Monday, March 10, 2008

Probabilistic ethics

So, I was looking at these online games from The Philosophy Magazine. In particular, I want to discuss the game "Taboo". They ask you some questions about the morality of things that are taboo. You can be assured that involves things like eating cats, incest, and so forth. The basic idea is that people usually don't actually think about these things rationally. They just have an emotional reaction to the situation, and then they come up with some after-the-fact rationalization for the judgment. Afterwards, they often claim they had been using that reasoning all along.

So here's my rationalization. Of course, I'm going to claim that I was thinking about this particular line of reasoning all along. ;-)

I claimed that an action cannot be wrong if "it is entirely private and no-one, not even the person doing the act, is harmed by it at all." Later, it gives some situations that sound bad, but in which no one was harmed, not even emotionally. Well, I should have qualified my previous answer on the quiz, but it was difficult to do when the only choices are yes and no. An act cannot be wrong if there is no possibility of harming anyone.

To elaborate, consider a hypothetical situation. Let's say there is a lottery. Like all lotteries, the odds are not in your favor. Let's also assume that the lottery does not donate a few pennies to the education system like real lotteries do. All profits and losses go directly to Al Qaeda (or you can insert some other unequivocally evil cause here). Lastly, assume that your participation does not affect anyone else's chance of winning.

Obviously, participation in this lottery is wrong. But what if you ignore this fact and play anyway? What if you beat the one to a billion odds, and win, thus taking funds directly from Al Qaeda? Does the outcome make your previous action ethically correct?

I would argue no. The right/wrongness of a particular decision should not be determined by the outcome, but by all the possible outcomes and their respective probabilities. These probabilities are determined by the decider's knowledge (or by what the decider should know). The whole purpose of labeling past actions right or wrong is to better choose between future actions. So the label should not depend on the outcome of that particular situation, but on the possible outcomes of a similar future situation. So even though you won the lottery, your participation was wrong because you would probably not win if you participated again.

This is why most traffic rules are illegal to break, even when no one actually gets hurt.

Of course, such an ethical model would never work out in fiction. Think about all those heroes who take stupid risks, but are saved by some lucky coincidence. Would they be heroes in real life?

Other implications I like: Ethicists should study probability theory? Mathematicians are naturally ethical?

I leave it as an exercise to the reader to imagine how this relates to taboo.

So... that was easy to rationalize. For my encore, I will prove black is white so you no longer have to worry about those pesky zebra crossings.


Anonymous said...

Then of course there's mathematical morality (although in this case it has nothing to do with ethics…)

In general I'd agree with your assessment. There's a couple of things that make it hard to implement in practice, though--most significantly the difficulty of pinning down reasonable probabilities for the possible outcomes of an action in all but a select few cases (such as a lottery).

On the topic of rationalization, some rather interesting questions are raised (for me, at least). I for one would be adverse to eating cats (well, if I had to I'd eat one if it was already dead through no fault of my own.) I don't mind that a cow gets killed so I can eat it (when I don't think about it too much), but I'd be much more reluctant to allow a cat to be killed so I can eat it. My rationalization would be that in some sense I perceive cats to be "smarter" or to have more "consciousness" than a cow, but that's probably more due to the way I've interacted with cats than any actual difference between the two. No one's yet figured out a completely acceptable measure of intelligence or consciousness, so I can't even fairly say that a cat is more conscious than a cow, or that that's a good reason not to eat one (especially if I were starving).

Anonymous said...

"There's no way to rule innocent men. The only power government has is the power to crack down on criminals. When there aren't enough criminals, one makes them. One declares so many things to be a crime that it becomes impossible for men to live without breaking laws." Ayn Rand.

Every citizen a criminal, every jackbooted State enforcer a saint - Homeland Severity..

DeralterChemiker said...

There is a practical reason why we eat cows, sheep, pigs, chickens, deer, etc. These are all herbivores, while cats of all sorts, dogs and wolves, etc., are carnivores. The ratio of carnivores to their usual prey is pretty well fixed in nature; I believe it is something on the order of 1:10. Therefore we can raise and eat large numbers of herbivores, whereas raising carnivores for food would be very difficult. Of course, the animals we eat have generally been prescribed by religions, and whether we believe those religions or not, we are all affected by them psychologically. Thus, few of us eat horses, although they are herbivores; and depending on your religion, you may also avoid pigs. I have this suspicion that most of the laws of major religions simply codified what were considered practical matters at the time that the roots of these religions were established. Today some of these laws may seem nonsensical, and ethnic groups that developed without these practical and religious root beliefs don't feel bound by them. Nevertheless, we are all affected by the beliefs of the people around us.