The Trolley Problem and Why I Wouldn’t Pull The Switch

Its been almost fifty years since philosopher Phillipa Foot first started thinking of runaway trolley-cars running over workmen. Today, decades later, the trolley problem is a cultural phenomenon that’s made it’s way into school lectures, office small-talk and dinner conversations all over the world.

So why write this article then, if the topic in question has already been so extensively covered, summarized and investigated? Well, its mainly because of the following observations:

1. People tend to believe you can only either be a utilitarianist (someone who wants to maximize utility) or a deontologist (someone who wants to follow a strict set of moral rules) when it comes to the trolley problem.
2. An overwhelming majority look towards utilitarianism for an answer, which is something I disagree with.
3. The wealth of information that exists on approaches to the problem are usually hidden in esoteric philosophy papers that are completely inaccessible to a general audience.

As someone who disagrees with your standard utilitarian answer of sacrificing a few to save many, it’s hard to even rationalize why I disagree. The Internet says if I’m not a utilitarianist, then I must be a deontologist. I’m sorry, but seeing as how I’m not trained in philosophy, this is a very confusing thing to find out about yourself.

I can’t tell you how many times I’ve read that deontologists want to abide by a strict set of moral rules. But what does that even mean? Who doesn’t have moral rules? Do utilitarianists not have rules? Telling someone they’re a deontologist without actually explaining what that means in a manner they understand is pointless.

With that in mind, my goal is to offer a detailed explanation for why some people, like myself, choose to disagree with utilitarianism in the context of the trolley problem. But more importantly, I want to explain it in a way that anyone could understand. In doing so, I will have to reference some philosophical groundwork laid out by many people, chiefly Phillipa Foot, Judith Thomson, Thomas Nadelhoffer and Frances Kamm, without whom I would have had no idea about how to even piece together a proper answer to this timeless debate.

Driver In Trolley (DIT) Scenario

Before we ever had to consider the idea of bystanders and switches, we had a simpler scenario, created by Phillipa Foot, one of the founders of contemporary virtue ethics, with a simple premise.

Does the driver continue down the trolley tracks and kill five, or steer left and kill one instead?

In this Driver In Trolley (DIT) scenario, a runaway trolley car is hurtling down a set of trolley tracks towards a group of five men who will not be able to leave the tracks and escape in time. You, as the driver of the trolley-car find that the brakes are faulty and if you continue down the tracks you will kill the five men. However, if you choose to steer left and onto an alternate set of tracks, where one man is standing, you will only kill one man, while saving the lives of the other five. What do you do?

Same Ends, Different Means

This seems fairly simple. If I had to guess, I’d say that nearly anyone presented with the DIT scenario would opt to steer left and kill one man instead of five

Here’s how a utilitarianism would come to this decision. Utility is defined as a property of any object or action that brings about good, happiness, benefit and so on. Utilitarianists want to maximize utility, so they would choose whatever action saves the most lives and brings about the most utility, leading to their choice to sacrifice the lone man.

I would also choose to sacrifice the lone man in DIT But while I seem to have ended up at the same end-point as utilitarianists, I think I came to this decision through a very different set of means. This of course, I can’t explain using DIT alone, so I will move on to some more complicated scenarios in order to elaborate.

Bystander At Switch (BAS) Scenario

The popular BAS scenario that was first thought of by Judith Thomson, the philosopher responsible for popularizing the trolley problem, was meant explore the limits of our willingness to remain utilitarian.

The infamous trolley problem: do you flip the switch or not?

In BAS, the runaway trolley-car is hurtling towards five men who cannot get away in time. This time, you’re a bystander who’s stumbled onto the scene of this accident-waiting-to-happen. You’re standing beside a switch which, when pulled, will switch the trolley-car onto a different set of tracks that lead to where a lone man is standing. Do you flip the switch and sacrifice one person, or do you let five die?

Here’s where things start to get interesting. Utilitarianists might view BAS and DIT as two different ways of phrasing the same problem: Do you sacrifice one to save five? At the end of the day, whether you’re a driver or a bystander, it doesn’t change the fact that sacrificing one for the greater good of five is the right answer, at least for the majority of people, but not for me.

How Do We Measure Utility?

In BAS, I wouldn’t pull the switch. Yes, you read that correctly, I would allow the five men to die. But why?

First of all, the idea that five lives is always worth more than one life makes no sense. How do we even measure a person’s utility? Do we find the difference between all the good things and bad things he’s done? Or maybe we need to consider his future contributions to society? Is it about his status, wealth and other demographics? If you really think about it, there’s no simple way to sum up a person’ life and compare it with another’s in any useful way. Here, let’s try some examples. See if you can figure out who you’d rather save and who you’d rather let die:

1. Five registered sex offenders VS One doctor
2. Five decent middle class workers VS One wealthy philanthropist
3. Five high school students VS One elderly war veteran

Not so clear-cut is it? This is why we can’t be so quick to assume that five lives is always more valuable than one.

Furthermore, a key detail to note is that in the BAS scenario, the people trapped on the tracks are anonymous strangers. You, as a bystander have absolutely no information about their background, morals or future actions whatsoever. If their lives are in your hands, can you remain assured of each person’s overall utility if you don’t even know who they are?

Utilities VS Rights

We’ve established that measuring people’s utility is impossible and often unreliable. So is there a better metric we can use to make a decision?

To me, basic human rights seem like they would fit the bill perfectly. A basic right is something that you as a human being are entitled to from the moment you are born and one globally recognized and enshrined basic right is the right to life.

I’m sure everyone can agree that the right to live is universal to all humans. So to take away someone else’s right to life (a.k.a murder) would then be morally wrong.

Going back to BAS, we cannot ascertain the individual utilities of the six men on the tracks, so we shouldn’t assume that the utility of five trumps the utility of one. But we do know that all six men have a universal and basic right to life and that none of them have any less of a right to live than the others.

As an observer, you would be an outside party. If you don’t pull the switch, five men would die, but that wouldn’t be your fault because even if you didn’t happen to be there to witness it, the five men would still have died to the trolley-car. So by not pulling the switch, you would be letting five men die, but you wouldn’t have infringed on their right to live, which is an important distinction.

If you pull the switch however, that would be the moment where you interfered with the situation and ceased to be a bystander. Now, you are a part of the situation and your action will cause one man who, if not for you, would never have been in danger, to lose his life. You would have infringed his right to live. You would have killed him.

If pulling the switch leads to you infringing upon someone’s right to live while not pulling the switch does not infringe anything, then the morally correct thing to do would have to be to allow the five men to die.

Difference In Answers Between DIT and BAS

The main reason why I would sacrifice one life to save five in DIT, but would not do the same in BAS is because in BAS, as mentioned above, I am a bystander. It would be morally wrong for me to interfere with a situation if by doing so, I end up killing a man.

In DIT, I am the driver of the trolley-car and an active party in the situation. No matter what I choose to do, I will end up violating someone’s right to life. Faced with such a choice, I would have to minimize the amount of lives lost. So as a driver, I would have to kill the lone man.

Bystander On Footbridge (BOF) Scenario

This scenario is Thomson’s follow-up to BAS. Here, she further elaborates the arguments she makes in BAS, which I will explain.

If you would flip the switch in BAS, would you push the man?

In BOF, a trolley-car is still hurtling towards five unsuspecting men who cannot escape in time. You are again a bystander in this situation, but now you’re standing atop a footbridge that overlooks the trolley tracks. You notice a rather large man with his back turned away from you, also standing on the footbridge. You know that if you pushed the man over the railings and onto the tracks that he would collide with the trolley-car and die, but due to his size, he would be able to stop the trolley-car from ploughing into the other five men. Do you push him to save the other five men?

What makes BOF so interesting is that, for most people, utilitarianists included, pushing the man in order to save the lives of five men is morally abhorrent, yet pulling a switch to send a trolley-car towards one man is morally permissible. This change in answer is well documented and is also the one of the biggest flaws of a utilitarianist approach to the trolley problem in my opinion.

Image taken from article “Your Morals Depend On Language” by Albert Costa, Alice Foucart, Sayuri Hayakawa, Melina Aparici, Jose Apesteguia, Joey Heafner, Boaz Keysar

The Double Effect

See, if there is a legitimate reason for why we would be willing to pull the switch in BAS, but not willing to push the man in BOF, then there must be some fundamental difference between the two scenarios to cause this change. Thomson argues that this difference is caused by something known as the Doctrine of the Double Effect (DDE).

It sounds really complicated and believe me, the philosophical jargon that’s used to fully explain this doctrine is immensely confusing, but to put it simply: subscribers of this doctrine believe that we cannot do something good if by doing so, we also intentionally cause negative side effects. Conversely, we can do something good even if it leads to negative side effects as long as they were unintended.

Let me give some examples to make this easier. If a general in wartime chooses to carry out an air raid on some important military facilities in enemy territory, but in doing so, also causes collateral damage in the form of civilian deaths in the surrounding villages, his actions would technically be justified according to the DDE because the general didn’t intend to harm civilians. However if a terrorist seeks to detonate a bomb at a crowded area and hopes to cause civilian deaths while he’s at it to induce more fear, he is not morally correct in doing so since he intended to cause civilian deaths.

Thomson thinks that people have different answers for BAS and BOF because BOF involved actually pushing someone to their death, which is clearly intended. In BAS, Thomson sees the pulling of a switch as an indirect action that would unintentionally happen to lead to the lone man’s death.

But having different answers for BOF and BAS would be contradictory would it not? Utilitarianism maintains that saving five at the cost of one is correct. Both BOF and BAS give you the option to do just that. Yet 60% of people (at least, according to the graph above) wouldn’t maximize utility in BOF anymore because now they’d feel guilty about killing someone. Clearly, this inconsistency in answers exposes that choosing who lives and who dies isn’t purely a utilitarian numbers game.

My Personal Take on BOF

Personally, I don’t even fully believe that the double effect justifies a switch in answers between BAS and BOF. I think the manner in which a bystander sacrifices the one man is irrelevant. Can you imagine being told that you were about to be sacrificed but that it’s alright because nobody’s gonna push you, they’d just send a trolley to run you over?

Both scenarios require the bystander to knowingly adopt a plan that will kill one man, and so both scenarios will still unjustly infringe upon a man’s right to life. Of course, that’s just how I see it, but if anything, the significant number of people who change their answers from BAS to BOF seem to suggest that the double effect is very much a real factor to consider for most people.

My Argument Thus Far

Up to this point, we’ve gone over DIT, BAS and BOF, the three most well-known versions of the trolley problem. From which, we’ve identified some key flaws of a utilitarian answer:

But let’s say that in spite of all this, you still choose to believe that utilitarianism, however flawed, is still the most optimal approach to the problem. Luckily, Thomson had the same thoughts in 1985 when she came up with yet another version of the trolley problem to prove once and for all that one life shouldn’t be sacrificed to save five.

Bystander’s Three Options (BTO) Scenario

Now that sacrificing yourself is also an option, who would you sacrifice?

BTO has the exact same premise as BAS. Except you, the bystander, now have the option of either pulling the switch to the left and sacrificing the lone man’s life, or pulling the switch to the right, which would cause the trolley car to switch onto a set of tracks that you are standing on, meaning you are now an option in the problem.

The Issue of Self-Sacrifice

Thomson came up with this elaborate scheme to challenge those who still believe that sacrificing one to save five is the right thing to do. Well we can safely assume that neither you nor the lone man are particularly sacrifice yourselves for the other five men in this case since none of you were being threatened by the trolley-car to begin with. Since neither of you really caused any of this, neither of you are obligated to sacrifice yourselves.

The key here is that while you and the lone man are not obligated to sacrifice yourselves, only you, the bystander, have the power to decide who is sacrificed. Thomson states (and I completely agree) that if you are not willing to sacrifice yourself, then you would morally wrong to force the lone man to do what you yourself wouldn’t do.

This then means that if you weren’t willing to die as the lone man in BAS or die as the large man in BOF, then you would have absolutely no right to pull the switch or push the man in each scenario as a bystander.

Revisiting My Arguments Again

I think this reasoning of not doing unto others what you wouldn’t want done to you is pretty logical and is something most people probably picked up in kindergarten. This adds yet another point against utilitarianism:

However, just make sure that point number four is 100% airtight, let’s apply Thomson’s self-sacrifice theory to the original DIT scenario we started off with.

Driver’s Three Options (DTO) Scenario

To ensure that the idea that we should never do something to others if we aren’t willing to have it be done to ourselves holds really is the key to solving the trolley problem, we must investigate this idea not just from the perspective of a bystander, but also from the perspective of the driver.

If self-sacrifice is now an option for the driver, how does it affect his decision?

DTO follows the same premise as DIT, but now, the driver has the option of steering right and into a dead end, which would crash the trolley-car and end his life whilst saving the lives of the five men. Or, he could still steer left and sacrifice the lone man to save the five men without sacrificing himself.

Heroism or Lunacy?

Thomson actually argues that in a scenario like this, since you are the driver and not a bystander, you will end up either killing one man or five men if you do not sacrifice yourself. This is different from our bystander in BTO, who would only be letting five men die by not sacrificing himself.

As such, while Thomson feels that self-sacrifice in BTO is not required even if it’s an option, in DTO, self-sacrifice is required of the driver if it is an option, because he is fully able to do so given that there’s a dead-end track and should be doing so in order to prevent the trolley-car from killing any other people.

Looking at how Thomson puts forth her point in BTO and DTO, it is evident that her solution to the trolley problem relies heavily on the assumption that nobody in their right mind should be willing to sacrifice their own lives just to save others except where it’s absolutely necessary like in DTO.

This doesn’t sit right with me. I mean, firemen, soldiers and policemen often end up sacrificing their own lives to fulfill their civic duty to save complete strangers. And if the driver in DTO really did choose to end his own life to prevent the trolley-car from killing anyone else, he’d probably be seen as a hero.

So if the answer to why you should not sacrifice one to save five in BAS is that the bystander in BAS would never willingly sacrifice his own life, thus meaning he would never be allowed to sacrifice the lone man’s life, we would be assuming that nobody in the bystander’s shoes would ever be altruistic and selfless enough to even consider sacrificing themselves. But as per what I’ve mentioned regarding firemen, soldiers and policemen, its obvious that at least some part of the population really would die to save five others’ lives. Does this mean that this select group of people are then morally allowed to also flip the switch in BAS since they would be okay with sacrificing themselves if it were an option?

Essentially, this fourth point against the utilitarian choice of killing one to save five is wholly dependent on whether you believe that there are people selfless enough to sacrifice themselves to save others. If you do believe, then the point would be rendered invalid.

This brings us to Thomas Nadelhoffer, an associate professor at the College of Charleston, who also reached this impasse in the trolley problem that we find ourselves in now. He felt that Judith Thomson’s other philosophical work on the nature of self-defense actually provides an alternative argument against utilitarianism if the point about self-sacrifice isn’t something that resonates with you. But to prove this, he needed to look at the original BAS scenario through the lens of self-defense. So he decided to add guns to his own modified trolley problem. What could go wrong?

Bystander At Switch, Gun w/ Five Men Version

You’re about to die but you’re armed with a gun, who can you shoot to stay alive, or can you even shoot at all?

Okay, here we go. Its the BAS scenario all over again. But now, instead of you being the bystander or the driver, you’re now one of the five men on the trolley tracks who is about to die. You’re armed with a firearm that you can use to shoot any of the other parties if you think it might help you to stay alive. The question is, who are you morally justified in shooting?

Nadelhoffer thinks that switching the point-of-view of BAS towards the potential victims is the key to settling the trolley problem once and for all because by exploring who among the six potential victims has a right to self-defense (assuming they somehow possess guns), we can extrapolate that to determine which actions, when taken by the bystander, would not incur self-defense, and thus would not involve lives being wrongfully lost.

Who Can The Five Men Shoot In Self-Defense?

Let’s go over each of the parties from your perspective. You could shoot the driver of the trolley-car. But since the brakes don’t work and you wouldn’t be able to escape in time anyway, doing this is meaningless. While the trolley-car is an imminent threat, the driver himself isn’t really threatening you at all since this isn’t exactly his fault, so shooting him would be morally unjustified too.

Next, you might think: “ Hey, if I shoot that guy on the other set of tracks, then the bystander wouldn’t have a problem sending the trolley-car over there since he dead anyways” and shoot the lone man. But since the lone man wasn’t actually threatening the lives of the five men in any way, this isn’t self-defense either, it’s still murder.

If you sense that the bystander is going to choose not to pull the switch, you could shoot the bystander in hopes of getting his body to fall on to the switch and send the trolley-car towards the lone man instead (I’m trying my best to stay within the realm of possibility here). Once again, although the bystander has the ability to save you and five men, if he ends up choosing not to exercise this ability, he still can’t be seen as a threat to your life (as we have established in the previous scenarios), so you also aren’t justified in shooting the bystander.

Summary Of The Five Men’s Case For Self-Defense

None of the five men are eligible for self-defense against any of the other parties in the scenario

To sum it all up, if any of the five men were given the capacity for self-defense through the usage of a firearm, they still wouldn’t be justified in shooting the lone man, the bystander or the driver because none of these three parties have threatened or wronged the five men in any way. This includes the option where the bystander decides not to pull the switch and allow the five men to die by the trolley-car. Even then, the five men wouldn’t have been able to claim self defense against any of the three parties.

Bystander At Switch, Gun w/ Lone Man Version

Now, we shift to the perspective of the lone man with the same premise as before.

You’re about to die but you’re armed with a gun, who can you shoot to stay alive, or can you even shoot at all?

Let’s do this one last time. BAS yet again, but you are adopting the perspective of the lone man and you are armed with a firearm. Just like before, you can choose to shoot any of the other parties if you think it would help you to stay alive. Who are you morally justified in shooting?

Going over each of the parties one-by-one, it becomes clear that the lone man’s case for self-defense is stronger than the other five men.

Who Can The Lone Man Shoot In Self-Defense?

You obviously are not morally justified in shooting the driver for the same reason as the five men, because the trolley-car is the imminent threat and not the driver.

You could shoot the five men dead so that the bystander no longer needs to choose who’s lives to sacrifice, but that’s also obviously morally wrong.

Lastly, if you think that the bystander might try to pull the switch and sacrifice you, you can shoot him so that he can’t get to the switch in time. Nadelhoffer believes that in this scenario, you would actually be able to claim self-defense.

He references Thomson’s own paper titled “Self-Defense and Rights”, where she argues that if an aggressor is about to cause you great harm or violate your rights unless you stop them by causing them great harm or by violating their rights, then you would be justified in causing them great harm or violating their rights, even if the aggressor did not intend their actions.

That’s a lot of information to process, so here’s two examples. If a murderer is going to kill you unless you hurt or kill them first, you are justified in doing so. If a schizophrenic individual is attempting to kill you because of their hallucinations unless you hurt or kill them first, you are also justified in doing so even though they may not actually intend to do you any harm.

So if a bystander is actively formulating a plan to save five people, but this plan will also definitely kill you in the process, regardless of whether you think he is doing so intentionally or unintentionally, you would be justified in shooting him first in self-defense.

Summary Of The Lone Man’s Case For Self-Defense

The Lone Man can only be justified in shooting the bystander if the bystander attempts to pull the switch

For the lone man, if he were to have the capacity for self-defense by virtue of having a firearm, he would not be justified in shooting anyone except the bystander if the bystander attempts to pull the switch.

In other words, the only party that would have wronged the lone man in any way in this scenario, would be the bystander if he pulls the switch.

Hence, taking both this scenario and the previous scenario where the five men were in possession of a firearm into account, the only action taken by the bystander that would wrong anybody in either scenario and invite retaliation, would be him pulling the switch. Which means that by using the right to self-defense as a decision-making calculus, the morally correct thing to do if you were a bystander, would be to not pull the switch.

Conclusion and Final Thoughts

All in all, by systematically going through a whopping seven variations of the trolley problem, and by adapting some of the fantastic philosophical heavy-lifting of great philosophers with my own interpretations, I believe a case against utilitarianism in the context of the trolley problem can be summarized by the following five points:

Some points may seem more appealing or less palatable depending on your own unique moral compass, but I think it would be exceedingly difficult to disagree with all five. As such, I hope that if you had stumbled upon this article by chance, that you can leave with a better understanding of why some would choose not to pull the switch.

But before I end off, I do want to address a final point about the trolley problem, which is exactly that: a point. Is there really any point in thinking about a solution to the trolley problem? Over the last five decades since its inception, many have to come to the conclusion that debating about whether we would pull the switch is a complete waste of time.

Despite having spent quite a considerable amount of effort researching and writing this article, I do agree that to a certain extent, it is a waste of time. I mean, how often are you going to find yourself in a situation where you hold the lives of six people in your hands and trolley-cars are just mysteriously running astray? The stakes of the trolley problem are just too extreme to hold much real-life application. In fact, the internet has even started to satirize trolley problems with its own brand of trolley memes. Just take a look:

So has this entire article and perhaps the decade’s worth of philosophy papers and forums and discussions been pointless then? Well, I wouldn’t go so far as to say that.

Yes it is true that most of us will never have to face a real-life trolley problem (probably for the best too) and yes it is also true that the stakes of the trolley problem are far too high to be applicable. But in the process of thinking through and formulating a coherent, logical and defensible answer to the moral dilemmas posed by thought experiments such as the trolley problem, we get a rare opportunity to do some introspection, figure out what it is that makes us tick and identify important moral principles to guide us in life. It’s not going to cure cancer or solve global warming, but if it helps us to better understand ourselves and how we should be seeing the world around us, then the trolley problem is far from useless in my eyes.