I agree that factual differences of opinion are often responsible for disagreement, and these are often more difficult to persuade people about than some value disagreements, but I think there are obviously value disagreements as well. For example, I find the idea that an opponent of foreign aid is likely to place exactly as much value on a foreigner‘s life as an effective altruist to be obviously preposterous, even if the fact that the fellow doesn’t have a coherent utility function means that in principle, it’s possible, though in practice, practically impossible to get him to change his view. Even Scott doesn’t seem to actually believe that people place exactly the same weight on different things and in practice how much weight you place on different things can lead to dramatically different outcomes in terms of policy implications. Also given what he writes on the topic, I get the impression that Scott is doing a bit of motivated reasoning because he does not like the implication that people have value disagreement because that sounds a bit too much like conflict theory, although obviously, I think he’s also trying to find out What’s true. That said, I think it’s a bit irrelevant in practice, whether people have value or factual disagreements because often in practice they are equally difficult to get people to change their mind about. People don’t have coherent utility functions. So in principle, I think a super intelligent AI could get you to change your values with persuasion alone. And often, it’s very difficult to get people to change some of their factual opinions either because of just how much probability they place on the accuracy of these beliefs or because they are dogmatically attached to them for a number of reasons.
This feels like a sloppier version of Haidt's Moral Intuitionism. You center facts in the story more than he does, but your notion of fact-meme seems very close to his notion that there are trusted people/groups who we trust for various reasons, and so what they express about reality is the stuff we trust to guide us.
I think your point about Iran is great, but generalization is a bit off. Tl;dr; I do not believe in checklists, rather in conflict between different values, as Berlin bequeathed.
There is multiple things going on here. First, Scott says that there is a continious line between nativist and inteventionist, which determins how open you are to foreign aid. Second, he argues that "[the world is] made of a giant mishmash of provisional things that solidify into values at some point".
The second is a great point. You narrow it a bit in your essay to "fact-based memes" that people run through their lists, but IMO Scott interpretation is more correct. I became consciously Zionist probably at the age of 10-12 (not Jewish). I definetely didn't run tests on Zionism to check if it's a good idea; rather, I grew up in Russia, happened to be drawn to liberal ideals, many Russian liberals are Jews, so I just got support for Israel as an extra. For other people, it may be even less formed: somebody may be a Zionist because his girlfriend is Jewish, or because first essay they read on the topic was arguing this point. I eventually got better reasons to support Israel, but that happened much later. Anecdotally, I would assume it is more common to just adopt the same ideas as your friends than actually think about them.
On the other hand, great point about Iran debate being about nukes, not about nativism vs interventionism. Scott shouldn't have flattened "mishmash" into a line. Real world is complicated, and as Isaiah Berlin noted, even your good values would likely come to conflict: in this case, "anti-war" is pulling towards nativism, while "anti-nukes for Axis of Evil" is pulling towards interventionism. Or less beningly, you may be pro-Mexican immigration, because you want cheap labor, your personal experience with Mexicans is nice, and you watched a touching movie about them, but against Muslim immigration, because you think they chant "globalize the Intifada" too often. It's not really a point on interventionist-nativist line, but rather a bunch of motivations, some a good and some are bad, that may or may not apply to each particular case.
Which also explains why checklists don't work. Japanese are more sympathetic, Western, allied, portrayed as hard-working, suffered a natural disasters, and, yes, less corrupt than most African countries. This together could outweight general lack of desire to send money abroad. African countries with their permanent state of destitution is a different thing. Even if you come up with the studies to cross the corruption argument, you won't make people like the country they won't even be able to find on the map more.
> you may be pro-Mexican immigration, because you want cheap labor, your personal experience with Mexicans is nice, and you watched a touching movie about them, but against Muslim immigration, because you think they chant "globalize the Intifada" too often.
Again, I'm tempted to draw this up as a matter of fact-memes rather than values: if your worldview is "Mexicans are nice; Muslims are mean" and everyone's checklist has something like "if they're nice, and if they're not mean, then I want them to live next to me," so the pro-Mexican / anti-Muslim "values" pop right out.
Similar thing for the Japanese: it's a series of fact-statements about their character, a series of fact-statements about whether that character bears on "deservingness" (or maybe whether it's indicative of using aid poorly, etc.), and then a mutually-agreed-on principle of "we should give aid to people who satisfy x y z criteria because they won't waste it or embezzle it or whatever."
> Even if you come up with the studies to cross the corruption argument, you won't make people like the country they won't even be able to find on the map more.
This is a good point. I'm just still uncertain of what to call that second-order preference for non-updating! "Fundamental value" feels too immutable, like Scott pointed out, but even "continuum of values" seems... too charitable, maybe? There's a risk that this reductive fact-meme view I've taken will deteriorate into the left-liberal "stop misinformation at all costs" thing, but I'm not sure it *has* to—I would rather it pop out as "LessWrong for the masses," or something along those lines.
I think there’s an amount of truth in this. But I don’t think it’s universally true. I think for some people, for proposition x there exists no evidence y that will make them support x. I don’t think this is exclusively true for my enemies either, for instance there’s probably no evidence or conditionals that would make me oppose gay marriage, or accept Christianity (even if it was factually true, I just wouldn’t worship) and there’s nothing else I could really call those except “values differences”.
But your acceptance of Christianity surely isn't dependent only on whether you think the religion itself is true! What if you believed that it was true, believed that Christians were generally good and pure-hearted people, believed that your whole social group & family & so on would continue accepting you if you converted, believed that worship was essential to securing a spot in heaven, etc. etc.?
Or for gay marriage: what if you earnestly truly believed that a nation which made gay marriage legal would be forsaken by God and left to lie in ruin? I think you'd have good reason to oppose it!
Those seem like more fact-y conditionals than value-y ones to me, but maybe you disagree..?
Whether Christians are good people inherently is value-laden, as it requires a standard by which I judge “good”.
Whether God would forsake my nation *does* seem more fact based, though in the case that I believed it I would oppose gay marriage on a consequentialist basis only, whereas I think many people would condemn it because they believe God’s opinion is morally relevant irrespective of consequences.
I agree that factual differences of opinion are often responsible for disagreement, and these are often more difficult to persuade people about than some value disagreements, but I think there are obviously value disagreements as well. For example, I find the idea that an opponent of foreign aid is likely to place exactly as much value on a foreigner‘s life as an effective altruist to be obviously preposterous, even if the fact that the fellow doesn’t have a coherent utility function means that in principle, it’s possible, though in practice, practically impossible to get him to change his view. Even Scott doesn’t seem to actually believe that people place exactly the same weight on different things and in practice how much weight you place on different things can lead to dramatically different outcomes in terms of policy implications. Also given what he writes on the topic, I get the impression that Scott is doing a bit of motivated reasoning because he does not like the implication that people have value disagreement because that sounds a bit too much like conflict theory, although obviously, I think he’s also trying to find out What’s true. That said, I think it’s a bit irrelevant in practice, whether people have value or factual disagreements because often in practice they are equally difficult to get people to change their mind about. People don’t have coherent utility functions. So in principle, I think a super intelligent AI could get you to change your values with persuasion alone. And often, it’s very difficult to get people to change some of their factual opinions either because of just how much probability they place on the accuracy of these beliefs or because they are dogmatically attached to them for a number of reasons.
This feels like a sloppier version of Haidt's Moral Intuitionism. You center facts in the story more than he does, but your notion of fact-meme seems very close to his notion that there are trusted people/groups who we trust for various reasons, and so what they express about reality is the stuff we trust to guide us.
Always a good sign when I've done a sloppy job of something that someone smarter already did...
I think your point about Iran is great, but generalization is a bit off. Tl;dr; I do not believe in checklists, rather in conflict between different values, as Berlin bequeathed.
There is multiple things going on here. First, Scott says that there is a continious line between nativist and inteventionist, which determins how open you are to foreign aid. Second, he argues that "[the world is] made of a giant mishmash of provisional things that solidify into values at some point".
The second is a great point. You narrow it a bit in your essay to "fact-based memes" that people run through their lists, but IMO Scott interpretation is more correct. I became consciously Zionist probably at the age of 10-12 (not Jewish). I definetely didn't run tests on Zionism to check if it's a good idea; rather, I grew up in Russia, happened to be drawn to liberal ideals, many Russian liberals are Jews, so I just got support for Israel as an extra. For other people, it may be even less formed: somebody may be a Zionist because his girlfriend is Jewish, or because first essay they read on the topic was arguing this point. I eventually got better reasons to support Israel, but that happened much later. Anecdotally, I would assume it is more common to just adopt the same ideas as your friends than actually think about them.
On the other hand, great point about Iran debate being about nukes, not about nativism vs interventionism. Scott shouldn't have flattened "mishmash" into a line. Real world is complicated, and as Isaiah Berlin noted, even your good values would likely come to conflict: in this case, "anti-war" is pulling towards nativism, while "anti-nukes for Axis of Evil" is pulling towards interventionism. Or less beningly, you may be pro-Mexican immigration, because you want cheap labor, your personal experience with Mexicans is nice, and you watched a touching movie about them, but against Muslim immigration, because you think they chant "globalize the Intifada" too often. It's not really a point on interventionist-nativist line, but rather a bunch of motivations, some a good and some are bad, that may or may not apply to each particular case.
Which also explains why checklists don't work. Japanese are more sympathetic, Western, allied, portrayed as hard-working, suffered a natural disasters, and, yes, less corrupt than most African countries. This together could outweight general lack of desire to send money abroad. African countries with their permanent state of destitution is a different thing. Even if you come up with the studies to cross the corruption argument, you won't make people like the country they won't even be able to find on the map more.
This is a wonderful comment, thanks!
> you may be pro-Mexican immigration, because you want cheap labor, your personal experience with Mexicans is nice, and you watched a touching movie about them, but against Muslim immigration, because you think they chant "globalize the Intifada" too often.
Again, I'm tempted to draw this up as a matter of fact-memes rather than values: if your worldview is "Mexicans are nice; Muslims are mean" and everyone's checklist has something like "if they're nice, and if they're not mean, then I want them to live next to me," so the pro-Mexican / anti-Muslim "values" pop right out.
Similar thing for the Japanese: it's a series of fact-statements about their character, a series of fact-statements about whether that character bears on "deservingness" (or maybe whether it's indicative of using aid poorly, etc.), and then a mutually-agreed-on principle of "we should give aid to people who satisfy x y z criteria because they won't waste it or embezzle it or whatever."
> Even if you come up with the studies to cross the corruption argument, you won't make people like the country they won't even be able to find on the map more.
This is a good point. I'm just still uncertain of what to call that second-order preference for non-updating! "Fundamental value" feels too immutable, like Scott pointed out, but even "continuum of values" seems... too charitable, maybe? There's a risk that this reductive fact-meme view I've taken will deteriorate into the left-liberal "stop misinformation at all costs" thing, but I'm not sure it *has* to—I would rather it pop out as "LessWrong for the masses," or something along those lines.
I think there’s an amount of truth in this. But I don’t think it’s universally true. I think for some people, for proposition x there exists no evidence y that will make them support x. I don’t think this is exclusively true for my enemies either, for instance there’s probably no evidence or conditionals that would make me oppose gay marriage, or accept Christianity (even if it was factually true, I just wouldn’t worship) and there’s nothing else I could really call those except “values differences”.
But your acceptance of Christianity surely isn't dependent only on whether you think the religion itself is true! What if you believed that it was true, believed that Christians were generally good and pure-hearted people, believed that your whole social group & family & so on would continue accepting you if you converted, believed that worship was essential to securing a spot in heaven, etc. etc.?
Or for gay marriage: what if you earnestly truly believed that a nation which made gay marriage legal would be forsaken by God and left to lie in ruin? I think you'd have good reason to oppose it!
Those seem like more fact-y conditionals than value-y ones to me, but maybe you disagree..?
Whether Christians are good people inherently is value-laden, as it requires a standard by which I judge “good”.
Whether God would forsake my nation *does* seem more fact based, though in the case that I believed it I would oppose gay marriage on a consequentialist basis only, whereas I think many people would condemn it because they believe God’s opinion is morally relevant irrespective of consequences.
“May not morality have its origins in error” -Francis Picabia