Ends Don't Justify Means (Among Humans) • Hacker News Board

Re: Ends Don't Justify Means (Among Humans)

#61

Post by JoachimSchipper » Wed, Jan 12, 2011, 12:41 PM UTC

samd wrote: post #58

JoachimSchipper wrote: post #43

Earlier quoted context omitted.

I think it's a bit more subtle than that, although he does point out the danger of imitators. See my reply at http://news.ycombinator.com/item?id=2088562 .

I'm not sure. Here is the rule: "Do no harm." Why should we follow it? A rule-consequentialist would say, "Because if everyone followed that rule all the time it would lead to better consequences, on average, than if people tried to calculate the consequences of each individual action and act to maximize them." Why would that lead to better consequences? A rule-consequentialist could say, "Because people are quite ba…

Put like that, I agree - I had a different meaning of "rule-consequentialist" in mind than you, apparently. (I'm not too familiar with the vocabulary of such discussions, sorry!) Thanks for clarifying.

Re: Ends Don't Justify Means (Among Humans)

#62

Post by JoachimSchipper » Wed, Jan 12, 2011, 1:18 PM UTC

yters wrote: post #59

JoachimSchipper wrote: post #44

Earlier quoted context omitted.

I do not understand how that follows. It's entirely possible to imagine a world in which a god/God judges you by the results of your actions instead of your intentions. (E.g. a woman may have been raped, but pre-marital sex still condemns her to Hell; or, perhaps less unfairly, you're not "saved through faith alone" but rather judged by how many people you've positively affected.)

That's deontological, i.e. certain events have inherent value, instead of being judged by their effects. Deontological morality means the effects chain is terminated. Otherwise there is no termination and no value ascribed to an action.

[I'm not too familiar with the (English) vocabulary used in these discussions, so I'll try to expand my previous comment a little and lay off the jargon. If you still disagree, can you point out where?]

> If you agree moral relativism is bad, you agree consequentialism is bad.

I disagreed with this statement.

My examples were meant to say: it's possible to have "god-given morals" (and thus no "moral relativism") while still judging acts by their consequences ("consequentialism").

In such a universe, wearing a sexy dress into a bad neighborhood may be morally fine (act ok), but if this causes you to get raped (pre-marital sex bad) you're still going straight to Hell (so act ok, results bad -> bad). Conversely, if I'd kill my neighbour for no good reason (act bad), and my neighbour happened to be 1920's Hitler (results good), I will be rewarded richly (so act bad, results good -> good).

Some disclaimers: this may be based on a misunderstanding of the words you used; I don't think the universe I sketched is the universe in which we live; and of course rape is not the victim's fault, and not wearing sexy dresses may not be enough to prevent it from happening (for the sake of discussion, though, in this particular instance it wouldn't have happened if the victim had worn a more conservative garment instead.)

Re: Ends Don't Justify Means (Among Humans)

#63

Post by samd » Wed, Jan 12, 2011, 5:23 PM UTC

JoachimSchipper wrote: post #61

samd wrote: post #58

Earlier quoted context omitted.

I'm not sure. Here is the rule: "Do no harm." Why should we follow it? A rule-consequentialist would say, "Because if everyone followed that rule all the time it would lead to better consequences, on average, than if people tried to calculate the consequences of each individual action and act to maximize them." Why would that lead to better consequences? A rule-consequentialist could say, "Because people are quite ba…

Put like that, I agree - I had a different meaning of "rule-consequentialist" in mind than you, apparently. (I'm not too familiar with the vocabulary of such discussions, sorry!) Thanks for clarifying.

Don't worry, like any academic field philosophy is filled with jargon, but knowing the jargon doesn't make you any smarter and not knowing doesn't make you any dumber.

Both you and Eliezer made good arguments and points, they just happened to already have a name in the philosophical jargon.

If you're interested, the Stanford Encyclopedia of Philosophy at plato.stanford.edu is the greatest repository of philosophical knowledge on the Internet.

They have an article about rule-consequentialism: http://plato.stanford.edu/entries/consequentialism-rule/

Re: Ends Don't Justify Means (Among Humans)

#64

Post by syllogism » Sat, Jan 15, 2011, 8:55 AM UTC

yters wrote: post #34

syllogism wrote: post #10

Earlier quoted context omitted.

What do you mean by that? Why is it so? What is a "right", in fact? You've just offered another slogan. You're going to have to work harder than that if you want to argue against consequentialism --- the idea that what matters are the consequences of each action, not the principles it contravenes.

What ends are good in consequentialism? De-ontological morals to one degree or another are necessary to avoid moral relativism. If you agree moral relativism is bad, you agree consequentialism is bad.

Ultimately consequentialism requires you to adopt some set of moral axioms that define what ends are good. If you adopt zero such axioms, then you're back at nihilism. Once you have your minimal set of axioms, you then reason about consequences of the action to try to figure out its likely effects on net utility.

Most consequentialists adopt a "golden rule" sort of axiom about minimising suffering. An alternate formulation is about maximising agents' preferences. Generally people's intuitions about morality are broadly similar, so the challenge is to handle the corner cases most efficiently. But the basic moral relativism problems don't occur for consequentialists.

It's difficult to arrive at a plausible set of moral axioms that are going to lead you to positions such as "homosexuality is unethical", "might is right", or "slavery is ethically neutral". Very few consequentialists arrive at these positions, as far as I'm aware. Some consequentialists do end up back at what are essentially deontoligical positions, however, by re-deriving them consequentially. They argue that it's ineffective to try to reason ethically on a case-by-case basis. I think this is an empirical question about what is practical for most humans.

Re: Ends Don't Justify Means (Among Humans)

#65

Post by syllogism » Sat, Jan 15, 2011, 9:07 AM UTC

JoachimSchipper wrote: post #62

yters wrote: post #59

Earlier quoted context omitted.

That's deontological, i.e. certain events have inherent value, instead of being judged by their effects. Deontological morality means the effects chain is terminated. Otherwise there is no termination and no value ascribed to an action.

[I'm not too familiar with the (English) vocabulary used in these discussions, so I'll try to expand my previous comment a little and lay off the jargon. If you still disagree, can you point out where?] > If you agree moral relativism is bad, you agree consequentialism is bad. I disagreed with this statement. My examples were meant to say: it's possible to have "god-given morals" (and thus no "moral relativism") whil…

In my opinion this hypothetical God that judges actions based on their consequences simply allows you to derive a consequentialist position in a deontological framework.

In consequentialism, you adopt some set of moral axioms, and say "this is how I'm going to define what worlds are good and bad, and I'm going to judge actions according to the worlds they are likely to create". This deontological version is instead saying, "I'm going to imagine there's a god, who reasons morally as follows...", and then saying the god reasons consequentially.

I think this derivation path does get you away from the "moral relativism" that's at the bedrock of a consequentialist position --- you've got to adopt some axioms. But it only does this by imagineering this "god" that behaves in an arbitrary way. All this is doing is pretending that the axiom you desire is a property of the universe you inhabit, rather than a property of you.