Lumifer comments on 2013 Survey Results - Less Wrong

74 Post author: Yvain 19 January 2014 02:51AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (558)

You are viewing a single comment's thread. Show more comments above.

Comment author: Sophronius 27 January 2014 08:53:15PM *  0 points [-]

"I think women should be barefoot and pregnant" is a factual statement about Charlie's preferences which are themselves reducible to the physical locations of atoms in the universe (specifically Charlie's brain). Futhermore, the statement "We should make sure women remain barefoot and pregnant" follows logically from Charlie's previous statement about his preferences regarding women. I would expect you to say that Charlie is factually wrong. In which way is he factually wrong and Bob isn't?

Charlie is, presumably, factually correct in that he thinks that he holds that view. However, while preferences regarding colour are well established, I am sceptical regarding the claim that this is an actual terminal preference that Charlie holds. It is possible that he finds pregnant barefeeted women attractive, in which case his statement gives valid information regarding his preferences which might be taken into account by others: In this case it is meaningful. Alternatively, if he were raised to think that this is a belief one ought to hold then the statement is merely signalling politics and is therefore of an entirely different nature.

"I like blue and want the town to be painted blue" gives factual info regarding the universe. "Women ought to be pregnant because my church says so!" does not have the primary goal of providing info, it has the goal of pushing politics.

Imagine a person holding a gun to your head and saying "You should give me your money". Regardless of his use of the word "should", he is making an implicit logical argument:
1) Giving me your money reduces your chances of getting shot by me
2) You presumably do not want to get shot
3) Therefore, you should give me your money

If you respond to the man by saying that morality is relative, you are rather missing the point.

The statement "We should paint all the buildings in town blue!" is not a claim in need of evidence. It is a command, an expression of what Bob thinks should happen. It has nothing to do with how many people think the same.

I think you are missing the subtle hidden meanings of everyday discourse. Imagine Bob saying that the town should be painted blue. Then, someone else comes with arguments for why the town should not be painted Blue. Bob eventually agrees. "You are right", he says, "that was a dumb suggestion". The fact that exchanges like this happen all the time shows that Bob's statement is not just a meaningless expression, but rather a proposal relying on implicit arguments and claims. Specifically, it relies on enough people in the village sharing his preference for blue houses that the notion will be taken seriously. If Bob did not think this to be the case, he probably would not have said what he did.

Comment author: Lumifer 27 January 2014 09:07:33PM *  1 point [-]

I am sceptical regarding the claim that this is an actual terminal preference that Charlie holds

Given that you know absolutely nothing about Charlie, a player in a hypothetical scenario, I find your scepticism entirely unwarranted. Fighting the hypothetical won't get you very far.

So, is Charlie factually wrong? On the basis of what would you determine that Charlie's belief is wrong and Bob's isn't?

Imagine a person holding a gun to your head and saying "You should give me your money". ... If you respond to the man by saying that morality is relative, you are rather missing the point.

Why would I respond like that? What does the claim that morality is relative have to do with threats of bodily harm?

I think you are missing the subtle hidden meanings of everyday discourse.

In this context I don't care about the subtle hidden meanings. People who believe they know the Truth and have access to the Sole Factually Correct Set of Values tend to just kill others who disagree. Or at the very least marginalize them and make them third-class citizens. All in the name of the Glorious Future, of course.

Comment author: Sophronius 28 January 2014 09:37:03AM *  0 points [-]

Well, given that Charlie indeed genuinely holds that preference, then no he is not wrong to hold that preference. I don't even know what it would mean for a preference to be wrong. Rather, his preferences might conflict with preferences of others, who might object to this state of reality by calling it "wrong", which seems like the mind-projection fallacy to me. There is nothing mysterious about this.

Similarly, the person in the original example of mine is not wrong to think men kissing each other is icky, but he IS wrong to conclude that there is therefore some universal moral rule that men kissing each other is bad. Again, just because rationality does not determine preferences, does not mean that logic and reason do not apply to morality!

In this context I don't care about the subtle hidden meanings. People who believe they know the Truth and have access to the Sole Factually Correct Set of Values tend to just kill others who disagree. Or at the very least marginalize them and make them third-class citizens. All in the name of the Glorious Future, of course.

I believe you have pegged me quite wrongly, sir! I only care about truth, not Truth. And yes, I do have access to some truths, as of course do you. Saying that logic and reason apply to morality and that therefore all moral claims are not equally valid (they can be factually wrong or entirely nonsensical) is quite a far cry from heralding in the Third Reich. The article on Less Wrong regarding the proper use of doubt seems pertinent here.

Comment author: Lumifer 29 January 2014 05:13:16PM 0 points [-]

Well, given that Charlie indeed genuinely holds that preference, then no he is not wrong to hold that preference.

I am confused. Did I misunderstand you or did you change your mind?

Earlier you said that "should" kind of questions have single correct answers (which means that other answers are wrong). A "preference" is more or less the same thing as a "value" in this context, and you staked out a strong position:

I reject your notion of a strict fact-value distinction: I posit to you that all statements are either reducible to factual matters or else they are meaningless as a matter of logical necessity. ... but questions about morality ... should ... be answered in a rational and factual manner all the same.

Since statements of facts can be correct or wrong and you said there is no "fact-value distinction", then values (and preferences) can be correct or wrong as well. However in the parent post you say

I don't even know what it would mean for a preference to be wrong.

If you have a coherent position in all this, I don't see it.

Comment author: Sophronius 30 January 2014 06:03:49PM *  1 point [-]

I think you misunderstood me. Of course I don't mean that the terms "facts" and "values" represent the same thing. Saying that a preference itself is wrong is nonsense in the same way that claiming that a piece of cheese is wrong is nonsensical. It's a category error. When I say I reject a strict fact-value dichotomy I mean that I reject the notion that statements regarding values should somehow be treated differently from statements regarding facts, in the same way that I reject the notion of faith inhabiting a separate magistrate from science (i.e. special pleading). So my position is that when someone makes a moral claim such as "don't murder", they better be able to reduce that to factual statements about reality or else they are talking nonsense.

For example, "sex is sinful!" usually reduces to "I think my god doesn't like sex", which is nonsense because there is no such thing. On the other hand, if someone says "Stealing is bad!", that can be reduced to the claim that allowing theft is harmful to society (in a number of observable ways), which I would agree with. As such I am perfectly comfortable labelling some moral claims as valid and some as nonsense.

Comment author: Lumifer 30 January 2014 06:37:29PM 1 point [-]

I don't see how this sentence

Saying that a preference itself is wrong is nonsense in the same way that claiming that a piece of cheese is wrong is nonsensical. It's a category error.

is compatible with this sentence

I reject the notion that statements regarding values should somehow be treated differently from statements regarding facts

Comment author: Sophronius 31 January 2014 11:13:18AM *  1 point [-]

I am distinguishing between X and statements regarding X. The statement "Cheese is wrong" is nonsensical. The statement "it's nonsensical to say cheese is wrong" is not nonsensical. Values and facts are not the same, but statements regarding values and facts should be treated the same way.

Similarly: Faith and Science are not the same thing. Nonetheless, I reject the notion that claims based on faith should be treated any differently from scientific claims.

Comment author: Eugine_Nier 01 February 2014 03:44:52AM 0 points [-]

Similarly: Faith and Science are not the same thing. Nonetheless, I reject the notion that claims based on faith should be treated any differently from scientific claims.

Do you also reject the notion that claims about mathematics and science should be treated differently?

Comment author: Sophronius 01 February 2014 10:03:20AM 0 points [-]

In the general sense that all claims must abide by the usual requirements of validity and soundness of logic, sure.

In fact, you might say that mathematics is really just a very pure form of logic, while science deals with more murky, more complicated matters. But the essential principle is the same: You better make sure that the output follows logically from the input, or else you're not doing it right.

Comment author: Eugine_Nier 01 February 2014 10:57:56PM 0 points [-]

In the general sense that all claims must abide by the usual requirements of validity and soundness of logic, sure.

My point is that what constitutes "validity" and "soundness of logic" differs between the two domains.