I have some sympathy for the idea that we have realist moral intuitions but I think these intuitions are quite superficial and do not stand up to much scrutiny.
What is the nature of these "realist moral intuitions"? Are they a fragile echo of a the truth that is "out there somewhere"? I say fragile echo because we know we have moral disagreement and that must surely mean the "real" truth is patchy. What is the mechanism by which this fragile echo is constructed? We could hypothesize evolution. If you're looking through the things perceived as moral facts how do you establish which are purely evolutionary artifacts and which ones are "real" moral facts? Presumably there all perceived moral facts that are consistent with evolution (otherwise we'd have evidence against evolution). In which case the parsimonious explantion for these is evolutionary selection. Unless you're going to claim some more profound linkage between evolutionary selection and moral facts. Without this explanation I think intuitions of "real" moral facts seem like them must just be a mirage.
It's sort of like having intuitions about sentient life on other planets and jumping to the conclusion that all those UFOs being seen are real spaceships.
I used to be a moral anti-realist, and I'm now in the moral realist camp. I totally understand both the frustration Christian is expressing here, as well as your stance. You believe you're correct and you're not seeing substantive arguments. I also agree that moral realists need to bite the bullet and make predictive claims.
I think the right way to frame moral realism as a predictive claim is to say:
- moral intuitions evolved because they helped our ancestors perpetuate their genes into the future
- thus, moral intuitions are approximations of strategies that actually help genes perpetuate themselves into the future
- thus, if moral realism exists, we should expect some belief structures to lead to thriving, others not to, over very long periods of time. The objective reality that moral intuitions point at is 'what actually works for life, long term'.
So here's a solid prediction i'm happy to make: no "group" that believes morality isn't real is real is going to be able to reliably perpetuate itself into the future more than a few centuries. If morality really is about long term survival and thriving, we should expect that groups will thrive in proportion to how accurately they understand morality.
hat word "group", it problems needs more defining. The time horizons here are necessarily large, because we're talking about perpetuating yourself arbitrarily far into the future, and likely talking about probabilistic outcomes morose than guaranteed outcomes.
No, you can't do experiments on this - but that may be a 'not yet' thing. If we had, say, very realistic reality simulators and could run them for very long periods of time, we maybe very well could put some of these ideas to the test. Or, if we had WAY more written history than we currently do, we could look at which groups are able to keep existing multiple generations and which ones aren't.
The fact that we can't test these ideas now probably has more to do with technological limitations. But, among other things, this theory explains why some traditions from the ancient world - like, say, the Jews, or reading Confucius, ore the teachings of the Buddha - are still around today. Longevity of a set of moral ideas becomes evidence for their fiteddness to moral reality.
"If morality really is about long term survival and thriving, we should expect that groups will thrive in proportion to how accurately they understand morality."
I do not understand exactly what you mean by this. Survival depends on doing what works - I don't see the connection with accurately understanding why what works...works.
"The objective reality that moral intuitions point at is 'what actually works for life, long term'."
I agree. But it does not follow that understanding why it works is necessary - nor 'believing' morality is 'real'. Believing moral norms are 'necessary' is all that is required. But believing they are true, whatever that means, is not.
Anthropology will show you this. There many hunter gatherer groups that have very specific and complex steps in food preparation. Failure to follow these steps leads to poisoning. They know that failure to follow these steps will lead to poisoning, but they don't know why. They just know what works - but they don't necessarily know why.
"no "group" that believes morality isn't real is real is going to be able to reliably perpetuate itself into the future more than a few centuries."
Again, I don't follow this reasoning. Following the rules of grammar are necessary for communicating with others. But 'believing' in the rules of grammar: I don't know what that could mean?
"thus, if moral realism exists, we should expect some belief structures to lead to thriving".
Yes, but thriving in one environment does not equate to thriving in a different environment. Hence, the difference we find in moral norms from one culture to the next. Species can survive over great periods of time so long as the environment remains relatively stable. Once the environment changes they can find themselves in trouble.
Anyway, I am sympathetic to your views. But I think you would be better to describe your position as a realist in spirit, rather than realist in any philosophical, meta ethical sense.
Thank you for the answer here. Your objections are helping me evolve how I'm phrasing what I believe, which is great.
The example of hunter gatherers following a complex recipe they don't understand really sharpens the point. I agree, that approach works, only until the environment changes. And then it breaks.
But if you understand _why_ the process works, now you have a shot at changing it to adapt for the changing environment. I think the reality that tells us whether moral claims are true, is if they can 'scale' across large periods of time, large groups of people, and large groups of different environments.
I see your point on grammar, but i think a better analogy is the periodic table of the elements. It's clearly socially constructed. You could draw it any which way. But only one way of drawing it actually helps you navigate the reality of chemistry. If all you have are norms, but no theory of _why those norms work_, you can't intentionally adjust the norms when the environment changes. You can't say, "this norm used to work because the environment was like X, but now it's not, so we need to change the norm. " Likewise, if, for some reason, our universe had fluctuations in fundamental constants, say the relative strength of the weak vs the strong force, we _would_ have to rewrite the periodic table. But we could only do that if we understood _why_ it was shaped that way to begin with.
On thriving not translating between environments: I think this is only sort of true. Some norms work across environments. One example would be, "you should try to understand the environment." That one will help you regardless of where you are. The term AI safety researchers use is "convergent instrumental subgoals". I'm not sure how widespread that term is, but what I believe is that convergent instrumental subgoals are at least a good approximation of moral reality, maybe in the same way that Newtonian physics is a good approximation of the true laws of physics, at least in the domains we could access prior to ~1850 or so.
Some will object that convergent instrumental subgoals don't just give you morality. I think that's ultimately wrong, but that's a much larger claim. I have substack post here, for example, arguing that love and sacrifice have powerful mathematical properties:
"But if you understand _why_ the process works, now you have a shot at changing it to adapt for the changing environment."
I am not saying one is helpless, but it really depends on how much the environment changes. There is countless examples of western explorers dying in harsh environments in which natives to the same area have lived for thousands of years. This because they do not possess the accumulated cultural know how that has been passed down though generation after generation.
"I think the reality that tells us whether moral claims are true".
It tells what is true in the environment we currently inhabit. But it does not tell us what is true independently of any and all environments. The realist position is that what is true is true independently of any and all environments. This is why I think you are mistaken calling yourself a moral realist in any meta ethical sense. A realist in spirit is not committed to meta ethical realism.
As a side note, I do not agree with Lance's anti-realism, for it commits one to a meta ethical position. A a realist in spirit I am not committed to any meta ethical position. Meta ethics in my view is unintelligible regardless of where one stands.
"I think the reality that tells us whether moral claims are true, is if they can 'scale' across large periods of time, large groups of people, and large groups of different environments."
I don't know of any moral norms that have scaled across 'large groups of different environments'.
I do believe in universal moral claims, yes. I think these exist for reasons outside of the scope of physicalism, but they have physical consequences. So i'm hoping to talk about those physical consequences in a way that will work for people who want predictions.
I don't think moral behavior is _guaranteed_ to keep you alive or thriving in any environment. An example here would be, "treat the people around you well ." I think that strategy works for reasons that will show up in _any_ multiplayer game with agents and uncertainty.
One prediction that comes from this theory: we should expect certain practices to emerge in all large scale cultures, and we should even expect narratives that describe how these practices are important and things 'fall apart' when they are broken.
And we see just this. In china, there is the notion of the mandate of heaven being lost. In Hindu culture there is this idea of the cycle of Yugas. In the west we have a notion of empires rising and falling, as they become 'decadant' and 'corrupt'. In the old testament, a similar pattern in show, as the old testament god says "if you don't follow these practices, your neighbors will conquer and enslave you."
All of those stories have this similar shape: "there's a right way to be, and tings fall apart when you deviate from it."
I liked your comment and your attempt at making a predictive claim for realism. I feel some problems in how you seem to be defining moral realism, though. "Moral intuitions are approximations of strategies that actually help genes perpetuate themselves" feels eminently reasonable -I don't think I have a dispute with this-, but this is not a definition most moral realists would accept, as it makes the value of moral principles dependent on a non-moral, pragmatic principle (reproductive/memetic success). More than that, it reduces it to that (so all core 'objective moral truths' that most realists believe would have to be sacrificed, if necessary, to this new Golden Idol, and not be objective and true in themselves).. It also opens itself up to counterexamples: one can imagine that an evolved, transhuman homo sapiens might easily employ strategies for longevity that would likely be considered immoral by realists (i.e, devouring other inhabited worlds in Galactus-like fashion to sustain our individual and/or group flourishing, power and continuity).
I think you're probably right that lots of moral realists would not accept this definition. I think many would refuse _any_ definition, and I agree with Lance here that that if you can't make predictions, and open yourself to counterexamples, you're a lot harder to take seriously from the outside.
The risk that this becomes the 'golden idol' is mediated, I think, by the reality that nobody knows the distant future. Anyone claiming certainty about "this is sustainable" is fundamentally at odds with physics: chaotic systems are real, and play big roles in human affairs. This means we can't calculate precise outcomes across long time horizons.
Yes I get that people can imagine all kinds of things, but you could have made the same objection to the law of conservation of energy: "I could imagine a machine that were capable of spinning indefinitely" is something it's not hard to imagine someone saying.
Moral realism, if it's true, actually make some predictions. For one, it says you should _not_ expect this giant space-faring civilization that cannibalizes and destroys other worlds. Also, you should expect history to get "roughly better" over long periods of time, absent some cataclysm like an asteroid wiping everything out. That 'roughly better' is a consequence of evolution pushing things in that direction.
There's a separate way of defining morality which also makes predictions, and i think caches out to the same thing: moral behavior is that which scales, over large groups of people and over long periods of time. So moral claims just become claims about which kinds of behavior are generative and which self destructive at the individual and collective levels.
"…moral antirealism is consistent with thinking that morality is real." I mean, that moral beliefs correspond to something real, and thus a person can have incorrect moral beliefs. I thought anti-realism was the idea that there's no reality to which morals correspond. Do i have that wrong?
Moral antirealism is simply the view that there are no stance-independent moral truths. It is not the view that morality isn't "real." That would never be an acceptable definition because "real" is extremely ambiguous, especially among philosophers. Antirealists might say they think morality isn't "real," but some wouldn't say this.
Correspond? Real? Moral beliefs? Reality? I don’t think those words are helping you clarify what you mean. It kind of just seems like you’re saying that moral antirealists believe morality is fake or something similar. If that’s the case then yes you’ve got that wrong.
I get very frustrated with these sorts of arguments about intuitions. Like, if anyone's argument reduces to 'you need to have x intuition' and intuitions are all you have to base your opinion, I just don't care about what you do. Your field is of no value/interest/truth-potential producing to me whatsoever.
As for what evidence I'd personally admit for moral truths, it is rather easy. First, empirical evidence, like that of the natural sciences. Second, lacking that, something like indispensability of mathematics in the physical sciences. Third, perhaps (more tentative in this one), a solid, watertight, irrefutable logical argument that all arguments for ethical antirrealism are logically inconsistent and self-contradictory. If you have none of these, I am sorry, I will be “deliberatively out of reach” of any such arguments.
Typo here, btw: "Since I think the notion of irreducibly normativity is meaningless, I don’t think it’s possible in principle for there to be any evidence for it, even broadly construed". You mean 'irreducible'.
"This was a direct response to my original note where I made a shortened version of the some of the remarks above," -> of some of the remarks...
"Note only" -> Not only
"how the is that not the sort of thing" -> How is that...
"and either are empirical tools aren’t good" -> and either empirical...
I have some sympathy for the idea that we have realist moral intuitions but I think these intuitions are quite superficial and do not stand up to much scrutiny.
What is the nature of these "realist moral intuitions"? Are they a fragile echo of a the truth that is "out there somewhere"? I say fragile echo because we know we have moral disagreement and that must surely mean the "real" truth is patchy. What is the mechanism by which this fragile echo is constructed? We could hypothesize evolution. If you're looking through the things perceived as moral facts how do you establish which are purely evolutionary artifacts and which ones are "real" moral facts? Presumably there all perceived moral facts that are consistent with evolution (otherwise we'd have evidence against evolution). In which case the parsimonious explantion for these is evolutionary selection. Unless you're going to claim some more profound linkage between evolutionary selection and moral facts. Without this explanation I think intuitions of "real" moral facts seem like them must just be a mirage.
In my opinion, you are the most worthy opponent of metaethical realists.
Do you have a core article that best articulates your stance on normative facts being “real” or “not real” and in what ways?
It's sort of like having intuitions about sentient life on other planets and jumping to the conclusion that all those UFOs being seen are real spaceships.
I used to be a moral anti-realist, and I'm now in the moral realist camp. I totally understand both the frustration Christian is expressing here, as well as your stance. You believe you're correct and you're not seeing substantive arguments. I also agree that moral realists need to bite the bullet and make predictive claims.
I think the right way to frame moral realism as a predictive claim is to say:
- moral intuitions evolved because they helped our ancestors perpetuate their genes into the future
- thus, moral intuitions are approximations of strategies that actually help genes perpetuate themselves into the future
- thus, if moral realism exists, we should expect some belief structures to lead to thriving, others not to, over very long periods of time. The objective reality that moral intuitions point at is 'what actually works for life, long term'.
So here's a solid prediction i'm happy to make: no "group" that believes morality isn't real is real is going to be able to reliably perpetuate itself into the future more than a few centuries. If morality really is about long term survival and thriving, we should expect that groups will thrive in proportion to how accurately they understand morality.
hat word "group", it problems needs more defining. The time horizons here are necessarily large, because we're talking about perpetuating yourself arbitrarily far into the future, and likely talking about probabilistic outcomes morose than guaranteed outcomes.
No, you can't do experiments on this - but that may be a 'not yet' thing. If we had, say, very realistic reality simulators and could run them for very long periods of time, we maybe very well could put some of these ideas to the test. Or, if we had WAY more written history than we currently do, we could look at which groups are able to keep existing multiple generations and which ones aren't.
The fact that we can't test these ideas now probably has more to do with technological limitations. But, among other things, this theory explains why some traditions from the ancient world - like, say, the Jews, or reading Confucius, ore the teachings of the Buddha - are still around today. Longevity of a set of moral ideas becomes evidence for their fiteddness to moral reality.
"If morality really is about long term survival and thriving, we should expect that groups will thrive in proportion to how accurately they understand morality."
I do not understand exactly what you mean by this. Survival depends on doing what works - I don't see the connection with accurately understanding why what works...works.
"The objective reality that moral intuitions point at is 'what actually works for life, long term'."
I agree. But it does not follow that understanding why it works is necessary - nor 'believing' morality is 'real'. Believing moral norms are 'necessary' is all that is required. But believing they are true, whatever that means, is not.
Anthropology will show you this. There many hunter gatherer groups that have very specific and complex steps in food preparation. Failure to follow these steps leads to poisoning. They know that failure to follow these steps will lead to poisoning, but they don't know why. They just know what works - but they don't necessarily know why.
"no "group" that believes morality isn't real is real is going to be able to reliably perpetuate itself into the future more than a few centuries."
Again, I don't follow this reasoning. Following the rules of grammar are necessary for communicating with others. But 'believing' in the rules of grammar: I don't know what that could mean?
"thus, if moral realism exists, we should expect some belief structures to lead to thriving".
Yes, but thriving in one environment does not equate to thriving in a different environment. Hence, the difference we find in moral norms from one culture to the next. Species can survive over great periods of time so long as the environment remains relatively stable. Once the environment changes they can find themselves in trouble.
Anyway, I am sympathetic to your views. But I think you would be better to describe your position as a realist in spirit, rather than realist in any philosophical, meta ethical sense.
Thank you for the answer here. Your objections are helping me evolve how I'm phrasing what I believe, which is great.
The example of hunter gatherers following a complex recipe they don't understand really sharpens the point. I agree, that approach works, only until the environment changes. And then it breaks.
But if you understand _why_ the process works, now you have a shot at changing it to adapt for the changing environment. I think the reality that tells us whether moral claims are true, is if they can 'scale' across large periods of time, large groups of people, and large groups of different environments.
I see your point on grammar, but i think a better analogy is the periodic table of the elements. It's clearly socially constructed. You could draw it any which way. But only one way of drawing it actually helps you navigate the reality of chemistry. If all you have are norms, but no theory of _why those norms work_, you can't intentionally adjust the norms when the environment changes. You can't say, "this norm used to work because the environment was like X, but now it's not, so we need to change the norm. " Likewise, if, for some reason, our universe had fluctuations in fundamental constants, say the relative strength of the weak vs the strong force, we _would_ have to rewrite the periodic table. But we could only do that if we understood _why_ it was shaped that way to begin with.
On thriving not translating between environments: I think this is only sort of true. Some norms work across environments. One example would be, "you should try to understand the environment." That one will help you regardless of where you are. The term AI safety researchers use is "convergent instrumental subgoals". I'm not sure how widespread that term is, but what I believe is that convergent instrumental subgoals are at least a good approximation of moral reality, maybe in the same way that Newtonian physics is a good approximation of the true laws of physics, at least in the domains we could access prior to ~1850 or so.
Some will object that convergent instrumental subgoals don't just give you morality. I think that's ultimately wrong, but that's a much larger claim. I have substack post here, for example, arguing that love and sacrifice have powerful mathematical properties:
https://apxhard.substack.com/p/the-surprising-computational-properties
"But if you understand _why_ the process works, now you have a shot at changing it to adapt for the changing environment."
I am not saying one is helpless, but it really depends on how much the environment changes. There is countless examples of western explorers dying in harsh environments in which natives to the same area have lived for thousands of years. This because they do not possess the accumulated cultural know how that has been passed down though generation after generation.
"I think the reality that tells us whether moral claims are true".
It tells what is true in the environment we currently inhabit. But it does not tell us what is true independently of any and all environments. The realist position is that what is true is true independently of any and all environments. This is why I think you are mistaken calling yourself a moral realist in any meta ethical sense. A realist in spirit is not committed to meta ethical realism.
As a side note, I do not agree with Lance's anti-realism, for it commits one to a meta ethical position. A a realist in spirit I am not committed to any meta ethical position. Meta ethics in my view is unintelligible regardless of where one stands.
"I think the reality that tells us whether moral claims are true, is if they can 'scale' across large periods of time, large groups of people, and large groups of different environments."
I don't know of any moral norms that have scaled across 'large groups of different environments'.
I do believe in universal moral claims, yes. I think these exist for reasons outside of the scope of physicalism, but they have physical consequences. So i'm hoping to talk about those physical consequences in a way that will work for people who want predictions.
I don't think moral behavior is _guaranteed_ to keep you alive or thriving in any environment. An example here would be, "treat the people around you well ." I think that strategy works for reasons that will show up in _any_ multiplayer game with agents and uncertainty.
One prediction that comes from this theory: we should expect certain practices to emerge in all large scale cultures, and we should even expect narratives that describe how these practices are important and things 'fall apart' when they are broken.
And we see just this. In china, there is the notion of the mandate of heaven being lost. In Hindu culture there is this idea of the cycle of Yugas. In the west we have a notion of empires rising and falling, as they become 'decadant' and 'corrupt'. In the old testament, a similar pattern in show, as the old testament god says "if you don't follow these practices, your neighbors will conquer and enslave you."
All of those stories have this similar shape: "there's a right way to be, and tings fall apart when you deviate from it."
I liked your comment and your attempt at making a predictive claim for realism. I feel some problems in how you seem to be defining moral realism, though. "Moral intuitions are approximations of strategies that actually help genes perpetuate themselves" feels eminently reasonable -I don't think I have a dispute with this-, but this is not a definition most moral realists would accept, as it makes the value of moral principles dependent on a non-moral, pragmatic principle (reproductive/memetic success). More than that, it reduces it to that (so all core 'objective moral truths' that most realists believe would have to be sacrificed, if necessary, to this new Golden Idol, and not be objective and true in themselves).. It also opens itself up to counterexamples: one can imagine that an evolved, transhuman homo sapiens might easily employ strategies for longevity that would likely be considered immoral by realists (i.e, devouring other inhabited worlds in Galactus-like fashion to sustain our individual and/or group flourishing, power and continuity).
I think you're probably right that lots of moral realists would not accept this definition. I think many would refuse _any_ definition, and I agree with Lance here that that if you can't make predictions, and open yourself to counterexamples, you're a lot harder to take seriously from the outside.
The risk that this becomes the 'golden idol' is mediated, I think, by the reality that nobody knows the distant future. Anyone claiming certainty about "this is sustainable" is fundamentally at odds with physics: chaotic systems are real, and play big roles in human affairs. This means we can't calculate precise outcomes across long time horizons.
Yes I get that people can imagine all kinds of things, but you could have made the same objection to the law of conservation of energy: "I could imagine a machine that were capable of spinning indefinitely" is something it's not hard to imagine someone saying.
Moral realism, if it's true, actually make some predictions. For one, it says you should _not_ expect this giant space-faring civilization that cannibalizes and destroys other worlds. Also, you should expect history to get "roughly better" over long periods of time, absent some cataclysm like an asteroid wiping everything out. That 'roughly better' is a consequence of evolution pushing things in that direction.
There's a separate way of defining morality which also makes predictions, and i think caches out to the same thing: moral behavior is that which scales, over large groups of people and over long periods of time. So moral claims just become claims about which kinds of behavior are generative and which self destructive at the individual and collective levels.
…moral antirealism is consistent with thinking that morality is real. I don’t see what moral realism has to do with evolution.
Can you help me understand this:
"…moral antirealism is consistent with thinking that morality is real." I mean, that moral beliefs correspond to something real, and thus a person can have incorrect moral beliefs. I thought anti-realism was the idea that there's no reality to which morals correspond. Do i have that wrong?
Moral antirealism is simply the view that there are no stance-independent moral truths. It is not the view that morality isn't "real." That would never be an acceptable definition because "real" is extremely ambiguous, especially among philosophers. Antirealists might say they think morality isn't "real," but some wouldn't say this.
Correspond? Real? Moral beliefs? Reality? I don’t think those words are helping you clarify what you mean. It kind of just seems like you’re saying that moral antirealists believe morality is fake or something similar. If that’s the case then yes you’ve got that wrong.
Moral realism:
1. Moral statement is truth-apt/truth-bearer/propositional
2. Some moral statements are true.
3. The truth or falsity of these moral statements are stance-independent/objective.
Moral anti-realism:
Rejects at least one of the above.
I get very frustrated with these sorts of arguments about intuitions. Like, if anyone's argument reduces to 'you need to have x intuition' and intuitions are all you have to base your opinion, I just don't care about what you do. Your field is of no value/interest/truth-potential producing to me whatsoever.
As for what evidence I'd personally admit for moral truths, it is rather easy. First, empirical evidence, like that of the natural sciences. Second, lacking that, something like indispensability of mathematics in the physical sciences. Third, perhaps (more tentative in this one), a solid, watertight, irrefutable logical argument that all arguments for ethical antirrealism are logically inconsistent and self-contradictory. If you have none of these, I am sorry, I will be “deliberatively out of reach” of any such arguments.
Typo here, btw: "Since I think the notion of irreducibly normativity is meaningless, I don’t think it’s possible in principle for there to be any evidence for it, even broadly construed". You mean 'irreducible'.
"This was a direct response to my original note where I made a shortened version of the some of the remarks above," -> of some of the remarks...
"Note only" -> Not only
"how the is that not the sort of thing" -> How is that...
"and either are empirical tools aren’t good" -> and either empirical...
Thanks for catching the typos.