Last night I saw The Road, a truly bleak post-apocalyptic movie. [Warning: Minor spoilers.] As I watched, I realized that I was witnessing a mighty counter-example to my views on the propriety of statistical discrimination. In the movie, about 80% of the people seem to be murderous cannibals. This is common knowledge. As a result, everyone is tempted to shoot first and ask questions later. After all, even if two perfectly innocent human beings bump into each other, each can rationally assume the worst about the other.
Notice the tipping point. Once p(a stranger is a murderous cannibal) gets high enough, morally confident statistical discrimination spirals out of control. Even if the stranger down the road isn’t a cannibal, he has a strong motive to preemptively murder you – which gives you a strong motive to preemptively murder him.
Given a grim starting point, statistical discrimination can quickly lead to mass human extinction even if decent folk adhere to a “beyond a reasonable doubt” standard to convict. For example, if your threshold is 99% certainty, everyone in a world of 99% murderous cannibals has a license to shoot on sight. And if, like me, you accept the preponderance standard for conviction, the situation’s even worse.
My best reply, I’m afraid: Just because you probably have a right to do X, doesn’t mean that it’s right to exercise that right. Non-cannibals in The Road usually have a right to shoot first, but it’s morally wrong to do so when they can gather information at moderate cost, or simply avoid conflict by hiding or running.
Got a better story?
READER COMMENTS
Liam
Nov 29 2010 at 1:07am
You just saw The Road now? That was released last year. You need to get out a little more. Vigo did a good job but the kid’s reaction was totally unbelievable. He was raised after the apocalypse and yet was completely unable to fend for himself in any way.
Better example is The Road Warrior. Women, children and seniors were given menial positions except for the little kid with the razor boomerang. Though, the Director did try to overcome this in Beyond the Thunderdome when Tina Turner ruled Bartertown until she fell to the economic pressures exerted by a dwarf and a mentally handicapped giant collectively called Master Blaster.
Hyena
Nov 29 2010 at 1:09am
The view you expressed earlier was that “statistical discrimination” is both inevitable and useful while being largely unaddressable ex post insofar as it is unjust. So I’m not sure how a counterexample could be drawn in this way.
You noted at the end of your post:
…and…
So it seems like you had this situation well-in-hand before The Road.
Nick
Nov 29 2010 at 2:01am
I’m not sure that the issue you raise with statistical discrimination is really an issue. After all, in a world where 80% of human beings are cannibals the survival of the human race is in question whether people choose to shoot or eat each other. In such a situation shooting first might give the non-cannibals a greater advantage, even if the occasional non-cannibals shoot each other.
Doc Merlin
Nov 29 2010 at 6:53am
Pfft I call shenanigans! There have been societies in history where far more than 80% of the people are murderous canibals, and it doesn’t devolve into a free-for-all. People form small groups that agree they won’t eat each other and then build from there. The same can be seen in a game of Assassins. People end up forming gangs which agree not to murder each other in exchange for help murdering everyone else.
Henry
Nov 29 2010 at 7:53am
Firstly, if there is any problem with the above scenario, it’s that shooters do not consider the external costs of their shootings, not that they statistically discriminate. If each person received a reward for each cannibal they killed and a punishment for each innocent person they killed, they would be motivated to employ moderate cost methods to identify people as innocent or cannibal. Of course, government wouldn’t exist to provide these incentives in such a scenario, but efficiency could still exist as one’s ethical code.
Secondly, most cases of statistical discrimination in real life do not involve external costs at all, but denial of external benefits. (You could count external costs if you considered psychic harm, but you can use that to justify absolutely anything, so I’m sceptical of its use in cost-benefit analysis). Furthermore, they usually don’t have efficiency consequences because their external effects tend to be solely distributive. The benefits of a job may be denied to one person, but are usually given to someone else instead. You would only want laws or social norms restricting statistical discrimination/incentivising information seeking if a) you consider the distributive effects sufficiently negative to warrant a response and b) if finding information is cheaper than redistributing after the fact.
Liam
Nov 29 2010 at 9:58am
Henry brings up an interesting point when he said, “If each person received a reward for each cannibal they killed and a punishment for each innocent person they killed, they would be motivated to employ moderate cost methods to identify people as innocent or cannibal.”
But what if the canibals are also zombies (as zombies tend to the consumption of human flesh). Zombies tend to be very cheap labor and are capable of coordinated group activity. Sure then tend to be slow and favour unionism, but they work long hours and have extensive background in Human Resources, which is another group that tends to demand more brains.
Kurbla
Nov 29 2010 at 11:27am
Doc is right. Tribe is the solution, and it can solve even more difficult problems. Even if all people (including me) are cannibal murderers I am still interested in searching for people who can delay their cannibal murder impulses long enough so we can hunt other people efficiently. Eventually, we might make civilization just like this one we have, except that we’ll breed people for food, and pretend we are completely innocent.
We actually already live in that civilization, just we call the kind of people we breed for food “pigs” and we quite well pretend that there is no moral problem with that.
Steve Roth
Nov 29 2010 at 11:56am
Speaking of statistical discrimination:
http://www.angrybearblog.com/2010/11/proposed-bet-for-professors-bryan.html
Doug
Nov 29 2010 at 12:35pm
I think evolution already solved this problem in the formation of social primate groups and it didn’t have anything to do with Locke or federal courts…
Philippe
Nov 29 2010 at 12:46pm
This “first mover advantage” you describe (or to the extent I understand it) is also prevalent in financial markets. Maybe one may choose an even stronger word than “prevalent”, such as “defining feature”. If prob(sovereign is not willing to / capable of repaying bond debt) is sufficiently high, people will only be willing to lend at higher rates (prohibitively high in the extreme) to lend to this country, in turn increasing this probability. That is, the market can become highly illiquid / inexistent due to high prior beliefs of default.
Is that a valid example for the same concept in another context?
Isegoria
Nov 29 2010 at 2:50pm
A true crisis inverts many of our moral intuitions.
When the number of humans suddenly outstrips food production, how bad is homicide? Killing people now may reduce the number of even more painful deaths in the near future.
For instance, after the limited nuclear strike or asteroid collision that sets off our apocalyptic scenario, our local community can expect a few years of crop failures, but they have enough canned food and dry grain to feed all 1,200 people for one month.
Should they feed everyone for one month, and then starve en masse? Should they draw lots and euthanize 1,150 people, so that 50 can live through two years? Is that practical? Should they send all the young men to seize food from any nearby communities? That’s not so different from drawing lots — some die, and the survivors get more food.
It’s a different world from the one we expect, where we presuppose law and order and nutritional plenty.
(Imagine trying to bootstrap society after such a cataclysm.)
Silas Barta
Nov 29 2010 at 2:54pm
A few points:
1) Someone willing to kill everyone they encounter no longer counts, morally, as a human. (How do they differ from a man-eating lion or a psychopath?) So, don’t do that if you want others to still respect your moral worth.
2) One easy countermeasure against mass cannibals is to attach high explosives to your body that you can activate (or which trigger upon cessation of your heartbeat). This would char you and make you useless as meat to cannibals.
3) Surely there’s some intermediate ground in which you use non-lethal weapons against attackers and “test” unknown humans by their reaction to (hopefully already dead) body parts?
4) Per all of the above, your knowledge about the likelihood of someone being a cannibal requires you to use that information as a good Bayesian, but not to kill.
david (not henderson)
Nov 29 2010 at 5:05pm
“Just because you probably have a right to do X, doesn’t mean that it’s right to exercise that right.”
Seems tenuous. Looking for a way to suppress movie-induced second thoughts about the morality of pre-emptive war?
Bob Murphy
Nov 29 2010 at 9:33pm
Bryan wrote:
Non-cannibals in The Road usually have a right to shoot first, but it’s morally wrong to do so when they can gather information at moderate cost, or simply avoid conflict by hiding or running.
I understand in general that you could have a right to do something, but it would still be immoral. (E.g. many libertarians think using cocaine falls into this category.)
But I’m not sure I see it in this context. If your life isn’t in immediate danger–because you can hide or run–then how do you have the right to kill somebody? The only way you can have a right to kill someone is if it’s self-defense, which it isn’t per Bryan’s assumptions.
Doesn’t the rule, “You can only kill someone if there is no other way to protect yourself” still handle this?
Comments are closed.