Monday, November 9, 2015

Four plausibilistic arguments for redirecting the trolley

Start with the standard scenario: trolley speeding towards five innocent strangers, and you can flip a lever to redirect it to a side-track with only one innocent stranger. Here are four arguments each making it plausible that redirecting the trolley is right. [Unfortunately, as you can see from the comments, the first three arguments, at least, are very weak. - ARP]

1. Back and forth: Suppose there is just enough time to flip the lever to redirect and then flip it back--but no more time than that. Assuming one shouldn't redirect, there is nothing wrong with flipping the lever if one has a firm plan to flip it back immediately. After all, nobody is harmed by such a there-and-back movement. The action may seem flippant (pun not intended--I just can't think of a better term), but we could suppose that there is good reason for it (maybe it cures a terrible pain in your arm). But now suppose that you're half-way through this action. You've flipped the lever. The trolley is now speeding towards the one innocent. At this point it is clearly wrong for you to flip it back: everyone agrees that a trolley speeding towards one innocent stranger can't be redirected towards five. This seems paradoxical: the compound action would be permissible, but you'd be obligated to stop half way through. If redirecting the trolley is the right thing to do, we can block the paradox by saying that it's wrong to flip it there and back, because it is your duty to flip it there.

2. Undoing. If you can undo a wrong action, getting everything back to the status quo ante, you probably should. So if it's wrong to flip the lever, then if you've flipped the lever, you probably should flip it back, to undo the action. But flipping it back is uncontroversially wrong. So, probably, flipping the lever isn't wrong.

3. Advice and prevention. Typically, it's permissible to dissuade people who are resolved on a wrong action. But if someone is set on flipping the lever, it's wrong to dissuade her. For once she is resolved on flipping the lever, it is the one person on the side-track who is set to die, and so dissuading the person from flipping the lever redirects death onto the five again. But it's clearly wrong to redirect death onto the five. So, probably, flipping the lever isn't wrong. Similarly, typically one should prevent wrongdoing. But to prevent the flipping of the lever is to redirect the danger onto the five, and that's wrong.

4. Advice and prevention (reversed). The trolley is speeding towards the side-track with one person, and you see someone about to redirect the trolley onto the main track with five persons. Clearly you should try to talk the person out of it. But talking her out of it redirects death from the five innocents to the one. Hence it's right to engage in such redirection. Similarly, it's clear that if you can physical prevent the person from redirecting the trolley onto the main track, you should. But that's redirection of danger from five to one.

12 comments:

Angra Mainyu said...

Lots of trolleys lately!
But it's fun, so I have some alternatives:

4. Alternative: The trolley is speeding towards the side-track with one person, and you see Bob about to redirect the trolley onto the main track also with one person. It's permissible to talk Bob out of that immoral behavior. But that's redirecting death from one to one. It follows that in that case, redirection of death from one to one is permissible. But it was not permissible if Bob had done it. So, some instances of redirection from one to one are permissible, and some instances of redirection from one to one are impermissible.
Maybe the same goes for cases involving redirection from two to one, or five to one. Sometimes they are permissible, and sometimes they are not.

More alternatives:

Push(1,5): Let's say if you flip the switch, a mechanism will push a man onto the tracks. That won't stop the trolley (clearly!), but the onboard computer will detect the impact and activate the brakes, saving the other five (the computer can't detect the five people ahead due to a virus, but can still detect impacts).

1. Modify push(1,5). Now there is a button to cancel. But now suppose that Bob is half-way through this action. You have flipped the lever. The trolley is now speeding towards the one innocent. The "back and forth" argument is committed to the following: either it's okay to flip the switch in Push(1,5), or else, it's okay to remove the man from the tracks, saving him from the trolley even if as a result, the trolley will kill five innocents.

2. Similar to 1.

3. Let's say that Bob decided to flip the switch in Push(1,5). Persuading Bob to change his mind will redirect death onto the five (you wouldn't be using the person as a means; Bob already was set to do that). But that's wrong. So, it seems that by the same reasoning, it follows it's probably permissible to flip the switch in Push (1,5).
Similarly, preventing Bob from flipping the switch is to redirect death from the one to the five.

Alexander R Pruss said...

Good job refuting most of the arguments here!

1. I am inclined to think that pushing a man onto the tracks is borderline permissible in the sense that there are some ways of filling out the case where it's permissible and others where it's not. For instance, I am inclined to think that in a volunteer military, an officer could command a soldier to jump on a grenade, and could even throw a soldier on a grenade. Likewise, the officer could push a soldier onto the track. On the other hand, in a typical civilian scenario, I think it would not be permissible without the permission of the person being pushed. Perhaps the fact that it's a borderline case makes it easier to have what seem to be minor variants get judged differently.

I think the one-one case is also borderline. It might be permissible, for instance, to redirect from one's child to a stranger.

2. However, a case that I don't think is borderline is that of intentionally killing one innocent to save ten. And I think my arguments can be adapted to that case, which now makes me very suspicious of my arguments. For instance, take "back and forth". It seems that there is nothing wrong with feeding a poison to the innocent if you're going to give him an antidote right away. But suppose you've fed him the poison. Now, it seems wrong to give him the antidote. For if you give him the antidote, you thereby cause the death of the ten. I think the answer in this case is that you shouldn't feed him the poison, because if you do, you won't be morally permitted to give him the antidote. And the points adapt to undoing and advice.

I guess I can still maintain that my back-and-forth/undoing/advice arguments show that there is a cost to saying that it's wrong to redirect trolleys, in that we do get violations of principles that are true for the most part, but the above arguments show that these costs are no higher than the costs of deontology.

The one thing that seems to me to survive this problem is Advice and Prevention Reversed. But it only shows that *sometimes* redirection is permissible.

In any case, we've learned some interesting things, such as that sometimes it's wrong to return things to the status quo ante.

Angra Mainyu said...

With respect to your point 1, I agree that in cases involving the military, or one's child, etc., the assessments may very well change (and sometimes do change). In particular, I agree redirection from one's child to a stranger may be permissible. But surely, redirection from five strangers to one's own child is not permissible, so if redirection from five strangers to one stranger is permissible (which I'm not sure about), then introduction of a child changes the assessment. But I was considering cases not involving the military, family members, etc., as a means of mirroring the original arguments.

I agree we've learned some interesting things. :-)

Alexander R Pruss said...

We have different intuitions. I am inclined to think redirection from five strangers to one's child is permissible, if only because most people would feel guilty about living at the expense of five strangers.

Angra Mainyu said...

Okay, maybe we have different intuitions about that, though just to make sure: I'm talking about a trolley case in which the trolley is speeding towards the five strangers, and one's child is on the other track.
By the way, I don't think most people would feel guilty if their parent refrained from redirecting in such a case, while I think that most people (including the other parent, siblings, friends of the child, etc.) would judge that the parent's behavior was immoral.

Okay, maybe we have different intuitions about that, though just to make sure: I'm talking about a trolley case in which the trolley is speeding towards the five strangers, and one's child is on the other track. I meant that introducing one's child, etc., would change the assessment in the original trolley case if redirection in that case is permissible, and it may well change the assessment in other cases. I'm not sure that was clear (maybe you were thinking about another case of redirection?).

By the way, I don't think most people would feel guilty if their parent refrained from redirecting in such a case, or that they would characterize their survival as living at the expense of five strangers.

On the other hand, I think that most people who assess the matter (including the other parent, siblings, friends of the child, etc.) would judge that the parent's behavior was immoral, even if only the closest ones (like the other parent) would probably be willing to publicly blame the parent.

An experiment to test that would be interesting.

Alexander R Pruss said...

Yes, that's the case I'm thinking about.

If we change the numbers, say to 10 or 100 or 1000, it becomes more plausible. But I think it's still true at 5. Maybe even at 2, but I'm not sure.

Alexander R Pruss said...

Let me add this. I would think that most people would hope that their children would have such a moral character that they would willingly sacrifice their lives for five strangers. In particular, they would hope that their children would ask them to redirect the trolley at them if that would save five lives. (But I could be wrong: what people do and do not hope for is an empirical question.)

Angra Mainyu said...

Personally, I still have the intuition that it's impermissible for 10 strangers, or even 100 or 1000 (I'm not thinking about a child they never met, but children they have raised for years).
Incidentally, do you find it intuitive that redirecting from five strangers to one's child is also obligatory?

Regarding the empirical question, I have no guess on this one. But I have the intuition that if the child asks, that may well change things regarding permissibility. It depends on factors such as the child's maturity, the numbers, her reasons for the choice, etc. I also think it's permissible to redirect from one stranger to one stranger if the stranger on the side track asks that one redirects, and there is no further reason not to. So, in general, I think what the person on the side track wants (if the person making the decision knows what that is) has at least sometimes an impact on what's permissible.
By the way, do you think it's also permissible for a parent to redirect from five to her child if the child asks her not to kill her?

Alexander R Pruss said...

I think at 5 it's not obligatory in your child's case. I am thinking that at around 10-20, it's obligatory to sacrifice your own life. I don't know about your child's, though.

But I am really uncomfortable with settling all this with bare intuitions like that.

Angra Mainyu said...

The lack of clear intuitions on the matter would make me more uncomfortable if the cases were more realistic, but that aside, what alternative method would you suggest?
I guess one might propose a general principle, test it against intuitions when the intuitions are clear, and if the principle passes the test, then apply it when the intuitions are unclear. However, it's very hard to find principles like that.

Do you have any such principles in mind (but then again, instead of being convinced by the principle, I would probably end up testing the principle against intuitions!), or an alternative method?

Alexander R Pruss said...

I agree: right now we don't have a good general principle available. (This is also related to my post on parameters.)

There is a general principle available, the utilitarian one. But it yields the wrong results.

One does a little better with the utilitarian principle plus a deontological prohibition on intentional killing. But that still gets the importance of special responsibilities wrong.

By the way, what's your intuition about this case. You're in a rescue boat. There is one person overboard in icy waters. He can swim to your boat and get out, as long as you stay put. But then you get a message that there are five people overboard some distance away. You can rescue them if and only if you leave right away, leaving the one person to drown. Does this differ significantly from trolley cases?

Angra Mainyu said...

If I try to think of a realistic icy waters scenario, then I would (probably! - there are many variables that one assesses intuitively in a real context) reckon the probability of saving more people is better if I wait (his chances are better than any of the others), and then go and try to rescue the rest; it's probable that the delay will not be the difference between five surviving vs. five dying, etc.
So, I would think I should wait. Probably.
But what if the scenario stipulates otherwise? (as it does).
It's difficult to say. Actually, when it comes to scenarios in which not only the setup is artificial, but also the probability one should assign to the different outcomes (i.e., who lives and who dies, etc.), I worry that one's normal intuitive probabilistic assessments when imagining the scenarios might alter what is one actually assessing (e.g., maybe when I'm applying my moral intuitions, I'm not succeeding at using them in coordination with the probability assignments the scenarios demand, but I'm intuitively assigning a different probability to the outcomes). That worry applies to this case, and also to trolley cases, to different degrees.

With that caveat, I would say it's permissible if the man didn't see you waiting for him. But if you're already waiting, he sees you, and you tell him to swim, etc., then I get a (very tentative) impression that abandoning him would be wrong. But then again, that's really uncertain, and my intuitive probabilistic assessment might be misleading me (then again, that might happen in some of the trolley cases too).

Side note: With regard to your post on parameters, I'm not sure we could make much progress there without discussing many other things first, because we have very different views of the whole issue (i.e., I would go with an evolutionary view, whereas yours is theistic and Thomistic). Plus, I don't agree with the fine-tune argument (though it's taken as a given in the post), but that would take just as long to discuss, if not longer, and additionally, I'm not sufficiently familiar with the physics to feel comfortable debating the fine details of the fine tune argument. If I think of something and it's not too long, I'll post it. :-)