Jordan comments on Strong moral realism, meta-ethics and pseudo-questions. - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (172)
Channeling my inner Eliezer, the answer is obviously that you should choose morality (since "should" is just "morality" as a verb).
No, because normative judgement = morality.
This is almost starting to make sense, except... Suppose I say this to a babyeater: "We should sign a treaty banning the development and use of antimatter weapons." What could that possibly mean? Or if one murderer says to another "We should dump the body in the river." he is simply stating a factual falsehood?
I wonder if this is a good summary of our disagreement with Eliezer:
Out of these two, 1 is clearly both a bigger problem and where Eliezer is more obviously wrong. I really don't understand why he sticks to his position there.
Supposedly the acceptable plan would both be the right thing to do and the babyeating thing to do at the same time: right given the presence and influence of babyeaters, and babyeating given the presence and influence of humans. So, when it is said, "Let us sign this treaty.", humans sign it, because it should be done, and babyeaters also do so, because it's a babyeating thing to do. The contract is chosen to compel both parties.
I agree with your explanation of the intended semantics of the sentence, which is also my explanation. What I disagree with is the suggestion that we denote that meaning using "Let us sign this treaty." instead of "We should sign this treaty." I believe the intended meaning is more naturally expressed using the second sentence, and trying to redefine the word "should" so that the second sentence means something else and we're forced to use the first sentence to express the same meaning, is wrong.
Also, since the first sentence is imperative instead of declarative, I'm not sure that it doesn't mean something else already, so that now you're hijacking two words instead of one.
There can be a separable sense of "should" that indicates rationality. Thus, "we should sign the treaty" can be an interesting truth for both parties when the "should" is that of rationality, and true for both parties but only interesting from the human side when the "should" is a moral should.
This commits one to what philosophers call moral externalism, namely, the view that what is morally required is not necessarily rationally required. Which is not a reason to reject the view, but I expect it will be criticized.
Where does he characterize it as realism? When he chooses the word, he always chooses "cognitivism"; if someone else says "realism," he doesn't object, but he makes sure to define it to match cognitivism and indicates that there other notions of realism that he doesn't endorse.
Thanks for pointing out the error. I changed it to "classify".
Should has many meanings. Which moral system I believe in is meta level, not object level and probably implies an epistemic-should or rational-should rather than moral-should.
Likewise, not all normative judgement is morality. What you should do to maximise personal pleasure, .lor make money, or "win" in some way , is generally not what you morally-should.