Trolleyology Done Right


I was thinking about “trolley car problem”s the other day and thought, “I have a friend who has kids, maybe I should give them a trolley car experiment playset for Xmas”*

But I’m not going to have to bother; there’s already one. I am pretty sure that the producers of the kit photographed it deliberately in this set-up. [pj]

See, the guy on the horse – you can tell to him and try to get him to catch and stop the trolley before the tragedy happens. But what if he does not speak philosophy? How do you know for sure?

And, of course there’s the overpass that you can throw the innocent off of, to create havoc and get the trolley operator’s attention:

The kit appears to be missing one thing: a track-switch with a lever.

Look at that smug little smile on the trolley’s face!

I just realized there is potential for cross-meme generation. What would Chuck Norris do with the trolley problem? What would Mr T do? (I think he’d just pity the fool) How many philosophers does it take to slow down a trolley if you pile them up in front of it? If you had a trolley and a switch and Sam Harris was on one track and Jordan Peterson was on the other, how do you arrange the switch so they both get hit?

------ divider ------* Yes, I actually do things like that. One person who shall remain nameless, had a particularly horrible brat that I had to put up with for an afternoon, so for Xmas I gave the brat a lovely semi-automatic paintball gun. Gifts from Marcus are usually single-edged but sometimes they are sharp on all sides.

Comments

  1. Reginald Selkirk says

    The quantum trolley: it goes down all tracks simultaneously, killing a probabilistic number of people in each scenario before the waveform collapses.

  2. komarov says

    What about engineering solutions to the trolley problem, do those count at all? Sure, they can’t fix the trolley problem you’re in right now, but they could mitigate future ones.* Do philosophers appreciate that sort of thing?

    In any case, your trolley problem is factually wrong. It should read, “everyone is in danger…” Some might think this problem too easy, but if the problem itself contradicts reality, how can we learn anything useful from it?

    And I see your footnote puts you squarely in the “blame the parents” camp, complete with revenge. (I’m not judging although kids are probably more complicated than that.)

    *Step 1: Cowcatchers. Minute odds that the victim might merely be maimed. More important is the significantly reduced clean-up cost and time with zero limbs stuck in the undercarriage. The trolley might even be on time, although it’ll still have to be hosed down at the next stop.
    Step 2: Fencing. If philosophers keep sneaking onto your tracks around every crossing and switch, put up some fences. Even cheap wire fencing can resist philosophising for hours at a time, long enough for a trolley operator to notice and call the authorities.

  3. says

    The quantum trolley: it goes down all tracks simultaneously, killing a probabilistic number of people in each scenario before the waveform collapses.

    So, if the trolley kills you, do you still count as an observer?

  4. consciousness razor says

    Build a fence on both sides of the tracks, to prevent ne’er-do-wells from putting their victims on the tracks, and place signs clearly warning people of the dangerous trolleys passing through on the other side. In order for people to safely cross them (which is what the bridge is for, not for throwing people off of it), you could put fences on the bridge too. An alternative to a fenced bridge is building a tunnel underneath for the pedestrian traffic, which may be a better or worse choice in different circumstances.

    That’s a rough idea of what we should do to minimize potential harm and make it a safe and useful trolley stop/crossing. So, it’s more or less what you actually see in properly-designed train stations, because the simple and well-understood measures needed aren’t as controversial or as mysterious as some would have you believe. (Don’t believe the lies that “Big Trolley” is peddling! Sorry, I’m only half joking.)

    In any case, this is a collective decision, one that we’re all responsible for making, when we design/maintain our trolley systems (to use just that one example). We should be doing that stuff before/during/after the scenarios described in typical “trolley problems,” in order to avoid scenarios like those. And if we come up with a particular plan, but it turns out there are still a bunch of injuries, deaths, crimes, traffic problems, accessibility problems, etc., then we should revise our plans to improve such outcomes as much as we can. This is a necessary part of it, because we didn’t and won’t accomplish it with just some abstract a priori principles alone. If things don’t really work out in the world that we actually live in (which may not be what we thought it was ahead of time), the way to determine that is by getting the evidence about this from the world. Armed with that kind of information, we can figure out if there are other choices that won’t lead to such failures.

    It should not be about a last-moment forced choice by a single individual, who may be incapable of (and/or not responsible for) preventing all of the harms that we want to avoid and the goods that we want to promote. If it really does come down to that, that’s probably a result of some bad choices that we as a society made, and those deserve more scrutiny and correction than the actions of a single person, who couldn’t do much about it all by themselves no matter how hard they tried.

  5. consciousness razor says

    Correction:

    who may be incapable of (and/or not responsible for) preventing all of the harms that we want to avoid and [enabling/causing] the goods that we want to promote.

    We obviously don’t want them preventing the latter.

  6. consciousness razor says

    One thing I didn’t mention…. controls for rail switches (if there are any in the vicinity) should not be easily accessible to random passersby. There are still some levers that a person can manually pull next to the tracks, but these days, it’s mostly done with a small electric motor (e.g.) which is remotely-controlled.

  7. cvoinescu says

    consciousness razor @ #6: So you’re saying that they’re not called trolley problems for nothing. Our reaction, if we had a trolley problem, should be to fix it. Or, ideally, set things up to greatly reduce the chance of problems in the first place.

    On the other hand, a fence can both keep innocent bystanders out of the path(s) of the trolley, and Sam Harris and Jordan Peterson in. (Not sure how I feel about the cowcatcher, given that, by now, Jordan Peterson must be at least 30% beef by weight.)

  8. says

    So many people misunderstand “the” trolley problem. The interesting part is when someone’s answers to essentially the same choice differ. If you’d pull the switch but wouldn’t push the fat guy off the bridge (or vice versa!). The reasons for the difference are something to talk about. But it’s been memed to death. Such a shame.

  9. consciousness razor says

    On the other hand, a fence can both keep innocent bystanders out of the path(s) of the trolley, and Sam Harris and Jordan Peterson in.

    It’s a never-ending task isn’t it? Always the little details.

    Put some doors on the fence here and there, so whenever it may be necessary, Harris and/or Peterson can be safely escorted to a psychiatric hospital, rehabilitation center, or jail. (Or possibly set free — it kind of depends on how they got there in the first place.)

    Assuming there’s enough space between the rails and the fence, a trolley could also simply open its doors to let them ride it to the next stop. However, it may have to be supplied first with some placebos (not actual tranquilizers) to entice them onto the train. If Harris happens to possess a ticking time-bomb (he does after all fit the profile), then someone will probably need to fetch the comfy chair.

  10. says

    Dave W@#10:
    So many people misunderstand “the” trolley problem. The interesting part is when someone’s answers to essentially the same choice differ. If you’d pull the switch but wouldn’t push the fat guy off the bridge (or vice versa!). The reasons for the difference are something to talk about. But it’s been memed to death. Such a shame.

    I’m going to disagree with you, there. It’s a pointless “thought experiment” that reveals nothing, because it’s about a ridiculously unrealistic scenario that nobody will ever encounter. It’s a long-term shame upon philosophy that trollyollogy was not laughed out of the academe for being completely pointless – it serves as nothing but a good stick for real sciences to beat philosophers with. Do they seriously want to pretend that we can learn something about how humans decide what is right, using such a flawed instrument? Gah!

    It has been memed to death because it only rises to the level of meme-worthy.

  11. says

    consciousness razor@#8:
    There are still some levers that a person can manually pull next to the tracks, but these days, it’s mostly done with a small electric motor (e.g.) which is remotely-controlled.

    Oh, great another life-impacting opportunity to install SCADA controls insecurely! There’s going to be an “internet of things” camera that feeds an image back to an outsourced :”operations center” where someone is sitting with a Windows PC and some software that sends switching commands to a programmable logic controller that has no default password set. Naturally, for cost reasons, the data will be sent across the internet. That way any hacker who checks SHODAN will be able to flip the switch on the trolley remotely without the system even keeping good diagnostic logs. Eventually the trolley company will hire a penetration tester, who will demonstrate the weakness, and they’ll spend a small fortune re-doing the whole system with better security – but they’ll still outsource the operations center and they won’t ask if the ops staff are adequately trained and they won’t screen out the philosophers. Then, one day, a philosopher will find the control panel and wreak bloody havoc and everyone will wave their hands in the air and declaim “how did this happen?” Then they’ll hire someone like me to unravel things and write a report, which will be ignored and the whole process restarts.

  12. brucegee1962 says

    I think the reason the trolley problem continues to be popular is that it’s an encapsulation of real-world ethical problems, so we can analyze the dilemmas we actually face in real life.
    In real life, instead of trolley switch, let’s say you are voting on the budget of a large industrialized country. If you push button A, you can save the lives of 50,000 children in a foreign country by sending them vaccines for easily-preventable childhood diseases. If you push button B, you can improve the educational system of your own country so that 10,000 children will have access to books that will make them more productive citizens in the future. If you push button C, you can buy a bomber that will flatten a few villages in a distant place you will never visit, and will make your friends in the military business very happy.

    Well, not much of a dilemma, is it? C, obviously. This is the real world — ethics has nothing to do with it.

  13. billseymour says

    consciousness razor @8:  all switches need to be manually operable so that traffic can still move when, not if, the automated system fails. The levers tend to be locked; but since they need to be operable by J. Random Railroader, all the locks have the same key. You and I don’t have a copy of that key it’s believed.

  14. cvoinescu says

    Marcus @ #13: Oh, great another life-impacting opportunity to install SCADA controls insecurely!
    You’re probably wrong about this. They are very simple systems, with separate cables going to each track switch, and no software at all. They have actuators and position confirmation switches that also interlock with the signals. I’m not sure how it’s done in the brand-new systems (do they even build new rail anymore?), but anything older uses relay logic, with critical parts duplicated and cross-checking. Among other things, they detect current through the bulbs in the signals, so that if a bulb breaks they know the signal is inoperative right away. For the switches, the in-position detectors are sometimes duplicated on the opposite side from the actuator, so that if the linkage bar gets bent, they fail to engage and the signal remains at danger. A lot of industries could learn a lot from how railroads do it (in the parts of the world where they’re regulated properly).

Leave a Reply