Monday, March 28, 2011

Mailbag Monday: Trolley Problems

Mailbag Monday: A weekly segment that covers readers' questions and concerns about all things Philosophy, Bro, and Philosophy Bro that don't quite fit anywhere else. Send your questions to with 'Mailbag Monday' in the subject line.

Brad writes,
Hey Bro,

Do you think you could go over the Trolley Problem? You know, where a trolley full of bros is barreling toward a cliff and the only way to save those bros is to divert the path of the trolley so that it hits and kills some other bro. What's a bro to do?
Absolutely, Bro. I love the trolley problems more than Rebecca Black loves Friday. They're used to examine important ethical dilemmas; wrestling with them is a staple of any education in philosophy, like being asked how you know over and over or pulling out your hair and yelling, "THAT'S A FUCKING CIRCLE, DESCARTES". It's a moment we've all had.  But they're used in a whole host of other cool ways, too.

So let's start with the basic TP, stated in the question above. If you pull this lever, you switch the tracks and save the lives of, say, 5 bros - but the track that you switch it to has a bro on it who can't get out of the way in time, and he'll be killed. Decision time!

Utilitarians fucking own this trolley problem. "Um, fuck yes. Save five instead of one. Do people really have a hard time with this one?" It's pretty much game, set, fucking match if you're a utilitarian. 

I don't think even Kant would have a problem pulling the lever in this case - you're not treating the bro on the track as a means to save these lives, his death is just an unfortunate consequence. Fucking sucks to be him, bro, but I obviously have a duty to save lives when I can do so without doing something wrong, and pulling a lever isn't intrinsically wrong. Seems simple enough.

But the first trolley problem is often used as setup for the trap: thought experiments are flexible, and we can hack this one to test exactly the notion Kantians would have to rely on - means and ends. The Fat Man Trolley Problem is probably the second-most famous TP:
A trolley with five bros on it is hurtling toward a cliff, but walking on the side of the track is a big, fat motherfucker. Now, if you push him over, the trolley will run him over and kill him, but it'll stop in time so that it doesn't go off the cliff and the five bros survive. What do you do?
Suck it, Kant. Suddenly, you have to use the fat man as the means to stop the trolley - this is no accidental consequence. What do you do? Kant would say something like "Well this fucking sucks, but killing a guy isn't acceptable as a means. So... Sorry, bros." And now it looks like Kant is a fucking dickhead for killing five bros and utilitarians rule. 

Of course, not all deontologists (moral philosophers who believe in duty-based ethics, not consequential considerations) agree with Kant. Last week I mentioned W.D. Ross, who believes that we have a bunch of obvious duties. Among them, he lists the duty to not be a dick, and the duty to be good to people when we can. When our duties conflict, we weigh them to decide what our overall duty is - like utilitarian thinking, but for duties (heh) instead of pleasures. Now, pushing a guy into an oncoming train is easily in the top ten dickest acts of all time - just ahead of 'torturing a baby' and slightly less dick than 'being The Situation' - so we better have one hell of a strong duty to override that. Is being good to five bros enough? Hard to say. But maybe. Consider the 'Dirty Hands Problem', which is similar to this problem:
A dictator is massing military forces on the border, preparing an invasion of his neighboring country, which is pacifist. Hundreds of thousands will be killed. His son, his successor, disagrees with this policy and will not carry out the invasion if his father dies. Yours is the only military well-trained enough to take him out successfully; you won't get caught. However, both international law and the laws of your country forbid assassination. What do you do?
Kant would still say, "Sorry bro, you can't pull that trigger. You have a duty to uphold the laws of your country. The consequences are irrelevant." He's the BYU of philosophers - uphold the rules, consequences be damned. Ross, though, might say that your duty to prevent the massacre of hundreds of thousands, or the death of tens of thousands in a military intervention, outweighs your other duties in this case.

Of course, through all this, utilitarians are like, "Seriously? You bros are still talking about that bullshit? We pushed the fat man and pulled the trigger twenty minutes ago; we've been drinking that whole time. Honestly, we thought you were right behind us."

There are tons of variations on the TP; sometimes the fat guy is the asshole who locked the bros in the trolley to begin with, sometimes he's an unrelated criminal, sometimes the bros are suspected thieves who've lost control of their getaway trolley. Philosophers like Judith Jarvis Thomson and Peter Unger have proposed all sorts of variations - sometimes instead of a trolley, its five bros who need organs, and instead of a fat man, its a homeless guy with compatible organs. These questions challenge the 'moral intuitions' which some philosophers are so fond of invoking by intentionally putting us in surprising settings with similar stakes to what many of us take for granted in everyday scenarios.

Trolley Problems also get used by a new strain of philosophers who call themselves Experimental Philosophers, or ExPhis. "What? I thought if it could be tested, it's not philosophy!" Well, almost all philosophers agree with you there, and many see experimental philosophy as severely misguided. But, well, they're out there, and they'd say, "Bro, if you're going to go on and on about moral intuitions, let's fucking test them. If our intuitions are reliable guides to right and wrong, then shit, son - we might just crack this puppy yet." So they present people with TPs and see what they say. Most people, it turns out, pull the lever in the first TP and refuse to push the fat man in the second. "Obviously, if the first is intuitively right and the second intuitively wrong, then the difference between the two must be where morality lies - it's wrong to actively hurt someone, but not wrong to let harm happen to prevent greater harm. Fuck yeah, we can all go home now." 

Of course, some philosophers hate the Trolley Problems, and are real fucking skeptical about experimental philosophy. Oh, sure, they see the link between the TPs and psychological intuitions just like ExPhis, but rather than embrace experimental methodology they discard psychological intuitions. Elizabeth Anscombe wrote in her essay "Modern Moral Philosophy" that our 'intuitions' about duties and pleasure and everything else are outdated and not well defined at all - what exactly is justice? She did for morality what her mentor Wittgenstein did for language; she pointed out some real fucking important ambiguities that exist in our intuitions about 'justice' and 'duty' and even 'consequence' and 'good' - systems which try to pin down those words take them out of the context in which they developed - thousands of years of moral striving, development and discussion.

So why does the utilitarian answer 'Kill the fat man' seem so repulsive, so intuitively wrong to so many? Because utilitarianism removes important notions from our intuitions, changes them, and then tries to patch them back in and our intuitions go, "What is this shit? That's... that's not what I mean by 'good' or 'pleasure' at all!" And no moral system is free from that. We need an account of intuitions before we can account for morals.

Alasdair MacIntyre has built upon the work of Anscombe by working through the history of ethics and showing how important notions or assumptions just sort of get buried and passed along wholesale. Why would some people push a fat man but not kill a homeless guy for his organs? Because underneath all that are tied up questions of bodily ownership and trust in doctors, pleasure and pain, societal rules, and tons of other shit. When you say, "What? It's just killing one guy to save five," you're not isolating that question from all the other problems - you're just hiding them, so don't act so fucking surprised when people give different answers. MacIntyre believes that Aristotelian virtue ethics, and the improvements that Thomas Aquinas made, are the closest we've yet been - doing the right thing comes from being a good person, not having the best system

So if a deontologist is on a runaway trolley, and a fat utilitarian is walking along the side of the road, what would a virtue ethicist do? We don't have nearly enough information to know.

Wikipedia has a page on Trolley Problems and some of the modifications philosophers have made.

W.D. Ross' The Right and the Good lays out his own deontological system, which he describes as 'intuitionist'. 

Peter Unger makes use of trolley problems and other incredibly creative moral scenarios to make the case for act utilitarianism in his highly controversial but unquestionably challenging Living High and Letting Die: Our Illusion of Innocence.
You can read Elizabeth Anscombe's "Modern Moral Philosophy" online here.

Alasdair MacIntyre lays out a case for a return to virtue ethics in his After Virtue: A Study in Moral Theory and examines the history of ethics and the impact those traditions continue to have on reasoning today in Whose Justice? Which Rationality?


  1. Sup, bro.

    I was recently introduced to your website. I always had an interest in philosophy and its concepts, but the writing is so dense I fall asleep within 20 minutes.

    Way to make important questions obtuse to a mass audience, brolosophers.

    But I've read a few of your writings now, and you make dense concepts immediately accessible. Kudos to you, bro. Do you like beer or wine? Let me know.

  2. Actually, even utilitarians have their disagreements about the trolley problem. The argument is as follows:

    Assume we accept that it's right to throw this fat man in front of the train to save five lives. Yay, five lives are saved, and all else being equal, that's good. But is all else realy equal? After all, now you have set a precedent - that anyone, for reasons having nothing to do with them personally, can be killed simply because it provides benefits to other people.

    Surely, they say, being able to live your life in the comfortable knowledge that you won't be murdered for reasons having nothing to do with you is a good thing. After all, that's why we have a police force - it certainly isn't because we like paying for them, or because everyone wants a prison in their neighborhood.

    If we say that your life is forfeit as soon as a situation arises where your death serves the common good, then we destroy that security for pretty much everyone. Certainly, there is no one person on this planet who produces so much utility that there is no combination of other people for whom their life could not be traded.

    True, we save five lives now, but surely that doesn't outweigh the fact that EVERYONE, FOREVER, must now worry that their life might be the next one to be sacrificed for the greater good.

    Of course, other utilitarians disagree vehemently with this argument, but that's kind of my point - even among utilitarians, the trolley problem is far from cut and dried.

    Also, a side note - this blog keeps losing my comments when I try to post them using my google account... I've had to retype this comment, and I'm pretty sure the first version was much nicer. Same happened the other day and I was in too much of a rush to retype, so everyone lost out on whatever wisdom (or possibly bullshit) I was about to share.

  3. "like utilitarian thinking, but for duties (heh)"

    We don't have this joke in England, because we pronounce "duty" more like "dew-tea". Sorry this post is not a deep philosophical point, but this thought keeps occurring to me.

  4. Excellent post. I've always been fascinated by moral philosophy and this post summed up utilitarianism vs Kant's whatever-the-hell-it-is-ism quite nicely.

  5. For me, reading this, the key distinction that presented itself to me was the one between:

    "What should one do?"
    "What would I do?"

    Do you think these two questions are different? I realise this shifts the discussion a little, but isn't the creation of moral certainty a form of philosophical death?

  6. Thanks for writing this, bro. I wasn't aware that folks were still inventing new versions of the trolley problem. Really well done, as always.

  7. Pureprimitive,

    The framing I'm most comfortable with is "What ought I to do", which combines your two versions. I think it's important for ethics to be a personal question; otherwise, it's too easy for selfish people to tell others to be unselfish. "One should give 20% of their income to charity" has a very different meaning than "I should give 20% of my income to charity".

  8. Thank you ever so much for this! Just the right mix of humor and real educative stuff, i love it.

  9. The answer is "mind your own damn business and do nothing." Period. Self-sufficiency. Thin the herd. Survival of the fittest. Freedom. You know, all those pesky things that scare people shitless.

  10. I would not "save the 5 people" in either case. In both instances, my direct action results in the death of another person, something I find to be morally repulsive. By neglecting to act, it's not that I am "killing 5 people," but rather, they are simply being killed for having been unfortunate enough to be in such a situation. This is, of course, assuming that I am not the one that has set the trolley in motion to begin with. If I were the one that set it in motion, then I would mitigate the viciousness of my negligence by flipping the switch, but I would not murder the fat man. (cf. murder: intentional vs manslaughter: negligent)

  11. I think the answer is to raise taxes on the poor and middle class, while subsidizing already profitable companies, eliminating capital gains taxes, and cutting income taxes on the top tax brackets. Or Freedom - just what mindctrl said.

  12. Clearly the best is action to take is the one that results in the mist bro's coming to harm. That is the only action that betters the world.

  13. @Anonymous: freedom doesn't exist within a society. Your definition of freedom doesn't align with mine.

    All of these problems are a result of the existence of society itself. Society exists because of our intelligence level. I'm reminded of the quote "It has yet to be proven that intelligence has any survival value". Obviously it does on a short term basis, but long term is another beast.

  14. Things I've learned from this post: I'm an Experimental Philosopher. And all this time, I've been considering myself a Psychologist!

    There's gotta be more of a difference, right?

  15. Carrie - Psychologists are interested in the descriptive, rather than prescriptive, side of these questions. That is, they are using them as a means to study the human mind, rather than as a way to study morality. It's like the difference between reading a book to learn about the subject matter, and reading it to learn about the author.

  16. Very pleasant read and well summarised as always. The thing that always baffles me about these morality problems is why it's always a toss up between consequentialism, deontology and virtue ethics.

    The concept of moral particularism seems to be able to answer all of these problems yet it is hardly ever mentioned. Admittedly it is not a system, more of a lack of one to be honest, but like virtue ethics it avoids the problem of inflexibility in an ethical system and is superior to virtue ethics in that it does not have to deal with the vague concept of a "good" human being.

  17. As a layperson, and before I say this I'll straight up admit I haven't followed the further reading links, but PhilBro doesn't even hint at this, so...

    The one answer that none of these scenarios accounts for is the possibility of the bro on the spot sacrificing themselves instead of the bystander.

    This seems like a rather glaring hole in the question. Or rather, in the possible set of answers to the question. Or is scribbling in a third answer and ticking that one just not brosher?

  18. @Degraine - it's easy to hack the problem so that's not an option - say the switch is too far away from the track, or the bro at hand is too light to stop the trolley, so it HAS to be a fat dude.

  19. trolley problems? I have heard of those in terms of a serious note and a much more comical sense as what you suggest. Trolley do have the emergency brake as a safety mechanism.

  20. @Stuart Andrew (First Comment)-

    I think (and anyone correct me if I'm wrong) the distinction lies between act-utilitarians and rule-utilitarians.

    Usually when we say "utilitarians" we think of act-util; the idea that each action is deemed ethical or not based on the consequences of that one action being taken. Among these utilitarians, the issue is cut and dried; one action of pushing the far guy or killing the hobo is not going to have a serious impact on people's future expectations or level of comfort, certainly not enough to outweigh the lives of the 5 bros saved.

    Rule-utilitarians have a different system. They agree that we should care about consequences, but think that instead of looking at one action in isolation, we should think about what would happen if this course of action was universalized as a rule for all future people to follow? What rule would result in the best consequences generally over time? Among this school, it depends on the specific TP. Any of them that involve actually killing someone are probably bad; the organ "donor" one especially has the result of "no one goes to the doctor" which pretty clearly does not result in the best consequences.

  21. Surely this post is quite fun to read. I have only one suggestion, please add pictures when your planning to tell a story. It looks much nicer if it have at least one.
    self storage ottawa

  22. Regardless of flipping the switch or shoving the fat guy, aren't you killing the person in either case?

    And another toss up to the TP would be if the one person (or fat guy) on the rail was the guy who would kill the next Hitler or something. By having him killed sure you'd save 5 lives initially but in the grand scheme of things millions of other people would die, so that kind of thing can't be set by precedent.

  23. Mate, I'm pretty sure torturing a baby is worse than pushing a man into an oncoming train. Depending on your theory, anyway ;/