Friday, November 6, 2015

Pacifism and trolleys

In the standard trolley case, a runaway trolley is heading towards five innocent people, but can be redirected onto a side-track where there is only one innocent person. I will suppose that the redirection is permissible. This is hard to deny. If redirection here is impermissible, it's impermissible to mass-manufacture vaccines, since mass vaccinations redirect death from a larger number of potentially sick people to a handful of people who die of vaccine-related complications. But vaccinations are good, so redirection is permissible.

I will now suggest that it is difficult to be a pacifist if one agrees with what I just said.

Imagine a variant where the one person on the side-track isn't innocent at all. Indeed, she is the person who set the trolley in motion against the five innocents, and now she's sitting on the side-track, hoping that you'll be unwilling to get your hands dirty by redirecting the trolley at her. Surely the fact that she's a malefactor doesn't make it wrong to direct the trolley at the side-track she's on. So it is permissible to protect innocents by activity that is lethal to malefactors.

This conclusion should make a pacifist already a bit uncomfortable, but perhaps a pacifist can say that it is wrong to protect innocents by violence that is lethal to malefactors. I don't think this can be sustained. For protecting innocents by non-lethal violence is surely permissible. It would be absurd to say a woman can't pepper-spray a rapist. But now modify the trolley case a little more. The malefactor is holding a remote control for the track switch, and will not give it to you unless you violently extract it from her grasp. You also realize that when you violently extract the remote control from the malefactor, in the process of extracting it the button that switches the tracks will be pressed. Thus your violent extraction of the remote will redirect the trolley at the malefactor. Yet surely if it is permissible to do violence to the malefactor and it is permissible to redirect the trolley, it is permissible to redirect the trolley by violence done to the malefactor. But if you do that, you will do a violent action that is lethal to the malefactor.

So it is permissible to protect innocents by violence that is lethal to malefactors. Now, perhaps, it is contended that in the last trolley case, the death of the malefactor is incidental to the violence. But the same is true when one justifies lethal violence in self-defense by means of the Principle of Double Effect. For instance, one can hit an attacker with a club intending to stop the malefactor, with the malefactor's death being an unintended side-effect.

This means that if it is permissible to redirect the trolley, some lethal violence is permissible. What is left untouched, however, by this argument is a pacifism that says that it is always impermissible to intend a malefactor's death. I disagree with that pacifism, too, but this argument doesn't affect it.

9 comments:

Angra Mainyu said...

Alex:

Mass production of vaccines doesn't kill people, so if anything, it would be a parallel between coerced vaccination and the trolleys. However, I think there are a number of potentially morally relevant differences between vaccines and trolleys, for example:

1. In the case of coerced vaccination of children, the risk of dying actually decreases for each child who is inoculated. That seems to justify mandatory vaccination of children even against the wishes of their parents, in order to protect them. But in the case of the trolley, the person on the side track (whether adult or child) is killed, so the risk for that person goes up radically.
2. In the case of coerced vaccination of adults (or children who don't want to be vaccinated, but let's leave that aside to simplify), the risk of dying also decreases.
Now, there is a question of freedom at play: what if the adult does not want to be vaccinated? What's the justification? Shouldn't an adult be allowed to take the risk?
It seems to me coerced vaccination in that case might be justified (or it might not; it depends on the case I think) as a means of protecting other people. So, in this case, some people are having their freedom restricted in order to keep them from spreading dangerous microorganisms to other people. But that does not seem relevantly similar to the case of the innocent bystander, who is no threat to others.

So, a pacifist might reject the argument for the permissibility of the trolley case, while accepting mandatory vaccination campaigns.
That said, I agree that pacifism is mistaken.

Alexander R Pruss said...

The manufacturer is counterfactually responsible: were the vaccines not manufactured, no one would die from them.

The individual risk of dying decreases, yes. But the handful people who die from the vaccination are likely to be different people from the ones who would have died from the disease.

But you do make me think that maybe it's not quite like the trolley case. Let me, though, try to argue that it's similar enough, by thinking about two randomized trolley cases.

Common core: After the trolley passes the junction, a randomizing device kidnaps and place a person on the side track and five people on the main track.

Case 1. It is not causally determined whom the randomizing device chooses to kidnap and place where until after the trolley passes the junction.

Case 2. The randomizing device's choices are causally determined by random events that took place an hour earlier.

In both cases, it seems right to say that everybody's chances of death go down from 5/N to 1/N by switching. However, Case 2 is just like the original trolley case: there are five specific people who will be on the main track and one on the side track. They're just not placed there yet and you don't know who they are. (But your knowing or not knowing who they are is surely not relevant, assuming they're all strangers.)

Case 1 is pretty much like the vaccine case.

It would be weird to think that the fine details of when the randomizing machine has picked people would matter.

Angra Mainyu said...

Alex:

In re: counterfactual responsibility:

The people who vaccines, or more generally medicines for their children, are in that sense responsible too: were the medicines not bought, no child would die from them.
Then one could argue if it were impermissible to redirect the trolley, it would be impermissible to buy any medicines for one's children, as long as there is a nonzero chance of dying - no matter how low.
But that only seems to make the cases even less intuitively analogous.
You might say that in the parents case, there is no analogy due to different intentions: the parents aren't trying to redirect risk. But if so, then the manufacturers may well not be trying to redirect risk, either - they may well be trying to make money by selling vaccines, regardless of who uses the vaccines for what purpose.


In re: randomized cases:

a. I don't know how the randomizing device picks, but I get the impression you're saying the proper probabilistic assessment here is that each person has 1/N chances of being picked. But in case 2, the proper probabilistic assessment depends on the information available to one - at least, given that the event is causally determined already, the information about who gets picked seems to exist already. If so, whether the chances of each person go up or down depend on the info available to the person judging whether they go up or down.

b. It seems to me that in case 1 (and perhaps 2, depending on how it's set up; what's the info available to the relevant people?), the probability of dying does go down for each person. But then again, on the basis of that, one could argue that the randomize trolley cases are relevantly different from the original trolley case.
One way to argue intuitively in support of a morally relevant difference between the original trolley case and case 1 is as follows:
Let OTC(n,k) the original trolley case, but with n people on the first track, and k on the second, and RC1(n,k) just the same but in randomized case 1.
Now, take n=k=1.
Wouldn't you say that pushing the button in RC1(1,1) is either permissible, or at any rate less immoral than pushing the button in OTC(1,1)?
If the answer is "yes", arguably a relevant moral difference may well carry on to other k, n. If not, why not?
One candidate to be such relevant difference is precisely the fact that in OTC(n,k), the risk for some people goes up to almost 1.

Alexander R Pruss said...

1. Given that the chance of a particular child dying from the vaccine is vanishingly small (I don't know the statistics, but I wouldn't be surprised if the danger of death from the vaccine wasn't much bigger than the danger of death from driving the child to the doctor), I think the individual case is different from the case of someone making central decisions. Imagine that there are a million trolley setups, each with a different agent deciding whether to switch the trolley. In each trolley setup, there are five people on the main track. But in only one trolley setup is there anyone on the side-track, and in that case there is only one person there. Every agent controlling a trolley knows all these statistics, but cannot see if there is anyone on the side-track of the trolley that she is controlling. This is an easy case: almost surely no one will die from your switching the trolley, since almost surely the side-track is empty. But suppose that there was a centralized control system and you could simultaneously switch all the trolleys to the side-tracks (or order all the individuals to switch). Then this would be a slightly less easy case, because it would be a standard trolley case, albeit with n=5000000 and k = 1.

2. In the second case, I am assuming that you have no information on who was picked out to be in which group, so your credences assign 1/N to each person's being picked for the side-track. I also assume that you aren't one of the group of N people. I forgot to specify that.

Now I contend that the following shouldn't be morally different.

Case 2. Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and five on the main track, and they will be transported to the tracks as soon as the trolley passes the junction.
Case 3. Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and five on the main track, and placed them on the respective tracks. They are, however, too far away from you for you to be able to discern any differences between them.
Case 4. Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and five on the main track, and placed them on the respective tracks. They are close enough to you that you can see individual differences.

Now, Case 4 is just the original trolley problem. I don't think there is much difference between Case 3 and Case 4, and I think there is no difference between Case 2 and Case 3.

However, your n=k case is really interesting and I don't know what exactly to say. This is like Sophie's choice, but with strangers. I think that for serious reasons, you can switch the trolley from one track to another, but not for trivial reasons. And I do have an intuition that the reasons you need for switching from one track to another don't need to be as compelling when the identities aren't yet picked out. I am not sure what to make of this. Maybe it's something symbolic: switching the trolley in the n=k case, when one's reasons for switching are trivial (e.g., arriving a minute earlier at your destination), seems to trivialize the death of the person on the side-track. And something about proximity seems to matter. In a chaotic universe, quite possibly my scratching my head will affect who lives and who dies in natural disasters in a thousand years. But it's not wrong to scratch my head for a trivial reason when the expected numbers of deaths are equal either way.

Angra Mainyu said...

I think I had misunderstood your "counterfactual responsibility" argument, but in light of your reply, I would say that the mandatory part of the vaccination does not seem to play a role. If it's about probability that some other person might be hurt and "counterfactual responsibility", then we can go with mass production of medicines, as long as very probably, at least someone will die from them. Similar cases would be massive sales of guns to the police (because they'll kill an innocent at some point), or selling medicines (or ambulances) to hospitals, etc.

In re: randomized cases.
Those are interesting cases; I don't have a clear intuition about them. I still get the impression that there might be something about proximity.

How about the following cases, in which the number is reduced from 5 to 2, and we increase proximity another notch?

Case 2(1,2). Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and two on the main track, and they will be transported to the tracks as soon as the trolley passes the junction.
Case 3(1,2). Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and two on the main track, and placed them on the respective tracks. They are, however, too far away from you for you to be able to discern any differences between them.
Case 4(1,2). Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and two on the main track, and placed them on the respective tracks. They are close enough to you that you can see individual differences.
Case 5(1,2). Out of the pool of N strangers, a machine has uniformly randomly picked one to be on the side-track and two on the main track, and placed them on the respective tracks. They are close enough to you that you can see individual differences. Moreover, there are cameras and microphones next to them, and next to you - so, they can see you as well. The two people on the main track are screaming "Please, help us! Don't let the trolley kill us."
The person on the side track is screaming: "Please, don't kill me!"

We can modify the setup to make it more personal than trolleys, perhaps with a pride of hungry lions going after them, or something like that, but I would say the matter is not clear to me.

Angra Mainyu said...

Addition: or maybe the issue is a combination of proximity and whether the probability decreases for everyone?

In any case, there still seem to be at least potentially relevant differences between the trolley case and the cases of mass production or sale of vaccines, medicines, ambulances, guns or bullets to the police, etc.

Alexander R Pruss said...

I don't know if proximity makes much of a difference, absent genuine interpersonal relationships (kinship, friendship, etc.).

Suppose it's a "forced" choice. The trolley is heading down track 0, which has a nuclear bomb on it that will wipe out all life on earth if the trolley hits it. On track 1, there are some people, and on track 2, there are other people. It's wrong not to redirect at this point, given that (a) deontic considerations aren't relevant due to double effect, and (b) if you don't redirect, the people on both tracks die. But now suppose that the people for track 1 have yet to be picked out by the randomizing machine, while the people on track 2 are already there, and you can see their individual differences (but you don't see anything else morally salient; for instance, you can't tell whether or not anybody is a serial killer).

It seems that this is a decision you should make by the numbers, if the numbers are unequal, ignoring the fact that you can see the differences between people on track 2 and that the people on track 1 haven't been picked out yet. If there are more people scheduled to be on track 1, you should redirect to track 2, and vice versa. If the numbers are equal, I would feel most comfortable flipping a coin, to avoid the worry that I am deciding on the basis of a bias that will haunt me for the rest of my life (did I opt for track 2, because they reminded me of myself, say?). In any case, I think there is very little if any preference to send the trolley on the track where the people haven't been picked out yet.

Angra Mainyu said...

Proximity may well not be an issue (I'd say it probably isn't, after further thought), but in that case, the question is whether redirection is permissible at all in Casej(1,1), for j between 2 and 5.
If I understand your point correctly, you hold it's permissible to redirect in 5(1,2). Did I get that right?
With respect to the new scenario you propose, your actions do not increase the risk for anyone: the people on track 1 would be killed by the nuke if you do nothing, so your switching either save the people on track 1, or leave the risk for them unchanged. There is a fix for that, I think. But still, there is the question of whether increasing the risks for specific people is never a relevant factor: it might be in some cases it is (e.g., the cases of equal numbers + redirection), but not in others.
At any rate, it seems to me a pacifist may hold that redirecting when decreasing the risk for all is permissible, but not when the risk to some specific person goes from near zero to near one (at least, not in some cases).
By the way, why not run the argument only on the vaccines (or medicines, or ambulances, etc.)?
After all, if the pacifist holds that even defensive war is impermissible because war will kill some innocents who otherwise wouldn't have been killed, then the same applies to the vaccines, so you can run the arguments on vaccines only, without having to rely on the somewhat more contentious trolley case. And if the pacifist holds that it's always impermissible to intend the death of any person, then the pacifist escapes both the trolley and the vaccines argument.
Granted, the trolley argument can cover cases in which the pacifist holds it's impermissible to do unintended lethal violence on malefactors, but you can make a self-defense argument against that, which seems to rely on stronger intuitions than those involving trolleys.

Alexander R Pruss said...

Yes, I think it's permissible to redirect in 5(1,2).

I was thinking of pacifists who had qualms specifically about killing aggressors.

By the way, the pacifist who holds that it's always impermissible to intend the death of any person can still consistently fight in a war, using means that are not intended to be lethal but that are foreseen to be likely to be lethal.