even be successful? Will all five transplants take, and give back full quality of life? From a strictly utilitarian viewpoint, this risk brings the judgement into question.

    (For a similar dilemma – imagine that the trolley loop only goes one way – e.g. if you run over all five, the train doesn’t enter the loop, but would go straight ahead, but if you go over the fat man, the train tracks have to loop back over the five. This makes the decision to hit the fat man – an action that, in reality, only _might_ stop the train – a more dubious choice)

    Second – is the backpacker even the only choice? Given the size of the population, other donors should be available. Why is this donor so special? If there are other choices, and there is time to spare, then there are chances that another donor could be found who has already died.

    Third – even if you accept that killing the backpacker to save five is acceptable, there is another solution to the problem that doesn’t kill the backpacker and still saves four lives: take the organs of the first patient to die and use them to save the other four patients. This gives you the same net benefit – one dead, five alive – and doesn’t require you to choose the victim.

    The other key part of the morality choice here is about implication. As soon as you add a person to the pool of victims intentionally, you are acting immorally. In the trolley situation, you have no control over the people on the track – your choice of which track to take doesn’t add a participant. Throwing the fat spectator over the rail (or taking a backpacker apart for involuntary organ donation) does.

    Certainty and the arrogance to choose on the behalf of others – that’s the flaw in these moral choices.

  25. @Robert – I’m just not at all convinced that the degree of certainty thing is the issue here. I (strongly) suspect that if one explicitly dealt with all those issues in the thought experiment (i.e., dissolved them) – and in fact a lot of them aren’t there if you take the backpacker scenario at face value – you’d still get different intuitions about the backpacker and trolley situations.

    I think the second point you make – about implication – is stronger (and, as I said earlier, it’s been debated in the literature, I think).

  26. Jeremy,

    The certainty factor is a big issue. Take the trolley example: in the scenario where the train will loop back over the five men, the only aspect that makes the choice to go over the single fat man potentially moral is the “certainty” that the fat man will stop the train. Otherwise all you do is risk six lives instead of five – and suddenly the option of letting the train go ahead, killing the five but not heading down the loop to kill the sixth person, becomes a moral option.

    Yes. when you dissolve the experiment, you strip the uncertainty away. But as soon as you put the words back around, you re-introduce the variable. Furthermore, putting different words around the problem – with different uncertainties in the resulting scenario – can change the answer.

  27. @Robert – We’ll just have to agree to disagree. I don’t think people’s intuitions are going to be different in the presence or absence of certainty. In other words, my view is that if people genuinely have no doubt that the thought experiment accurately describes the world – and that, as described, there is no uncertainty here – you’ll still get the same result. Most people will think it okay to turn the trolley (in the loop back scenario). Most people will think it not okay to operate.

    Also, you’ve got to remember here, there’s no requirement that thought experiments should describe an actual world (consider, Judith Jarvis Thomson’s famous violinist thought experiment, for example). All that is required is that they are logically consistent, that one accepts their terms, and then responds accordingly. In other words, if the thought experiment asserts there is certainty, then, in its own terms, there is certainty, and that’s how you should think about the issue.

    It’s true that some philosophers think that the kinds of extraneous variables you’re talking about will tend to come into play, but I see no reason to suppose that even if this is true in this case, it’s going to make much difference to how people respond here.

  28. s. wallerstein (amos)

    Perhaps it’s my particular problem, but the idea of a doctor
    has so many associations for me that I cannot see an “abstract” doctor, while for example, I can see an “abstract” trolley or an “abstract” backpacker.

    It horrifies me (and I suspect that it horrifies others) to think of a doctor cutting up a healthy person to use his or her organs to save the life of five others.

    Otherwise, I don’t have problems with the idea of sacrificing one person to save the life of five others.

    Thus, for me at least using a doctor in the example makes it difficult to answer the questions
    with the “objectivity”.

  29. Hey Amos

    I don’t think it’s a problem for the thought experiment that you’re horrified by the idea of a doctor cutting up people. I think it becomes a problem if you have good, consistent reasons for being horrified by the doctor’s actions, yet not horrified by the idea of turning the trolley.

    In other words, the worry here is that there is something about the patient/doctor relationship that means that one can respond differently to the two scenario on the basis of a single moral principle (e.g., some variety of utilitarianism.)

  30. “We’re definitely not interested in what people would actually do here”

    Fair enough. There still exists the problem that I think both options are perfectly morally permissible and wouldn’t want to judge the decision someone made in that terrible situation as ‘right’ or ‘wrong’, praiseworthy or blameworthy. I may well think that one is somehow ‘better’ than the other but that is a personal opinion and is based on certain values and traits I hold dear that I know not everyone will share and do not think they necessarily should share – hence both possibilities are morally permissible.

  31. The backpacker and loopback trolley are not morally equivalent and I can’t see why the author of the survey thinks that.

    The doctor in the backpacker scenario is watching five people die because she has no means to save them. When she chooses to harvest the backpackers organs and save the five, she has changed the scenario and inserted the backpacker into the equation. Neither chance nor the backpacker put the backpacker into the equation.

    In the trolley loopback scenario, you don’t choose to put the fat man or the five men on the track. When you come into the scenario there is only a binary choice of kill five or kill one. You don’t have to make a decision to put the fat man on the track because chance has already decided for you. You are not choosing to make the fat man a means to an end. Presumably he himself choose to be on the track. It’s like a man choosing to be a soldier. He has chosen to put himself in a scenario where someone else’s decisions may or may not get him killed. His commanders must consider him in the context of a larger group.

    The modified loopback scenario where you must actively throw the fat man onto the track is more analogous. Neither chance nor the fat man chooses to put the him on the track. If you throw him on the track then you have chosen to change the scope of the scenario, to change the moral question at hand. This is like drafting a man into the army for war. You’ve taken away his moral agency. He is no longer making the choices that lead to his death.

    Whether the fat man on the track knew that being on the track could kill him is not relevant. He put himself on the track and is therefore responsible for events that follow from that choice. If someone throws him onto the track, he has not made a moral choice.

    That’s the difference. The doctor makes a moral choice that the backpacker has no say in. The fat man on the track did make a choice (even if he couldn’t have known the result).

    If the backpacker had chosen to get in a car with a drunk man, he might not have expected it to end up with his organs being harvested, but that’s something that happens all the time and we don’t think of that as using a person as an end for a means. Why? Because we didn’t put him in the car with a drunk. He did. Or if he was merely walking down the street and got him by a drunk driver. We didn’t drive the car into him. He choose to be on the street and driver hit him. We correctly judge the drunk driver to be morally responsible and not ourselves, and we harvest his organs.

    I don’t know why the scenarios pretend that a persons own moral agency is not part of the equation. It creates a false equivalence.

  32. @James – Well, the author of the survey (i.e., me!) is merely following the line taken in the literature by philosophers such as Judith Jarvis Thomson.

    A lot of what you’re saying here just isn’t right. For example:

    “You are not choosing to make the fat man a means to an end.”

    Yes, you are – choosing to turn the trolley is a necessary condition of the fat man in the loop back scenario becoming a means to an end. Without doing that the fat man is just a guy on a track.

    “He is no longer making the choices that lead to his death.”

    Two points here:

    1. The fat man on the bridge certainly made choices to put himself on the bridge (just as the fat man on the track made choices that put him on the track).

    2. The fat man on the track has his agency taken away (as you put it), the moment you turn the trolley.

    “He put himself on the track and is therefore responsible for events that follow from that choice.”

    Again, two points:

    1. The fat man on the bridge also made the choice to put himself on the bridge (the backpacker made the choice to put himself in the hospital), and is therefore (by your lights) responsible…

    2. In neither trolley scenario do the events follow inexorably as a result of the choices made by either fat men. As I pointed out earlier, without the trolley being turned, nothing happens to the guy on the track. He isn’t responsible – in any way – for the trolley being turned.

    The basic point here is that you need to explain why any differences in terms of “agency” between the scenarios are morally relevant. Nothing you’ve said is convincing in this respect (though there might be some mileage in the idea that the fellow on the track has contracted in to a certain amount risk – albeit it’s hard to believe he’s contracted in to getting run down by a trolley).

  33. @Jeremy Stangroom
    The backpacker made a choice that led to meeting the doctor, but the doctor makes the choice to kill him to save five.

    The fat man chooses to be on the track (the loopback track for this discussion) and you choose to throw the switch and kill him to save five.

    They seem to be almost identical, but I’m sure that nearly everyone is horrified by the first but more ambivalent about the second. Why?

    I think that most people have an instinctual sense that the doctor is creating rather than responding to a scenario, choosing to make it about six people rather than five.

    Chance and happenstance put the fat man on the track. The bystander has no control and therefore no moral responsibility for him being there. The doctor is throwing his/her backpacker onto the track. The weight of the decision now lies on the doctor. The doctor put his/her thumb on the scale.

    Whether the backpacker or the fat man know what the results of their acts are isn’t relevant. Acts have consequences whether we know what they will be or not. If I perform act of my own volition, then who is responsible but me?

    If the fat man puts himself on the track, he doesn’t ‘choose’ to get run down, but he made an act that could lead to that result. The bystander chooses to share in the moral responsibility for that act by throwing the switch.

    The backpacker makes an act that puts him the sights of the doctor. Now, I admit that although the backpacker may not expect it, it’s possible his doctor will be a murderer so it follows from his act that he may be murdered. But how much moral responsibility does he have for getting murdered and how much responsibility does the doctor take on?

    I say the bystander takes on less responsibility than the doctor and the fat man bears more than the backpacker.

    I think that for the two scenarios to be equal then there must be an assumption that every act must be considered with regards to its result on every possible person. But that’s not what is presented. Nobody is considering the family of the fat man or the backpacker. Nobody is considering what the backpacker or the fat man will contribute to society. If killing the fat man and the backpacker are equally using a person as a means to an end, then why exactly are we limiting the end to those six people? Why not limit it to just immediate family? Or to all their friends and acquaintances? What is the unspoken principal at work that says just these five or six people need to be considered?

    I don’t think it’s possible to consider all possible ends to every act, so we never really know what the ends of our means are. We are forced to limit the scope of our consideration. I’m claiming that the bystander and the doctor have made different choices in defining the scope of their consideration.

    The bystander chooses to put the fat man within the scope of the problem, but crucially, the bystander has no option to replace the fat man with another fat man close at hand. The bystander can expand the scope from five to six people, but he can only expand it to the five thin men and that one fat man. The fat man already has one foot in the equation. If the scope expands then he is the closest at hand and the only choice.

    The doctor on the other hand, has a free hand to set the scope of her murderous plan. Presumably the doctor could consider other candidates. The doctor could consider volunteers, or accident victims, or those near death, for instance. The backpacker is an arbitrary choice of the doctor.

    I think this arbitrariness is what horrifies people and makes it different. The bystander is constrained by circumstance, like someone with a gun to their head, and the doctor is not. The bystander has not claimed the right to kill anyone, anywhere if they think it’s for the best. But that’s what the doctor is doing.

    (Please don’t try to argue that the scenario is that the doctor has only one potential victim. That’s like taking your ball and going home.)

  34. @James – You win. That’s way too long to reply to! 🙂

  35. Victory is mine, and yet the spoils of war taste so bitter.

  36. Jeremy,
    I went back to the page in the questionnaire in which being used for an end is introduced. You have, there:
    “According to some philosophers, . . . it is morally permissible to turn the trolley (in both scenarios), since a good outcome here does not depend on the presence of the single workman on the track.”

    I think it’s clear that the philosophers mentioned are in fact using (in conjunction with other facts or not) the logic:
    If A then not B
    Not A therefore B.
    Let A = the person on the tracks is being used as a means to an end.
    Let B = the switch should be thrown
    And which I stated above is wrong

    This same question was brought up by you in your last post, and at that time you admitted “It is a disturbing thought. I’m not sure I buy the argument.”

  37. Hi Gra

    Sorry, but that doesn’t follow at all.

    Some philosophers think that it is morally permissible to turn the trolley in both scenarios. But this isn’t based on them accepting a logical error!

    They’ll only think:

    If A, Not B.

    Not A, therefore, B not ruled out on the grounds of A.

    Your ellipses does violence to the meaning of the bit you quote. It refers to the difference between the scenarios (i.e., why for some philosophers it might be permissible to turn the trolley, but not to harvest the organs). However, it’s not functioning as a complete explanation of why it might be permissible in the first place: you’d have to add in a whole lot of other stuff for that to be the case (i.e., all the stuff that means that one can’t simply go “Not A, Therefore B”).

    Also, remember that “morally permissible” here doesn’t mean people “should” throw the switch, anyway. There’s no obligation. It only means it’s allowed.

  38. Jeremy,
    Thanks for the effort to get me to understand. I’m afraid with this last effort on your part I’ve gotten real lost.

  39. Re: Keddaw
    I also have difficulty with the question of moral permissibility. The first question seems to trick the reader. The question does not ask what action to do; the question asks if this is morally permissible. If one assumes that there is only one morally permissible action, then the question can be interpreted as asking which course of action to take. In a dilemma, there are sometimes many different courses of action that can be taken, and none may be right. The trolley problem might qualify as a dilemma. So, it is unfair to ask if the solution to a dilemma is morally permissible, and then criticize the reader for answering yes or no. That is, all answers to a dilemma could be morally permissible. However, the problems are also probably trying to test the reader for tension and moral consistency.

    If the trolley problem is not intended to be a dilemma, then it should state so. But then if it was not a dilemma, then there would be one correct answer. The trolley problem is what I have previously described as an incomplete dilemma as there are only two possible actions: (1) turn the switch or, (2) do nothing. Three criteria are required to solve this dilemma. If no criteria are offered, except post-suggestions and comments based on utilitarian judgement, then the problem is unsolvable.

    Re: S. Wallerstein
    I agree. This is an interesting problem in professional ethics. There is a long standing principle in Hippocratic medicine that a patient goes to a doctor to be healed and not to be killed. The backpacker has gone to the hospital knowing the ethical principle of permission (or exculpation) is enforced. The medical profession is obliged to follow this professional standard. There are those who may offer a utilitarian argument to suggest a maximization of life. However, I am not a utilitarian, and this is one specific example that supports my difference with utilitarians. A solid supporter of this theory and practice is Arthur L. Caplan from the Center for Bioethics at the University of Pennsylvania.

    I like how permission is introduced into the problems. I am not sure what percentage of people even understand what an ethical principle of permission is. “Permission” seems difficult to establish across all facets of life, so I have generally restricted its principles to professional ethics.

    Re: Gra
    Is there not another condition?
    C = the switch should not be thrown
    Thus, “if A not B” is in an incomplete logical deduction. That is, the condition C is not the inverse of B. C brings about a completely different result.

  40. Dennis,
    I’m not sure what you’re saying.
    If A not B is short hand for: If A is true then B is not true, e.g. If this is Oz then it’s not Kansas.
    Can you give me a specific example in which you want to use C?

  41. I scored 38% tension here. Morally permissible, is not always a good bedfellow with on the spot decisions; e.g.“what the hell should I do? I must decide quickly.” Thought experiments of this nature often ignore, and are isolated from so many other factors, which are manifested in real life situations. They seem to assume that all persons are alike in sex, innate personality traits, education, state of health, and so on. Additionally there are those who are reluctant to intervene in some states of affairs which are not of their making. For instance were I the passenger on the trolley, which is running out of control, I may elect to go for the single person on the grounds that his resistance to the trolley would be less than five persons. Impact with five persons and whatever equipment they have with them I would reason are sufficient to dis rail the trolley send it spinning with me as sole occupant. Hopefully one person would be less hazardous. Another point occurs suppose having decided to take the utilitarian one person option I suddenly see the that person is my mother or father would I not, assuming there is time, quickly change my mind?
    During the course of the test I was asked “would not be permissible for the surgeon to harvest the young backpacker’s organs. After all, by doing so she would prevent the deaths of five people in a situation that seems – at first sight, at least – to be structurally identical to the bystander scenario. “
    I do not think there are sufficient points in common with the two scenarios for them to be the subjects of analogous reasoning.
    1/ We have people in the case of the trolley, in state of affairs not of their making, who are suddenly and innocently plunged into a situation with a runaway trolley.
    2/The surgeon is a professional person who deals with matters of life and death on a daily basis. The moral constraints on him are modified by ethical practice in his profession. His job is to save life not kill it, Within the limits of his normal professional life killing does not justify saving (this may not hold on say a battlefield). He also has time to think, unlike the those in the trolley scenario. Additionally he has also had the opportunity to ask the backpacker if he can kill him and been refused. There is not time to discuss things in the Trolley scenario. Suppose he does successfully kill to save he may acquire a taste for it and continue playing God as he pleases. What use is he then as a surgeon?
    The Trolley problem has led to much interesting discussion, talk, lectures, learned papers over the years but it is so far as I can see, tightly bound by the few initial assumptions it is insoluble, I believe, because of the scarcity of other information. It reminds me of the Theory which does not work in Practice and that, as I think Schopenhauer pointed out, means there is something wrong with the theory.
    That said the test does show some interesting statistics. Males and Females are about the same measure of tension. How interesting that Muslims are far and away greater in tension than other religions, I am wondering why that is. In the loop scenario I don’t think it is made clear if the bystander knows or does not know there is a loop. Maybe I have missed something here.

  42. @Don Bird
    Yes! The bystander is morally constrained in a way the doctor is not. It’s a crucial difference. The bystander can save five, but only by killing one specific fat man. The doctor could choose anyone, even a volunteer. It doesn’t really follow from the trolley case that the bystander always thinks the ends justify the means. The doctor, however, claims a general right to kill anyone if he/she thinks its for the best. That’s much more terrible.

  43. I’m surprised by all the different explanations people are coming up with to explain this. Hasn’t research on this issue already shown – or at least strongly suggested – that humans avoid killing innocent people when doing so feels a little too personal? As in, the workmen working on the tracks are just numbers, but the fat man you would have to push onto the tracks is a person. You can feel him resist you as you struggle to push him off the bridge. You hear his cries, you see his expression of fear and anger. You know he’s thinking “How dare you decide to end my life to save the people in the trolley!”

    That shit’s not cool.

    But when you’re off somewhere in the distance throwing a switch? Not so big a deal.

    Of course, it could be a big deal. I imagine myself as a worker for the rail company. I’m standing safely in a monitoring office built into the side of the valley where the lone workman is working on the tracks. I can see him through a window, and communicate with him via speaker. It’s as if we’re standing side by side – the only difference being that if a train came down the track at that moment, it would only kill the worker, and not me. (Let’s also say that there is no entrance to the office from the track.)

    I imagine that a call comes in over my radio, informing me of the situation with the train. The workman hears this as well. Let’s say the controls for switching the track are located (solely) in my little office. I have a choice to make. The lone worker standing just five feet from me (on the track) can see the wheels turning in my head. “You’re not going to switch the tracks and kill me, are you?”, he says. “I don’t want to die!” I don’t know this guy at all, but that sort of pleading is hard to argue with.

    Whereas in the “impersonal” scenario I might have agreed to reroute the train, I think there’s a good possibility that in this scenario, I wouldn’t be able to do it. Not with the man whose life I would be taking standing right there asking me not to.

    Of course, things get more interesting when I personally know some of the workers involved. I can imagine myself letting five “numbers” die for the sake of one “real person.” But what if I know the other five? What if everyone involved is a person to me? In that case, would raw numbers make the difference? They might.

  44. Guys

    I’m still reading this stuff, but there’s no way I can reply to everything. Just too much!

    @Don – Thanks for the thing about it maybe not being clear that bystander in the loop scenario knows about the loop. I’ll add that in.

  45. Re: GRA
    “Can you give me a specific example in which you want to use C?”
    I could probably not give an example within the bounds of the trolley quiz here. My position is that the question is a dilemma, and therefore unsolvable. To solve the problem and consider the option of B or C, I would require additional criteria. An example of additional criteria might be empathy for someone who is on the track, such as my mother or father.

    However, if I am not mistaken, you are confusing options and criteria, or in the case of the wizard, places and actions. Oz and Kansas are places (A). Following the yellow brick road (B), or killing the witch (C) are actions.

  46. Dennis,
    I’m afraid my experience in logic is not enough to help you out.

  47. Jeremy,
    In your last response to me you said: “Your ellipses does violence to the meaning of the bit you quote. “ I’m sure you realize I wasn’t trying to be cunning by the way I used the quote. In any case, I reread the full selection with the original material replacing the ellipses, and if anything I feel it strengthens my argument. The reconstituted passage emphasizes that the surgeon is not permitted, morally, to do the operation on the backpacker because he is a means to an end, and that it is permissible for the bystander to throw the switch because the solitary workman is not. Now, I think there are a lot of people who like myself, a philosophically naïve smarty-pants ,who would take that paragraph at face value, i.e. assuming nothing about the philosophers mentioned except as presented in the paragraph. If you feel this is unreasonable, then I don’t think we can come to any agreement on this issue. However, if you find what I wrote has some merit, then I say the natural implication is that permissibility is necessary and sufficient if the person involved is not used as a means to an end. And that’s wrong.
    Jeremy, I really do appreciate your looking at this stuff. I’m sure you don’t realize how much, specifically I, and more generally, many of us get from participating in these discussions.

  48. Gra

    I’ll try one more time here. (Probably this is my fault for not being clear enough in my explanations!)

    “The reconstituted passage emphasizes that the surgeon is not permitted, morally, to do the operation on the backpacker because he is a means to an end, and that it is permissible for the bystander to throw the switch because the solitary workman is not.”

    Yes, that’s absolutely true – but it’s referring only to the moral difference between the two situations (and this is crucial).

    Here’s an analogy:

    You’ve got two plots of land. Everything is absolutely identical, except that one gets sunlight, the other doesn’t. Crops grow on the one with sunlight; not on the one without.

    In that situation, one might say when talking about why the crops grow on the one field, but not on the other, that the presence of sunlight is crucial. It’s the key difference.

    But… and this is the absolutely fundamental point:

    It doesn’t follow, generally, as a matter of logic, that crops will necessarily grow whenever and wherever there is sunlight. (They also need water, nutrients, etc). It happens all these things are already present in both the fields we’re talking about, which is why we’re able to isolate the sunlight as the key difference. But that’s the only reason.

    Or to put this another way, sunlight is a necessary, but not sufficient, condition of the crops growing. (It just happens that all the other necessary conditions are in place in the two fields.)

    It’s the same with the “treating people purely as a means to an end” thing.

    For some philosophers, ensuring that you don’t treat people purely as a means to an end is a necessary condition of acting morally. But you can’t assume that it is also a necessary and sufficient condition of acting morally. In other words, it’s entirely possible to think that other factors might also be in play.

    Okay, so why you’re going astray with the paragraph you’re talking about is that you’re failing to recognise that the reason we can talk in that situation as if all we have to worry about is ensuring that we don’t treat the single workman on the track purely as a means to an end is because it’s being assumed that all the other stuff required for moral behaviour (i.e., all the other necessary conditions of moral behaviour) is already in place in both scenarios (for example, that we have good reason to suppose we’re going to save more lives rather than fewer lives by turning the trolley).

    Now, if that’s true (if the other stuff is already in place in both situations), and only if that’s true, then it is true that in that very specific situation ensuring you don’t treat somebody purely as a means to an end is a necessary and sufficient condition of you acting morally (but only in that situation).

    However, to emphasize, this does not require that the philosophers are committed to the view that a necessary and sufficient condition of acting morally, generally, is just that we ensure we don’t treat people purely as a means to an end (because you also have to factor in all the moral calculus stuff that happens already to be present in the trolley scenarios – and possibly other stuff as well).

  49. Jeremy,
    Thanks loads. It looks as though you did a bank-up job on the explanation. I still have to take it apart and put it together again, but one way or another this suffices.

  50. Jeremy,
    I’m going to offer a proof. If you can tell me what’s wrong with it, I’ll happily concede I was wrong all along and beg your forgiveness, or just get lost, whichever you prefer.

    At what point was Kant’s principle ‘not using a person as a means to an end’ (we’ll call it KP) introduced? It was at a point of whether or not throwing the switch (we‘ll call it TS) was permissible, otherwise there was no point in looking at KP. That is,there was no definitive decision reached on TS. But KP is not applicable since no person was being used as a means to an end. Therefore, we should still be with TS permissibility undecided. But according to the paragraph in question TS is definitely permissible which means KP was improperly used.

  51. “But according to the paragraph in question TS is definitely permissible which means KP was improperly used.”

    Gra – I really think I’m done with this now!

    But very quickly:

    a) The paragraph *doesn’t* say TS is definitely permissible. It says some philosophers think it’s definitely permissible;


    b) It doesn’t say they think it’s definitely permissible (just) because turning the trolley doesn’t involve treating the single workman on the track as a means to an end. It says that some philosophers think it’s definitely permissible because *a good outcome* doesn’t depend on the presence of the single workman on the track (i.e., treating the workman solely as a means to an end). And, of course, the reason why it uses the phrase “good outcome” is precisely because it’s supposed to invoke all the moral complexity involved in the situation that goes beyond the simple Kantian thought (moral complexity that is seemingly held constant between the backpacker and trolley scenarios, thereby allowing us to talk of the Kantian thing being the key difference).

    That’s it, I’m done now! 🙂

  52. Jeremy,
    I truly apologize for dragging you into this as much as I have. You’ve shown outstanding patience. For me, it was so nice to think I understood what the issue was; I didn’t want to let it go.
    Thanks for the effort. My reasoning power is no where near as great as I think is needed to follow your explanation. Maybe I’ll give it another try later on.

  53. Gra – Absolutely no need to apologize! It’s just I don’t think I can get any clearer – which no doubt is a limitation of mine – so I’ve said my last on the issue. 🙂

  54. It seems to me that the difference between using someone as ‘a means to an end’ and ‘incidentally’ (but not accidentally!) killing someone is an emotional difference rather than a logical one.

    Neurological studies strongly suggest that direct, close-up harm and more dispassionate harm-at-a-distance activate very different centres within the brain that lead to the logical gymnastics we all try to perform in this scenario (in the beginning at least) as we try to make our natural reactions fit in with our supposedly reasoned and consistent morality.

    This is why I am tending towards the opinion that TS is equally as wrong, or right, as killing the backpacker. It just so happens that we can morally justify one using a modern part of the brain (CBA analysis in TS) and find the other abhorrent (due to our primitive wish not to harm our immediate social group).

    In reality we react to situations as they occur and I would very likely TS (even if it using someone as a means to an end) and almost definitely never push the fat man or kill the backpacker. Which I fully realise is inconsistent and not, in these very strictly defined scenarios, for the greater good. Sobeit.

    Incidentally, I think there is a massive difference between letting harm come to someone and actively inflicting that harm. Thankfully so does the law.

  55. Jim, I think the law is being pragmatic rather than moral/ethical/whatever. If I see someone eating something I believe will make them sick I CAN stop them but ultimately I do not know all the details so it is none of my business; whereas if I actively poison the food then I DO know all the details and it is entirely my fault.

  56. I had no tension until the very end, when it said because I didn’t accept either of the two moral reasonings critical to the process. I’ve skimmed comments but not read them all, so apologies if someone has come up with this before me. 🙂

    I saved the 5 people both times, and didn’t kill the backpacker. While I can’t point to a specific philosophical doctrine, I’ll outline my reasoning.

    In the trolley cases, it is clear that either 5 people or 1 person is going to die. it is imminent and inevitable. in the case of the backpacker, 5 people may die, or you can choose to kill one person, who entered into the doctor’s office under a the assumption that he would be dealt with according to a social contract of client/doctor relations and the hippocratic oath. That to me was the main distinction.

    There are also nagging doubts… can these organs not be gotten elsewhere? what if this sets a precedent that anyone entering a hospital may be killed so their organs could save the lives of two others? what if the opportunity cost of doing 5 transplants is larger than some other method of vaccination/treatment for others, why then kill this backpacker on purely utilitarian grounds.

    I do enjoy the issue that someone else brought up: harvest the organs of the heart (or other) patient. Assuming all of them have perfectly fine organs aside from their defective one, ask for a volunteer or even roll a dice. Then it’s having 5 people die, or having 1 person die that would have already and saving 4 others… which is somewhat unsettling if no one volunteers to do so, but is IMHO a simpler choice than harvesting a healthy person who put himself under the doctor’s power.

    One could also state that the rail workers knew there was some risk involved when they agreed to do their job, particularly in a place with no escape from the tracks. The backpacker in going to the doctor would have had no reason to believe that he would be killed for parts as it was a routine checkup and that is a decidedly non-routine practice.

    While I do generally believe in kant’s second categorical imperative (naively perhaps?) it wasn’t critical to my reasoning here. I am, as you can tell, somewhat of a contractarian. 🙂

  57. PS – having some type of threading to the comments system here seems like a good idea.

  58. to clarify my third paragraph a bit (I blame the small comment writing field / multitasking):

    In the trolley cases, it is clear that either 5 people or 1 person is going to die. It is imminent and inevitable. In the case of the backpacker, 5 people may die, or you can choose to kill one person who entered into the doctor’s office under the assumption that he would be dealt with according to a social contract of client/doctor relations and the hippocratic oath. The backpacker was not in a situation involving risk until he chose to put himself under the care of the doctor, and it feels like a gross abuse of that trust/agreement to take his life to save others. That to me was the main distinction, and I don’t felt it was adequately covered by killing vs. letting die or ends unto themselves.

    It might help to explicitly specify that this was a remote location / transportation failures / etc and there were no other sources of organs available in the time period before the 5 patients needing transplants would die. I suppose this is implied, but it might help clarify the urgency in killing the backpacker.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>