I don’t believe in libertarian free will; I’m a compatibilist at most. But watching the Sapolsky-Huemer debate on free will, Huemer brought up a moral consequence that I’d never thought about before. It starts with a classic argument against utilitarianism:
The Mob. An angry mob is convinced that Charlie did a heinous crime; if unsatisfied they are likely to turn violent and kill many people. The sheriff and Charlie are the only people who know Charlie is innocent—it really looks like Charlie did the crime. Should the sheriff hang Charlie to placate the mob?
And there are classic utilitarian responses: if legal authorities reasoned like this, it would create a culture of suspicion, people would overreach and start killing anyone for a “net good”, etc, we’re running on corrupted hardware, etc. Objectors can try to isolate the case by assuming the sheriff is going to die tomorrow and the decision will remain secret and all of that, and responders can argue about the appropriate intuitions for those cases. Let’s set all of that aside for now.
What Huemer added, that was new to me, is: even utilitarians get the sense that punishing Charlie for something he didn’t do is unfair. Similarly, it would be unfair to punish Charlie if he had a spontaneous out-of-the-blue epileptic seizure that killed someone. But the “unfairness” in both cases only makes sense if Charlie is said to have some kind of free will. Otherwise, the real murderer had no more responsibility for the act than Charlie did. So if the only reason for punishment is avoiding future harm, as determinists like Sapolsky believes, why should it be any wronger to imprison an innocent person for the same effect?
Regardless of whether there’s free will, I think it probably is wronger to imprison someone who hasn’t done anything wrong than to imprison someone who premeditated and did it (even if that premeditation was determined). But why should that be, if retributivism is false?
I think I have an answer now. Consider:
The Bizarro Mob: in this world, people have a different sense of justice than we do. No one thinks it is particularly unjust that an innocent person be punished for another’s crime, so long as it is effective at preventing further harms. Even Charlie agrees! While Charlie obviously personally prefers not to be executed, he would accept that this might be the right thing to do in the abstract. In this way, Charlie is like the loner on the left trolley track: he might agree that the right thing to do is to swerve and hit him, even though he really doesn’t want to get hit by a trolley.
Now, in The Bizarro Mob, I ask: are the people wrong?
That is, do they have some moral reason of “justice” to not execute Charlie, even if it really will avoid more bloodshed? Is there some means by which Charlie is incorrect to think that execution is the right thing to do?
There are two ways that the Bizarro people could be wrong:
They don’t fully understand the preferences of everyone involved and/or don’t respect preferences in proportion to their intensity. For instance, if the mob doesn’t appreciate how bad it is for Charlie if Charlie gets executed, or if Charlie believes in some moral authority that doesn’t actually exist.
They understand all the facts of preferences and consider them proportionally, but it is somehow objectively wrong, independent of what everyone wants, to punish Charlie.
I don’t think #2 is possible. Why? The true test of whether something is good for or bad for a person is whether, operating on perfect knowledge and free from external constraints, they would want that outcome or not.1 Nobody in this world cares whether they are treated “unjustly” by our standard; therefore “injustice” is not bad for anybody.2 But if it isn’t bad for anybody, it isn’t bad.
So what would make it wrong to punish Charlie in our world? The fact that we do value justice! If the mob actually wants revenge, then they actually care that they get the right guy—getting the wrong guy violates their preferences, even though they think their preferences have been satisfied. Similarly, you and I and Charlie and everyone else values that people not be punished for things they didn’t do. So it would be bad for a lot of people if Charlie were actually executed.
This is preference consequentialism, not moral relativism. In a society where everyone thinks brutal human slavery is okay—even the slaves!—that would not make slavery okay in my framework. Why? Because humans really, really, really don’t want to be beaten, tortured, coerced, and treated like property, and this is a much stronger desire than the greed of slave-owners. Anyone who thus believes that slavery is okay is therefore mistaken in the type #1 sense: they are failing to take into account all the preferences on the table. What is good for us depends on our preferences. But the fact that what is good for us depends on our preferences is not relative.
I admit my view has its own counterintuitive implications. For instance, it could be wrong for me to get gay married if enough other people preferred strongly that I didn’t—so long as those other people didn’t feel that this was a ridiculous moral standard. However, it’s possible the repulsive feeling I get here is just the strength of my own preference for cute twinks independence and autonomy showing through. Thoughts?
I draw inspiration from the wonderful Silas Abrahamson’s Should I Care Whether my Life is Objectively Bad? which is what started me on this preference-consequentialist track. Thanks, Silas!
Injustice in se. It is bad for Charlie that he be treated unjustly in this case but only because injustice means he will be executed, and it’s bad for Charlie to be executed. But that would not change even if Charlie actually did the crime.
A lot of this ethics stuff is new to me ... preferential consequentialism sounds so much better than the hedonist variety, thanks for explaining it!
Is there more to be said about how weighing preferences against each other works? I'm worried about a society where the slaves don't mind too much that they are slaves, but the slave owners are absolutely desperate to keep their slaves. Seems like slavery would still be wrong ....
I think your example is perfectly consistent with at least my intuition, and I agree with your verdict that if nobody cared about justice it would not matter. The problem with preference utilitarianism really gets off the ground if you combine it with the idea you mentioned in a comment on the previous post that the preferences of dead people also count, even if they are already dead because then it’s entirely possible that things like same sex marriage are actively wrong in the actual world, not just in hypotheticals and that sure feels weird. Like you said, we could be wrong about that, but the possibility that this is actually bad does sure strike me as a strange conclusion. Maybe dead peoples utility function should be dominated by the desire for resurrection or something. Although given how scientifically challenging that would be I’m doubtful. Maybe I’m letting my political sensibilities getting the way of a good ethical theory. After all, it does feel strange to start caring about what the implications of a theory are in the actual world given that morality is supposed to be equally applicable to all possible worlds you find yourself in so perhaps I am only post facto trying to get out of a reasonable possibility.