A Real-Life Trolley Problem

Many of you are probably familiar with the Trolley Problem, a classic ethics problem phrased something like this:

There is an unstoppable trolley hurtling down its track towards three innocent victims tied to the track. You’re fairly certain they will be killed instantly if the trolley reaches them. Fortunately, you find yourself standing next to a switch which can alter the route of the trolley away from those three people, saving them. Unfortunately, there is another innocent victim tied to the alternative stretch of track, and you’re fairly certain that the trolley would kill him if you diverted it. Do you flip the switch?

In its original form, the  Trolley Problem is one of many examples of problems where consequentialism (cost-benefit analyses) and deontology (“do no evil”) come into conflict. Consequentialists prefer the world where only one innocent person dies, all things equal, so they tell you to flip the switch. Deontologists prefer not to murder the one, so they would tell you not to. I’m probably oversimplifying, because at least to me, the consequentialist position in this case seems rather compelling.

So then it becomes a bit of a psychology question and they ask: “Instead of being next to a switch, you find yourself next to a fat guy on a bridge above the trolley. The only way to stop the trolley before it reaches the three innocent victims is to push the fat guy off the bridge onto the track, which you’re now certain will do the job but will unfortunately also certainly cause him to die.”

Of course, this adds some complications, and fewer people are willing to push the fat guy, even assuming by fiat that it will work. Many people point to the fact that pushing someone off a bridge feels more active than flipping a switch. But there are also universalizability concerns: If this happened consistently to fat people, they might either lose weight or stop hanging around bridges, defeating our solution. (Of course, I’ve imagined that trolley problems became common enough for us to care about as a society, so bear with that assumption for now.)

We can make further additions to the story on top of that, like:

  1. What if there is only one victim on the tracks?
  2. What if the fat guy didn’t choose to be on the bridge, but was dragged there by a friend of a victim?
  3. What if, instead of instant death, the trolley victim(s) only have their legs attached to the track, so they’ll “only” be paralyzed for life? Or (somehow) financially ruined? Socially ostracized? Excruciatingly pained?
  4. What if the fat guy is already paralyzed/poor/ostracized/pained and hates his life?
  5. What if he was already going to fall to his death, but you have the opportunity to time his descent appropriately and in doing so save the victim(s)?
  6. What if he was only going to probably fall, but by pushing him you can guarantee he does?
  7. What if he is comatose, and we can’t really agree on whether he’s alive?
  8. What if, rather than being captured, the “victims” were just foolish imbeciles playing around near trolley tracks, only to fall into traps on the tracks that you warned them about many times? (Like Simba in The Lion King?)
  9. What if these traps on the tracks are specifically fitted only to catch short people, but not tall people?
  10. What if tall people have been playing around those tracks for years without bearing the same consequences?
  11. What if tall people already have numerous other advantages in society as a whole?
  12. What if the tracks are in a dangerous neighborhood?
  13. What if these trolley accidents are a major component of the danger to that neighborhood?

Think through each of these considerations, and how big a weight they might play in your mind. Maybe even write them down. You can probably see where this might be headed, but let’s return to the original question first.

Unless they are forced to make up their mind, most people I notice try to back out of answering a trolley question conclusively, arguing that these convoluted scenarios are far from reality, and the closest thing you get is when sociopaths like the Joker in The Dark Knight run things:

And in some sense, that’s true. That movie is filled with trolley problems, which the Joker sets up as twisted experiments to get people to murder each other and lose those inhibitions and societal trust. Taking a bigger picture angle, another good response to the original question is something like, “Who tied the victims to the track in the first place? If it’s a sociopath, we might not want to trust the switch to actually save them.”

But its piercing gaze at our own ethical principles is one reason The Dark Knight is considered one of the best movies of all time. Can a carefully constructed consequentialist scenario overcome commitments to deontology? Can the Joker make Batman break his only rule? Should he even have had that rule, if running over the Joker in the middle would have kept the atrocities in the second half of the movie from happening?

Okay, so what am I really talking about? If you haven’t figured it out already, I’m wading into the abortion debate, trying out a perspective that I haven’t seen brought up much in this case. In fact, this isn’t new at all: Philippa Foot invented the trolley problem specifically to talk about abortion back in 1967.

I’ve made some substitutions, which I think you can figure out, but just to spell them out clearly: The fat guy that we can push off the bridge is the fetus being aborted. The victim(s) are the parents or possibly just the mother. Fetuses aren’t comatose, but they also aren’t conscious in our usual sense of the word, or in an observable way that, say, animals aren’t. Playing near the tracks is having sex, and short people are women.

Why make this analogy in the first place? Well, the key similarity between abortion and trolley problems is that we have two parties with strong competing interests, the victim(s) and the fat guy, and the mother/parents and the fetus. Since it’s the fetus’s death we’re talking about, we need dilemmas of similar magnitude to assess our moral intuitions.

The key, of course, is that the trolley problem starts with something where the tradeoffs are clear, one death versus three, and then mutates it so that they might no longer be obviously comparable, but the priming of the previous example keeps us from shutting our brain off with regards to both parties.

So where do your moral intuitions suggest you lean? Under which of these combinations of situations do you push the fat guy? Here’s that list again:

  1. What if we’re only deciding between the lives of the child and mother?
  2. What if the fetus had no choice in the matter, and was placed into this precarious circumstance through the actions of its parents?
  3. What if there is no risk of death to the mother, but just the burden of raising a child, the financial cost of the pregnancy and childbirth, and social ostracism and excruciating pain of pregnancy?
  4. What if the fetus would grow up to be a neglected child and wish he/she had been aborted?
  5. What if the fetus is early enough that there’s a significant chance of (potentially unnoticeable) miscarriage?
  6. What if it’s later, so there’s only a mild chance of miscarriage?
  7. What if the fetus isn’t a living human yet?
  8. What if the sex that produced the fetus wasn’t rape, but poor decision-making on behalf of the parents?
  9. What about the fact that mothers are affected much more by pregnancy than fathers?
  10. What about the societal perception that it’s okay for men to be promiscuous, while they don’t bear the same consequences?
  11. What about sexism in general? Shouldn’t we seek to advance women’s rights?
  12. What about the fact that unwanted pregnancies are much more common among the poor?
  13. What if fewer births among poor people cause a drop in crime?

With the full analogies, go back and evaluate: Which did you find compelling enough last time to toggle your answer? Which comparisons do you find unfair? Which caught you by surprise?

To me, the biggest leaps are in questions 1 and question 3, collectively going from “three people are going to die” to “a mother is going to have to go through a pregnancy and childbirth, and then choose to raise the child or put him/her up for adoption.” Of course, question 3 doesn’t apply if there is a significant risk of death to the mother as a result of the pregnancy.

We can even start to put some numbers to these things, converting to money simply as a useful common unit of measurement. The estimated cost of raising a child born in 2013, according to the USDA, is around $250,000, or $300,000 accounting for projected inflation. Notably, this doesn’t include the cost of college or pregnancy, or the time spent raising the child, but maybe we can agree on a number around $500,000. Meanwhile, the EPA and other government agencies have to assess how much money to spend on potential life-saving measures, and they place the monetary value of an average human life at around $8-9 million. (The kidney dialysis-derived yearly numbers come out similar for a 20-year-old.) Of course, these aren’t quite the same numbers because of transaction costs and emotional harm considerations, but they should be the right orders of magnitude.

My first reaction to that comparison: Parenthood is a really good investment, societally speaking. We put in less than 10% of what we get back!

My second reaction: This should give us an order of magnitude comparison between the value of an adult versus a child. Some might claim that if given the trolley-problem tradeoff between a young adult and a child, we should favor the young adult because we’ve already invested a lot into him/her. The first answer, of course, is that that investment is one or two orders of magnitude less than the value of life in general. The second answer is that the child has more years of quality life remaining in expectation, so even at only the lower $50k/year number, we should still consider childhood worth living, all things equal.

My third reaction: We are so far from a Malthusian / “Soylent Green” scenario, where people are so abundant that they are more valuable for cannibalism than anything else, that it’s not even funny. If there was a fierce competition for resources, the going price of saving a life would go way down. As it is, even in the developing world, we’re still in the millions of dollars per life range.

My fourth reaction: Wait, what about the fact that the Against Malaria Foundation can save a life at a cost of around $3340? That they haven’t already captured all that low-hanging fruit suggests we don’t actually value faraway life that much. Well, meta-level considerations (similar to the winner’s curse) suggest that as the lowest cost known to websites like GiveWell, it’s likely to be an underestimate. Probably not a 1000X underestimate, though. Still, the existence of low-hanging fruit like that does not suggest we shouldn’t spend money raising kids, because African children who we save from death still need to be raised, too.

Anyways, I’ve gotten a bit far afield. Back to the questions above and the abortion debate. If the mother’s life is not at risk, then there are a couple orders of magnitude to overcome in cost.

Question 7 is the strongest response, that we can’t really agree on when life begins. Christians believe a fertilized egg is life, generalizing from several passages in the Bible talking about God forming the author in his mother’s womb. While a little work is needed to distinguish those passages from claims of predestination (God simply foreseeing that we would be born, or setting up the context for it), that work is not hard and we can probably agree that the Bible considers fetuses human at any stage.

But outside of Christianity, there isn’t as broad a consensus. Various other angles like viability or self-awareness suggest we shouldn’t consider fetuses to be humans, while its unique genetic material and resemblance to human form both suggest rather early definitions. The closest thing to a consensus would be a gradual progression from non-life to life, which is why the current laws set a specific age of the fetus before which it can be aborted.

Questions 5 and 6 are meant to apply to debates over the morning-after pill, one of whose effects is to prevent a fertilized egg to becoming attached to the wall of the uterus, leaving it to die. There’s some evidence, though, that fertilized eggs often fail to become attached anyways, hence why question 6 might arise.

I think you can see how the other questions come into play. Question 2 highlights the helplessness of the fetus, and the inability of fetuses to form a Fetus Association and avoid entering wombs of women who will abort. Question 4 brings up the quality of life question: If the child will be raised in destitute poverty, are they going to wish they had been aborted? In non-rape cases, question 8 reminds us that the dilemma is a direct result of the parents’ voluntary actions.

Phrasing abortion as a “woman’s right to choose” is specifically pointing out questions 9-11, that pregnancy affects women much more than men, especially if the father is out of the picture. Throw in the class considerations of questions 12 and 13 (including the Freakonomics argument that legal abortion decreases crime), and you have the key attachment points of the pro-choice movement into general liberal concerns for women and the poor.

In my mind, the concerns of those five questions don’t seem to be on the same order of magnitude as the costs we are talking about, but they’re still worth addressing. Heck, the same should be true of the costs to the parents, in cases where the mother’s life is not at stake. The economic idealist in me sees an opportunity for some sort of compromise here, between a society that values a life on the order of millions of dollars and parents, women, and poor people with concerns on a smaller order of magnitude.

Perhaps this might take the form of a scholarship for poor women if they have kids. I can see pretty clear objections to that proposal, from concerns about incentivizing unwanted pregnancy to enforcement issues to make sure they’re actually spending that money raising the kids. But can’t we just be more creative? On a more local level, it’s best to combine any pro-life activism with actual assistance given to pregnant women and new mothers who might have otherwise considered abortion. This is actually something I see churches doing quite well, in my experience.

Why talk about this now? Well, making the rounds on Facebook is a video purporting to show some dastardly behavior by a prominent abortion doctor with Planned Parenthood. Some of the best coverage I’ve found comes from the Washington Post, Christianity Today, and Snopes.

I’m intrigued by the questions at the end of the Snopes article, regarding this “Center for Medical Progress” that recorded this undercover video. They apparently managed to disguise themselves as a stem cell research lab in order to gain access to Planned Parenthood executives, before flipping their purpose and releasing this video, which they claim is the first in what will apparently be a weekly series of revelations for the next few months. You don’t see that kind of drama every day.

Of course, I would be remiss not to address the stench of moral repugnance, the ugly sight of a doctor callously discussing fetal tissue donations over lunch. For me, this was most reminiscent of an eerie scene from Lord of the Rings: The Return of the King, after Denethor sends his son Faramir to foolishly try to retake the overrun river city of Osgiliath with far too few soldiers:

At the same time, doctors routinely become numb to the literally life-and-death issues they face on a daily basis. It’s one of the only ways to cope, and to be able to discuss this question in their arena, we need to set aside those feelings of disgust. This is why I started with the trolley problem analogy: We can also consider the tradeoffs in this case as well. The key question: Should it be legal to donate fetal tissue from abortions to research purposes in exchange for a small fee?

On the one hand, assuming we aren’t touching the legality of abortion itself, it seems pretty clear-cut. This is the situation of question 5, where the fat guy was going to fall to his death anyways, and we just timed his fall so that he might stop the trolley. While they might be repulsed by the idea for themselves, people are generally okay with voluntary organ donation (say, if you died in a car accident) and this situation is somewhat analogous.

On the other hand, there are some pretty big differences. The fetus can’t give consent the same way an organ donor can. In fact, they ask the mother to give that consent, which seems a little strange to anyone considering them as different entities, given that she just chose to terminate the fetus. (Then again, she donated nearly all of the energy/nutrients to create that tissue, so she probably has some claim to it. Pregnancy is a weird notion to reason about.)

Additionally, organ donation goes to save the lives of people who need a transplant, a very noble and valuable goal, and fetal tissue goes to… some kind of research? One thing I haven’t seen in all of this discussion is any attempt to assess the importance of fetal tissue to that research, and the prospective value of that research as a whole.

Once money starts changing hands, there is also a small cause for concern. If demand for more fetal parts motivates a regional Planned Parenthood clinic to encourage women to get more abortions, it starts to resemble pushing additional fat guys off bridges. However, the amounts mentioned in the video ($30-$100) are small enough to be described as a reasonable fee for the associated preservation costs, especially given how expensive medical procedures often are.

What does the current law say? Well, as discussed most clearly in the Christianity Today article, the clinic is allowed to charge for such preservation costs, within unspecified “reasonable” limits. The one part that Planned Parenthood might get in trouble for is changing the method of abortion based on demand for specific body parts, but on the scales we’re discussing, that seems like a minor overstep.

Should there be oversight? Sure, within reason. This video doesn’t give enough evidence that they are profiting off of these expenditures, but CMP also found an advertisement from a company that does purchase fetal tissue touting the financial benefits to the clinic. There’s no direct evidence that Planned Parenthood or its clinics profit off of these sales, however, and this is something they have vehemently denied. In that light, being compliant to some level of reasonable oversight would give them a defense to point to and everyone else a firmer peace of mind.

Of course, the bigger question is what this should mean for abortion rights in general. First, the usefulness of fetal tissue to research on humans isn’t in itself evidence that they are alive or conscious as humans. We experiment on mice and yeast all the time, because of various similarities with us, and seem fairly comfortable with allowing researchers to humanely kill the mice. While it might disgust us, on reflection, we probably are morally okay with casual discussion of the procurement of mouse hearts, lungs, and heads for research purposes. If the only revelation is that fetal tissue is useful for research, this doesn’t tell us anything morally.

The bigger effect, beyond the emotional content, is that the taped discussion serves as a reminder that some fetuses who are aborted do have intact heads, hearts, lungs, and livers. This vividly reminds all of us that similarity-based determinations of the beginning of life probably place the start date earlier than the legal abortion limit (at least in the US), and maybe motivates people to put more weight onto the similarity approach. For those with an early date already, though, it doesn’t change much.

Well, now that I’ve written this long 3500-word article, where have we gotten?

  1. We can use variations of trolley problems to assess our moral intuitions in cases where there are confusing tradeoffs, often involving death. I’ve hopefully modeled that by applying it to both the abortion problem and the related concern of using abortive tissue for research purposes.
  2. The cost we are willing to spend as a society to save a single (American) life is ~20 times higher than the cost of raising a child, suggesting that the price of a death of a human dwarfs the cost to parents. (But both are also much higher than the cost we can spend in the private sector to save lives internationally…)
  3. The biggest contributor to the abortion debate has always been, and remains, the question of when human life begins. Consensus will involve compromise on this question for the foreseeable future.
  4. The latest viral contribution to this topic probably creates more heat than light, but voices support to an earlier start to life for those who find similarity a substantial contributor to that discussion.

What are your thoughts?

11 responses to “A Real-Life Trolley Problem

  1. anon July 16, 2015 at 1:49 pm

    The question is not when life begins, it’s when life begins to have moral value. Surely we can all agree that a mosquito, a skin cell, and a fetus are all alive. Yet this doesn’t mean that it’s wrong to kill them.

    If you want to argue that abortion is wrong, then even from a Christian perspective, you must argue that God doesn’t want you to kill fetuses, and not merely that he views them as alive.


    • Sam Elder July 16, 2015 at 2:10 pm

      Right, I should have said “human” rather than “alive.” I’ve edited the post accordingly — thanks!


      • anon July 16, 2015 at 8:24 pm

        There’s a more subtle issue here, which is why killing a human is wrong in the first place. Is it because the victim doesn’t want to be killed? Is it because it’s bad to destroy thoughts and memories? Is it just because God said so?

        Note that (early-term) fetuses can’t want anything, don’t think, and don’t have memories. I also don’t know if the Bible ever says abortion is bad. The biblical quotes you cite don’t say it is wrong to kill a fetus. They talk about God forming a person in the womb, but that by itself doesn’t mean much: you could say that I was “formed” at conception, but that it was still moral to kill me until I developed thoughts and memories.


        • Sam Elder July 17, 2015 at 3:09 am

          In this post, I considered this something we’ve already decided as a society – and we’ve put the price we’re willing to pay to save a typical (American) life around $8-9 million.

          Why do we choose to spend resources to save lives? The main reason I can see is reflection: We would want to be rescued ourselves, up to some cost which we’ve settled on by consensus. We don’t want to draw too many lines between people which that reflection can’t cross, for parsimony and to prevent corruption, but some are natural, like the Secret Service protecting the President, the dependence on the amount of quality life you could lose in medical decisions, and the preference to spend government money on Americans.

          Reasoning based on reflection naturally leads to approaches like similarity in determining whether to extend that reflection. Unless I’m missing something, genetic and consciousness-based arguments are essentially avenues for exploring similarity. For Christians, there’s also another priority of preserving the “image of God” which we imagine to be some way that God interact with us that, based on those Bible passages, also seems to take place in the womb.


          • anon July 17, 2015 at 1:04 pm

            Why do you conclude from these passages that the “image of God” occurs in the womb? And where in the bible does it say that we must preserve this “image of God”?

            You’re skipping a bunch of steps here. Maybe they’re standard steps in Christianity, but if so, I’d like a link to a source explaining it. As I see it, Christians believe abortion is wrong only because other Christians believe abortion is wrong: it is entirely a social phenomenon, with very little biblical basis.

            Liked by 1 person

            • Sam Elder July 22, 2015 at 10:54 pm

              Good questions! I thought it wise to check with my pastor before responding, so apologies for that delay.

              The central passage tying murder to destruction of the image of God is in Genesis 9:6: “Whoever sheds the blood of man, by man shall his blood be shed, for God made man in his own image.” Here the theological consequences of murder are explicitly laid out, connecting back to statements in Genesis 1 and 5 that endow humans with the image of God, as God’s creation. The passages I cited invoke the same imagery of God creating humans, in the womb.

              Does that make sense?


  2. anon July 16, 2015 at 10:00 pm

    Would you donate your kidney to save a stranger?


    • Sam Elder July 17, 2015 at 2:01 am

      I’m signed up for organ donation if I die in a car accident. Just donating one kidney out of the blue is something I’ll probably consider later in life, assuming that there isn’t a foreseeable drop in demand on the horizon. (Donating now seems strictly worse than donating later.)


  3. Pingback: Singapore: General Impressions | The Christian Rationalist

  4. 86Randal August 21, 2017 at 2:19 pm

    I have noticed you don’t monetize your website, don’t waste your traffic, you can earn additional bucks every month because you’ve got high quality content.
    If you want to know how to make extra money, search for: best adsense alternative Wrastain’s tools


  5. Pingback: How I Blog | The Christian Rationalist

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: