I don’t disagree with Eliezer’s position for the most part, I just don’t see where he lays out a coherent foundation for why he believes certain things about human values. (Or maybe I’m just being uncharitable in my evaluation and not counting some things as “real arguments” that others would.)
By objective and discoverable, I meant in the sense of the values understood in and of themselves without reference to humans in particular. Obviously you can just model human brains and understand what they value, but I meant that you can’t learn about “beauty” or “friendship” or what have you outside of that. That part of the post was inelegantly worded, and I’d probably strike that out if this was a long post and not a comment.
I used “Moral Fictionalist” as a descriptor for Eliezer’s position because, although he probably wouldn’t ascribe it to himself, it seems to me to be the best fit for it. I’m not a rationalist, and I don’t have a rationalist background, I just like to read the site from time to time, and very occasionally comment. So my diction tends to sound “foreign” here.
I’m not sure what this is referring to…. I used the term “moral antirealist” and you used the term “moral fictionalist”, both of which are philosophy jargon, not Eliezer / rationalist jargon. (I assume you were using “moral fictionalist” in the standard philosophy jargon sense, right?)
Anyway, I have just read more of that SEP article, but remain confused by it. For example, if Mathematical Theorem X is provable from Axiom Set Y, is Theorem X “fictional” or not? My impression from the SEP thing is that philosophers sometimes argue about this question. But I don’t understand the nature of that argument. What’s at stake? Why can’t I just say “who cares, call it whatever you want to call it”.
I bring up that example because I think Eliezer sees “X is good” claims as being pretty analogous to “Theorem X is provable from Axiom Set Y” claims. Specifically, the analogy would be:
Axiom Set Y <--> the innate motivations and inclinations in human brains (ignoring interpersonal differences, which he views as sufficiently minor that this is OK to ignore)
Mathematical inference steps <--> the stuff that happens when people use their innate motivations and inclinations, along with their knowledge and reasoning abilitites, to reflect on the nature of The Good. Well, actually, some idealization of that.
“Theorem X is provable from Axiom Set Y” <--> such-and-such thing is Good
OK, if we accept all those parts of (my attempt to relay) Eliezer’s view, then is the statement “X is Good” a “fiction” in the moral fictionalist sense? I still think the answer is no, but I’m not very confident.
(You’re correct. I was using fictionalist in that sense.)
I think the equivocation of “Theorem X is provable from Axiom Set Y” <--> such-and-such thing is Good; would be the part of that chain of reasoning a self-described fictionalist would ascribe fictionality to.
As I understand it, it’s the difference between thinking that Good is a real feature of the universe and Good being a wordgame that we play to make certain ideas easier to work with. Maybe a different example could illuminate some of this.
Fictionalism would be a good tool to describe the way we talk about Evolution and Nature. As has sometimes been said on this site, humans are not aligned towards Evolution, since they aren’t inclusive fitness maximizers. We also say things like: such-and-such a feature evolved to do X function on an organism. Of course, that’s not true. Biological features don’t evolve in order to do a thing, they just happen to do things as a consequence of surviving in an ancestral environment.
We talk about organs and limbs “evolving to do” things, even when they do not, because it is a fiction that makes Evolution more palatable to intellectual examination, but unless you belief in weird stuff like teleology, it’s just a fiction, a story that is convenient, and corresponds to real features of the world, but is not itself strictly true. And it is not untrue in a provisional way that we expect to be overturned with later reasoning and evidence, but untrue by design, because the literal truth of biological features arising by chance and operating by chance is harder to talk coherently about, given human constraints on mental compute.
I think your presentation of Eliezer’s view is like that: one way it differs from a moral realist is not only that of a category error (objective morality vs aligning to human value) but that of a thought pattern deliberately constructed to aid human cognition vs a thought pattern attempting to align closely with correct mathmatical model of the object(s).
That’s my reading of why it would matter if you’re a moral antirealist (classical) vs a moral antirealist (fictionalist). I do consider fictionalist to be a subset of antirealist.
Update: I said above that Eliezer is a moral antirealist but upon further reflection I think I was wrong. (I’m still not confident though.) This post in particular strikes me as moral realist:
It may be that humans argue about what’s right, and Pebblesorters do what’s prime. But this doesn’t change what’s right, and it doesn’t make what’s right vary from planet to planet, and it doesn’t mean that the things we do are right in mere virtue of our deciding on them—any more than Pebblesorters make a heap prime or not prime by deciding that it’s “correct”.
The Pebblesorters aren’t trying to do what’s p-prime any more than humans are trying to do what’s h-prime. The Pebblesorters are trying to do what’s prime. And the humans are arguing about, and occasionally even really trying to do, what’s right.
I think he treats “what’s right” as having a status similar to “what’s provable from Axiom Set Y”. He thinks there are (something akin to) axioms for morality, and these axioms are downstream of random facts about human evolution; but he bundles those human-specific axioms into the definition of the word “right”.
In other words, Eliezer could have talked about “what’s righthuman brains” instead of “what’s right” (with the same definition, i.e. “what’s righthuman brains” is something like the limit of idealized human moral reflection, definitely not “what actual humans say is right today”), in which case he would have been a moral antirealist. In fact, I think he could have done that with almost no substantive change to anything he wrote in the metaethics sequence. But as written, I think Eliezer is closer to moral realist.
I’m not a philosopher and might be misunderstanding the terminology here. I’m also not Eliezer :)
“Fictionalist” seems to imply that human moral values are arbitrary, free creations. EY seems to be an anti realist , as far as basic ontology goes, but he also emphasizes that you can’t think outside of the human value structure, that they will always be compelling and seemingly real to you.
To me, that’s a quasi-realist position.
Thanks for the reply.
I don’t disagree with Eliezer’s position for the most part, I just don’t see where he lays out a coherent foundation for why he believes certain things about human values. (Or maybe I’m just being uncharitable in my evaluation and not counting some things as “real arguments” that others would.)
By objective and discoverable, I meant in the sense of the values understood in and of themselves without reference to humans in particular. Obviously you can just model human brains and understand what they value, but I meant that you can’t learn about “beauty” or “friendship” or what have you outside of that. That part of the post was inelegantly worded, and I’d probably strike that out if this was a long post and not a comment.
I used “Moral Fictionalist” as a descriptor for Eliezer’s position because, although he probably wouldn’t ascribe it to himself, it seems to me to be the best fit for it. I’m not a rationalist, and I don’t have a rationalist background, I just like to read the site from time to time, and very occasionally comment. So my diction tends to sound “foreign” here.
Thanks!
I’m not sure what this is referring to…. I used the term “moral antirealist” and you used the term “moral fictionalist”, both of which are philosophy jargon, not Eliezer / rationalist jargon. (I assume you were using “moral fictionalist” in the standard philosophy jargon sense, right?)
Anyway, I have just read more of that SEP article, but remain confused by it. For example, if Mathematical Theorem X is provable from Axiom Set Y, is Theorem X “fictional” or not? My impression from the SEP thing is that philosophers sometimes argue about this question. But I don’t understand the nature of that argument. What’s at stake? Why can’t I just say “who cares, call it whatever you want to call it”.
I bring up that example because I think Eliezer sees “X is good” claims as being pretty analogous to “Theorem X is provable from Axiom Set Y” claims. Specifically, the analogy would be:
Axiom Set Y <--> the innate motivations and inclinations in human brains (ignoring interpersonal differences, which he views as sufficiently minor that this is OK to ignore)
Mathematical inference steps <--> the stuff that happens when people use their innate motivations and inclinations, along with their knowledge and reasoning abilitites, to reflect on the nature of The Good. Well, actually, some idealization of that.
“Theorem X is provable from Axiom Set Y” <--> such-and-such thing is Good
OK, if we accept all those parts of (my attempt to relay) Eliezer’s view, then is the statement “X is Good” a “fiction” in the moral fictionalist sense? I still think the answer is no, but I’m not very confident.
(You’re correct. I was using fictionalist in that sense.)
I think the equivocation of “Theorem X is provable from Axiom Set Y” <--> such-and-such thing is Good; would be the part of that chain of reasoning a self-described fictionalist would ascribe fictionality to.
As I understand it, it’s the difference between thinking that Good is a real feature of the universe and Good being a wordgame that we play to make certain ideas easier to work with. Maybe a different example could illuminate some of this.
Fictionalism would be a good tool to describe the way we talk about Evolution and Nature. As has sometimes been said on this site, humans are not aligned towards Evolution, since they aren’t inclusive fitness maximizers. We also say things like: such-and-such a feature evolved to do X function on an organism. Of course, that’s not true. Biological features don’t evolve in order to do a thing, they just happen to do things as a consequence of surviving in an ancestral environment.
We talk about organs and limbs “evolving to do” things, even when they do not, because it is a fiction that makes Evolution more palatable to intellectual examination, but unless you belief in weird stuff like teleology, it’s just a fiction, a story that is convenient, and corresponds to real features of the world, but is not itself strictly true. And it is not untrue in a provisional way that we expect to be overturned with later reasoning and evidence, but untrue by design, because the literal truth of biological features arising by chance and operating by chance is harder to talk coherently about, given human constraints on mental compute.
I think your presentation of Eliezer’s view is like that: one way it differs from a moral realist is not only that of a category error (objective morality vs aligning to human value) but that of a thought pattern deliberately constructed to aid human cognition vs a thought pattern attempting to align closely with correct mathmatical model of the object(s).
That’s my reading of why it would matter if you’re a moral antirealist (classical) vs a moral antirealist (fictionalist). I do consider fictionalist to be a subset of antirealist.
Update: I said above that Eliezer is a moral antirealist but upon further reflection I think I was wrong. (I’m still not confident though.) This post in particular strikes me as moral realist:
I think he treats “what’s right” as having a status similar to “what’s provable from Axiom Set Y”. He thinks there are (something akin to) axioms for morality, and these axioms are downstream of random facts about human evolution; but he bundles those human-specific axioms into the definition of the word “right”.
In other words, Eliezer could have talked about “what’s righthuman brains” instead of “what’s right” (with the same definition, i.e. “what’s righthuman brains” is something like the limit of idealized human moral reflection, definitely not “what actual humans say is right today”), in which case he would have been a moral antirealist. In fact, I think he could have done that with almost no substantive change to anything he wrote in the metaethics sequence. But as written, I think Eliezer is closer to moral realist.
I’m not a philosopher and might be misunderstanding the terminology here. I’m also not Eliezer :)
“Fictionalist” seems to imply that human moral values are arbitrary, free creations. EY seems to be an anti realist , as far as basic ontology goes, but he also emphasizes that you can’t think outside of the human value structure, that they will always be compelling and seemingly real to you. To me, that’s a quasi-realist position.