This is Berk
by Benjamin Studebaker
I have invented a moral conundrum which I believe makes a strong point about moral philosophy. I wish to share it.
You are the chief of a Viking village called Berk. Berk has 1,000 Viking citizens. Berk is twelve days north of hopeless and a few degrees south of freezing to death, but your people are hardy and they manage. One day, a sentry reports to you. He informs you that a number of dragons are coming–they are going to attack the village! As chief, it is your task to defend your people and uphold just and good moral principles. You are also a master of military strategy and tactics. Consequently, you know the following:
- Vikings that fight dragons have a high mortality rate. The Vikings who are sent out to fight the dragons have a 75% chance of dying. This increases to 100% if the dragon-fighting Vikings are unable to stop the dragons from reaching the village.
- If the dragons succeed in reaching Berk, they will burn it to the ground, killing 30% of all the non-dragon fighting Vikings. This is a catastrophe.
- In order to successfully prevent the dragons from reaching Berk, 200 Vikings must be sent out to fight the dragons, of which 150 are expected to die. Sending out more than 200 Vikings makes no difference.
If fewer than 200 Vikings are sent out to fight the dragons, they will all die, 30% of the remaining Vikings will be killed, and Berk will burn. You synthesize this together:
- If you do not send out any Vikings to fight the dragons, 300 Vikings will die and Berk will be burned down.
- If you send out less than 200 Vikings to fight the dragons, all of those Vikings will die along with 30% of the remainder, and Berk will be burned down. This is worse than sending out no Vikings at all. The dragon-fighting Vikings die in vain.
- If however you send 200 Vikings to fight the dragons, only 150 Vikings will die, and Berk will be spared. This is clearly the best outcome.
You call for volunteers to fight the dragons. Your Vikings are brave and valiant, but nonetheless, dragons are very frightening, and most Viking citizens know that their chances of surviving as dragon-fighting Vikings are low. As a result, 150 public-spirited Vikings volunteer to fight. You know this is not quite enough to save Berk, but in order to get 50 additional Vikings to fight dragons, you will have to conscript them, coercing them to fight dragons against their wills and robbing them of autonomous choice. All Vikings you choose will be harmed–their chance of dying individually will be higher as dragon-fighting Vikings, even in victory, than it would be if they stayed behind and took their chances surviving the burning of the village. You have four options:
- You can refuse to let the 150 public-spirited Vikings fight, allowing Berk to be burned and resulting in 300 Viking deaths.
- You can let the 150 public-spirited Vikings to die in vain, allowing Berk to be burned and resulting in 405 Viking deaths.
- You can draft an additional 50 Vikings to fight the dragons via a lottery, saving Berk and resulting in 150 Viking deaths.
- A minority of the population of Berk is Christian. The Christians are widely disliked by the other Vikings. You can force 50 Christian Vikings to fight the dragons, saving Berk and resulting in 150 Viking deaths, a disproportionate number of which will be Christian Vikings.
Do you know which option you would choose?
A lot of moral theorists think that autonomy is very important. Autonomy is the principle of allowing people to make their own choices and refraining from using them for the benefit of others. If we ask people to do something as a society, autonomy requires that we be able to justify to them why consenting to doing what we ask is good for them. Some proponents of autonomy go further and oppose paternalism, the process of restricting the autonomy of someone for his own benefit. If autonomy is what matters, option 1 is wrong because it robs Vikings of the choice to fight and is a kind of paternalism. Option 3 is wrong because it uses some Vikings for the benefit of others–the conscripted Vikings are undoubtedly worse off as dragon-fighters, and we are using them in order to benefit others. Option 4 is like option 3, only it’s also unfair to Christian Vikings. Under a moral theory that places autonomy at the center, the best option is option 2, and options 3 and 4 are worse than option 1, because option 1 is merely paternalistic, while options 3 and 4 use others.
As far as I’m concerned, this proves that placing autonomy at the center of one’s moral theory is a mistake, because option 2 is clearly repugnant insofar as it is the option that results in the greatest amount of harm.
I think that option 3 is the best option. I agree that option 3 uses some Vikings for the benefit of others, but I think this use is acceptable because, unlike option 4, it is not exploitative. What do I mean by that? I offer a distinction between using people and exploiting them.
When we use people, we harm some of them for the benefit of other people. But if we do it through a procedure that recognizes the equal status of the interests of all of the citizens in our community, it is fair use, and is morally acceptable. In option 3, all of the Vikings have the same chance of being chosen for this duty. We do not single out some Vikings in a way that makes them second class citizens. Before the lottery is conducted, when all Vikings are in ignorance of whether or not they will individually be selected, all reasonable Vikings would agree that the people of Berk are collectively better off having the lottery than not having it.
In contrast, option 4 is exploitative. It treats the interests of Christian Vikings as being of inferior value. It does not uphold the state’s obligation to take seriously the equal standing of the interests of all of its people.
Often utilitarian moral theorists are accused of “using people” or “treating people as means rather than ends”. I contend that these behaviors are not wrong intrinsically; they are only wrong when the procedure by which we use people is exploitative, when it treats some citizens as being of greater or lesser moral importance than others. Critics of utilitarianism often conflate justifiable use with exploitation. When they act on those beliefs, they prevent society from using people in justifiable ways, resulting in harm. That is not okay.
Benjamin Studebaker 
Here’s my problem with all of this: You have a demonstrated preference problem. 85% of your Vikings implicitly prefer not to fight the dragon.
Let me anticipate a commons objection: you could point out that your Vikings are death minimizers; they’d rather have fewer deaths than more and this is their only consideration. However, only 15% (instead of the needed 20%) are willing to voluntarily put that into action. At that point, I’d agree it would be defensible for a fair conscription.
But to take that route is to result in a loss of generality. The essential problem in the Prisoners dilemma is that the prisoners would like to bind themselves to mutual cooperation, but can’t. Without stipulating that the Vikings would prefer bind themselves to a draft if they could, it is impossible to say if the 85% of Vikings holding out would prefer to risk getting drafted over getting torched in the inevitable attack.
Now, under your criteria, the odds of getting drafted and burnt to a crisp are 12.75%; the odds of becoming dragon food are 30% if you wait. At this spread, I think submitting to the draft is an eminently sensible move. But this is because I /personally/ would rather lower my odds of dying; I know I have resolutely anti-draft compatriots. To reduce our Vikings to death-minimizers is an interesting thought experiment; to reduce real people to that is to deny that people do not value things neatly along the rules of philosophical thought experiments.
And this is my problem with practical utilitarianism. It quickly becomes impossible to practically to determine and subsequently calculate the utility in a society.
None of the numbers necessarily have to hold for the experiment to work, the various characters just have to perceive their odds of survival and their various utilities in something like this way. Utility is always very difficult to calculate in practice–for some of the Vikings, honor and duty provide more utility than survival, unless they are fundamentally mistaken about their interests. All that really matters is that some Vikings want to fight, but clearly not enough, and that the rest are resistant to joining, but the average Viking’s odds are better if others are made to. This can be a matter of qualitative observation rather than quantitative figures.
The point is not to say that we can always calculate outcomes with accuracy and precision–we almost never can–but to say that whatever our best estimate of the outcomes is, that should form the basis of our moral judgements, and to argue the fundamentally arbitrary nature of alternative models.
Right, but your rebuttal misses my point exactly. First, not all Vikings prefer better odds. (Or, if they do, you lose generality.)
The problem is that “best model” is itself a difficult proposition. Indeed, your best model totally glosses over why 85% of Vikings did not demonstrate a preference for fighting.
In the case, I presume that 85% of Vikings do not demonstrate a preference for fighting because they are not as public spirited as the other 15% and consequently seek to maximize their survival odds. This doesn’t seem to be an unreasonable presumption. We might say that 15% of Vikings are irrational heroes, or that 85% of Vikings are moral cowards. Given that each one of them wants to survive but can only maximize his odds if a minority fights Vikings, the draft is the most reasonable solution because it creates the best odds for the average Viking.
I don’t think this is an unrealistic scenario, if that’s what you’re arguing (aside from the whole “dragons” thing).
In that case, I accept your conclusion since, by construction, 85% of vikings would rather risk being drafted than risk dying.
What about in a society where the issue was split? Or, if the the 85% were anti-draft and willing to accept higher odds of dying to stand by that point?
And, further, if the chiefs had no way of knowing /why/ their citizenry had spurned the called to arms. (And no time to make an accurate assessment before the window to fight the dragons closed.)
I disagree with preference utilitarianism, I don’t think that just because the Vikings express a preference for avoiding the draft even though it increases their risk of death that this preference expresses what is in their rightly considered interest. I haven’t presupposed that all Vikings would agree to the draft, I have only attempted to show why all rational Vikings should agree. It may nonetheless be the case that many or most Vikings disapprove of the draft, but these Vikings are mistaken–they are better off with the draft than they are without it.
The chief doesn’t really need to know the citizens’ reasons for spurning the call, he only needs to have a justifiable view of what is in their well considered interest. In this case, I take the chief’s well considered view to be that Vikings are much better off living than they are dying, even if ensuring survival requires coercion.
This is a fairly subtle sleight of hand. You are merely substituting the chief’s preferences for the village’s (ie, the sum of the Vikings.) And, if we get meta about this, your preferences since you’re the author arguing the chief is correct.
And therein lies your problem: this is preference utilitarianism; it’s just your preferences.
There’s an epistemic difference between a preference and a justifiable, well-considered interest. Everyone has preferences, but most people’s preferences do not meet a sufficient epistemic standard to be deemed well-considered interests. The chief’s view of what’s best for the village is not just another preference, it’s a preference that is supportable through moral argument. The chief is chief in the first place because he is deemed to be the best at sorting out the well-considered interests from misguided personal preferences.
[…] primary feature of this case is that it looks precisely like the sort of case in which it is almost universally considered reasonable to impose a military draft, the case in […]