The “Trolley Problem” is Fundamentally Flawed

The so-named Trolley Problem first originated in 1905, and since its full development as a thought experiment in 1967 by Philippa Foot, it has captivated the on-going attention of philosophers and psychologists for decades. More recently, as indicated on the Facebook page “Trolley problem memes,” developing various iterations of the Trolley Problem is a popular pastime for everyday people, too.

The basic scenario involves a runaway trolley that you observe. On its current path, it will kill five people who are working on the tracks (or who are tied to the tracks). If you pull the lever next to you, it will go down another set of tracks with one person in the way. What do you do, if anything? Death will occur no matter what. Unlike psychologists and philosophers who are interested in how people reason, apply ethics and morality, and about possible contradictions (or cognitive dissonance) in their ethical world views, I am not immediately interested in whether you testify that you would or would not pull the lever.

One particularly interesting alternative Trolley Problem, on the aforementioned Facebook page, calls into question our definitions of “life.” In this situation, the runaway trolley will kill one man tied to the tracks or if the lever is pulled, five plants–yes, plants–will die. But, again, for purposes here, I am not interested in definitions of life, but do like the added layer.   

Other classic scenarios say that you have an opportunity to save the five people by pushing one very large person in the way of the runaway trolley. Similarly, though, I am also not instantly interested in whether you would push this person in the way or not.

I am, however, interested in the observable identities of the people involved.

Quick: In the situations above, what identities did you immediately and mentally “assign” to the various people? Even, what kind of weather did you “assign” to the environment? How did this influence your answers and then your justifications?

Except for a handful of scenarios that involve someone who is overweight, a Nazi, or a researcher, for example, identity is not part of the equation in scenarios researchers (and internet users) regularly develop. All are basically “generic,” except for a few situations developed by high school student Tiffany Sun and to a lesser extent, some by researcher David Pizarro. Both found, unsurprisingly, that when minority variables are added, people are more willing to choose that person to die.

As a researcher always interested in the new and unusual and as a professor of both History courses and Women’s, Gender, and Sexuality Studies courses, I want much more information about the five people tied to the tracks, the one person on the other tracks, and the person who now has some control over life and death, for example. In what country does the Trolley Problem take place? Would society racialize the five as White, Black, Hispanic, Asian, or something else? Does this group include just men, just women, both men and women? Are these people gender-conforming or nonconforming? Are these people tall or short? Are they healthy or unhealthy? Are the people younger or older? Are they able-bodied or disabled? Are they rich or poor? From all kinds of quantitative and qualitative data from researchers across the spectrum, we know that such identity factors matter a great deal and that they matter in ways people are not even aware of and cannot be aware of.

The Trolley Problem is wholly incomplete without considering questions of identity and intersectionality, or how these identities interact and overlap. Consider the following variations of the Trolley Problem. What would happen if we found 100 people, divided them into four groups, and gave each person in each group one of the situations? How would people respond differently to the different examples? That would give richer results, especially if instead of hearing or reading the following situations, people saw an animation of it.

There is a runaway trolley. It is headed toward a group of White men. If you do nothing, this group will die. If you pull the switch a Black trans woman, who is young, tall, and physically fit will die instead.

There is a runaway trolley. It is headed toward a group of White men. If you do nothing, this group will die. If you pull the switch a Black (cisgender) woman, who is young, tall, and physically fit will die instead.

There is a runaway trolley. It is headed toward a group of White men. If you do nothing, this group will die. If you pull the switch a White trans woman, who is young, tall, and physically fit will die instead.

There is a runaway trolley. It is headed toward a group of White men. If you do nothing, this group will die. If you pull the switch a White (cisgender) woman, who is young, tall, and physically fit will die instead.

Given what we know about the history and psychology of racism, people in general would be more likely to save the White cisgender woman than a woman of Color or a trans woman. People might actually not even truly “see” the non-White woman.

Other variations could pull on the following:

There is a runaway trolley. It is headed toward a group of White men. If you do nothing, this group will die. If you pull the switch an Asian Child, who is a baby will die instead.

There is a runaway trolley. It is headed toward a group of five White men. If you do nothing, this group of five will die. If you pull the switch a group of four poor, middle-aged White man will die instead.

The Trolley Problem is much more complex, even much more beautiful by making it more detailed, by considering identity and by queering it. Why do the various iterations of the Trolley Problem involve generic “people”? Why hasn’t this conversation already occurred again and again? How does this introduce new ethical problems, ethical problems that philosophy–a discipline generally dominated by White men–might prefer to ignore or might be unaware?

The Trolley Problem could be made even further interesting by making it come to life–factoring in the weather and time of day, the political climate of the nation, the education of the people involved, and an infinite array of additional factors.

As Omid Panahi argues in a recent issue of Philosophy Now, there is no solution to the Trolley Problem. Therefore, I want people to embrace that there being no solution even available is all the more reason to make the examples more complex and more detailed, especially where identity is concerned. While I understand that the original experiment’s aim to be as clear and as simple as possible, this takes the situation completely out of the human experience, even somewhat eliminates ethics by making it ideal. (I mean, does the lever even work?) Depending on the detail provided, people will read any body language available and likewise use that to influence their unconscious decision and unconscious ethical reasoning. Embrace the power of thought experiments. And until researchers do their “thing,” we will not know how the inclusion of identity changes responses.  

Dr. Andrew Joseph Pegoda

Categories: Thoughts and Perspectives

Tags: , , , , , , , ,

8 replies

  1. Truly an interesting conundrum. You could put some version of this trolly problem to test on each of your classes, even filling in the details you suggest at some time, keep collecting over the years. You probably teach small classes where this can be done but only over time accumulating a decent sample size, but might still need to do time series analysis to rule out changes in answers due to normal generational times. But it could be really interesting. You might even enter a variable, like “you know the one person on the other track” and “you never liked him/her” or “you never really got to know him/her but can still recognize him/her,” and then balance the other track as including someone of equal “importance” but different “quality.” Or ask your students to describe a situation where they would still be very emotionally involved but likely to just freeze because they can’t choose. Of course, all submitted anonymously. In the end, if you are not wanting them to think ethically, but only to become aware of their own biases or categorizations of people, that last method might be the best.

    Liked by 1 person

    • I’ll have to think about all of this! I’m for sure going to have to do something with this in my Queer Studies classes this semester. It usually takes me at once or twice of doing an experiment like this to do it well enough to get meaningful results. I have to figure out, in the moment and based on how they respond, how to explain the task and whatnot. Part of me is tempted to do some version of this one-on-one during office hours, but that would take a lot of time if I did it with everyone. I could do it with some of the really good ones who come by to visit often – that would be a good way to kind of “test out” everything.

      Liked by 1 person

  2. Not all that dissimilar to the sinking lifeboat-in-a-storm scenario. Someone has got to go or everyone dies. You are the crewman in charge to make the decision. While the trolly person has the immediate inherent power of determining the minimizing potential of how many die depending on the throw of the switch, the lifeboat crewman has to decide in the face of people knowing that one must die. Such is the responsibility of command. Interesting scenarios all. In the end we can ALL acknowledge, no one wants to be in that position to determine the death of another. Or as Mr. Spock put it as an example of pure logic.. the needs of the many outweigh the needs of a few.

    Liked by 2 people

    • These questions help us realize that “morality” is subjective and relative! 🙂


    • I think its easy to think this way when in theory, one has no information about the people tied on the tracks, besides how many of them are on each rail. I think that for sure, the idea of them being complete strangers definitely facilitates that kind of objective thinking.But even if you were to choose to save the five instead of the one, it doesn’t take away the fact that you’d be responsible for the death of someone and that responsibility being something for you to carry with you the rest of your life. Like it was stated in the article, there is no right answer, but it doesn’t make it any easier to decide which option to go for, if any at all.

      Liked by 1 person

  3. I really thought Sheri Fink had some good points on this topic.
    Playing God via @Radiolab interview with Sheri Fink.

    Liked by 1 person

  4. How about you do nothing and ensure the maximim carnage and death because it will bring down the government for failing to enact trolley safety regulations and it will bring into power your preferred political party which will then enact stricter trolley operational practices?

    “They gave their lives for a safer tomorrow …”

    How about the question that you are commander of a submarine and an enemy navt ship appears to be leaving the conflict zone. Do you obey the order from your Prime Minister to kill all of the enemy on board even though it is an “act of cowardice” (only a coward stabs a man from behind) or do you disobey? And how would your decision be influence if you knew that as a result of your action, the enemy would retaliate by killing many more of your own men which is what the Prime Minister actually wants in order to ratchet up hostilities to full scale war even though shedding crocodile tears over the loss of life?

    “The enemy had to be killed to save our brave men’s lives” (Even though the government knew it would actually lead to greater loss of life.)

    Liked by 1 person


  1. 7 Things You Likely Have Wrong That Few Address – Without Ritual, Autonomous Negotiations
%d bloggers like this: