I feel like you’re not understanding the argument that’s being made. It’s not that counterintuitive case judgments don’t serve as counter examples. It’s that you shouldn’t trust real world case judgments if they were formed without considering pertinent information (say, that nearly all suffering on earth is had by wild animals). Also, for the record, I don’t support wiping out life on earth and I don’t think the others do either, largely for trans humanist reasons.
Can you trust the real world case judgment that drinking diet coke is ok if you hadn't considered the pertinent information that it frustrates the natural faculty of hydration? If you CAN trust your diet coke judgment and reject natural law theory on that basis, how is that disanalogous to trusting your judgment that human civilization matters more than shrimp, and therefore concluding that either shrimp pain is not morally bad or utilitarianism is false?
Well no, then you should think about whether it seems wrong to drink diet coke after being aware that it has the functions that it does. When I do that, however, it still seems diet coke is fine to drink.
It's good to imagine hypothetical cases--e.g. that of drinking of some special diet coke that doesn't bring about natural faculties--and form judgments about it. But what you should not do is cling strongly to actual judgments formed about cases before one was informed of pertinent details.
As an analogy, you should not be wildly confident that drinking diet coke is fine if you learned that, say, drinking diet coke causes thousands of far-away aliens to die. You should not say "well everyone knows drinking diet coke is fine, and it's as obvious as any judgment." You've now learned something that makes it no longer obvious.
In short, intuitive judgments should not gain extra purchase from being about the actual world, and generally should gain less. So let me give you an instance of a fine intuition and then a not fine intuition.
Fine intuition: error theory implies it isn't wrong to torture babies if doing so provided no benefits. This is clearly absurd. So error theory is false.
Not fine intuition: according to the many worlds interpretation, random actions we do sometimes bring about the creation of staggering numbers of people. If this is right, then it means pro-natalism has utterly ridiculous implications. It implies that, e.g., driving to the store to purchase a murder weapon is really good so long as it increases the amount of universe splitting (after all, most pro-natalists aren't extreme deontologists, so they'd hold that one should kill to bring about a sufficient amount of good). Thus, the pro-natalist should think that it's much better if you kill someone and increase the amount of universe splitting than staying home! In fact, the pro-natalist thinks if many worlds is true, then our intuition that it's better to help suffering people than go to the beach is flatly wrong--it's better to go to the beach and increase universe splitting.
This is a bad argument because it appeals to intuitions formed before we learned some surprising factual claim as binding. But as the facts change, so should your moral judgments! If the facts about the actual world are surprising, then your judgments about the actual world should be as well.
So now let's consider, for instance, the judgment that shrimp suffering is a much bigger deal than human woes.
Now it's true, that considered on its own is counterintuitive. But it's also quite obvious we shouldn't simply trust our untutored intuitions about it! We're systematically biased against shrimp because they look ugly and weird--we don't empathize with them. We also display a bias called scope neglect where, because a million and a billion hit us intuitively the same way, we intuitively grasp the importance of mindboggling numbers
So to eliminate that bias, we should instead imagine that the shrimp look humanoid. Imagine creatures that look like people cry out in pain. To correct for scope neglect, rather than comparing 8 billion people to 10^18 https://reducing-suffering.org/how-many-wild-animals-are-there/#Terrestrial_arthropods marine arthropods, we should compare each human to some fixed number of arthropods.
When we do this, we conclude there are about 100 million marine arthropods for every human. So now imagine: if you were in a cave with 100 million humanoid creatures, that probably suffered, potentially quite intensely, would your interests be essentially all that matters? If you were in a cave with 100 million creatures that were essentially like super cognitively disabled people--meaning this would have about as many creatures as there are people in the United States--would it really be plausible that your interests are basically all that matters?
When one takes seriously the realization that suffering is bad because of how it feels--your headaches are bad because they hurt, not because you're smart--and realizes that animals in nature experience potentially millions of times more pain than we do each year, then wild animal suffering begins to look very serious!
"Fine intuition: error theory implies it isn't wrong to torture babies if doing so provided no benefits. This is clearly absurd. So error theory is false."
Whether something is "clearly absurd" or not can only be judged from one's personal perspective. I find nothing at all absurd about this, and I don't care if you do. Different people will think different things are absurd. There is no dialectical force to appealing to your personal judgment on the matter. Others simply don't have the same intuitions as you.
Yes I would agree that if a person doesn't share the intuition they should not be moved by it. But I'd be very confident that most people would reject the statement "torturing babies for fun isn't wrong."
I would reject that statement. Such a response is consistent with being an antirealist. Just about the only antirealist position that wouldn't do so is error theory, and most laypeople wouldn't understand error theory or the context in which the error theorist would say such a thing to understand why they'd respond that way. Expressivists, quasi-realists, constructivists, appraiser relativists, and proponents of view like mine would all reject that statement.
Appealing to the responses of untrained people regarding normative sentences that don't explicitly outline a specific metaethical interpretation is useless, as is attempting to figure out whether people are moral realists from the armchair. It's especially useless if your only reference point is one or a handful of languages (out of the 7000+ in the world) and one or a handful of cultures. Armchair analysis of English is not a good way to figure out how human beings think.
There are no good arguments that most people are moral realists. Even "linguistic intuitions" don't consistent do so; see e.g.,:
Franzén, N. (2024). The presumption of realism. Philosophical Studies, 181(5), 1191-1212.
And in any case intuitions are not a good way to determine empirical truths about what other people think.
I believe you and I have discussed before my stance on the "most people are moral realists," claim.
As you'll note, the claim you were criticizing was a counterexample to error theory. So it doesn't matter if it's a broader counterexample to moral anti-realism--I wasn't intending to provide such a thing.
I basically agree with what BB said: We *should* start with particular examples; but we should start with particular *hypothetical* cases rather than actual cases.
This I think what you added in the square brackets when you quoted my post is actually very important. It shouldn't say "particular actions" but "actual actions." The problem is that the world is "descriptively opaque," if you will--we don't have direct access to all the descriptive truths of the world. But our moral judgements require that descriptive facts are assumed to get off the ground, hence why we should start from hypotheticals, not actual cases.
As for the theism point, that's interesting, but I don't think it works. Surely you think there has been moral progress, but someone before slavery was abolished could have said the same thing as you do. Maybe it's just part of God's plan that we morally grow as a species. Whatever you say about previous moral development given theism will likeky apply here too.
I agree that we should not dismiss hypotheticals. Some judgments about particular cases do require a background of empirical premises. But I think the objection to shrimp-welfarism is different. I claim that even in principle, human civilization could never be outweighed by shrimp suffering.
Can you elaborate on what you meant by "but someone before slavery was abolished could have said the same thing as you do."?
Well, the argument you're objecting to in this post (at least the argument I was making in my post), was simply that you shouldn't use your judgements about the actual world--which you might not have made if you were considering a hypothetical--to moorean shift away from any moral theory or descriptive claim. So if yours is an in principle argument, I don't think it's really relevant.
On the second thing, I mean that there have been developments in our "moral circle" throughout time. At one point many considered it fine to have slaves, or good to brutally kill others, simply because they were from a different tribe. Imagine people started making arguments that actually it might not be so good to rape and kill the tribe next-door. To such arguments Limon Saird could have said: "Well God wouldn't have wanted us to have faulty moral intuitions, and so far we have thought it fine to kill others. Everything we have worked towards, every tribe we have sacrificed ourselves to conquer, would have been in vain and been deeply immoral. Surely God wouldn't have allowed this, so it can't be that it would be wrong to kill and rape others." From our POV this is clearly a bad argument, and so it serves as a counterexample to your idea that God wouldn't have given us wrong moral intuitions: It can't be very unlikely that God would allow us to have acted immorally all this time, as he has done so in the past many times.
I don't believe in "moral circle" theory for a couple of reasons.
1. I don't think that concern for others is the only dimension of morality. The crusade against personal alcohol use was considered a moral crusade, and that was about upright personal conduct as much as concern for others.
2. I don't believe that people actually thought that slavery, rape and conquest were morally good. People thought they were inevitable facts of life, but that's not the same as thinking they are good. Americans in the 1700s openly admitted that slavery was a "necessary evil" and many, including Thomas Jefferson, hoped that black Americans could soon return to Africa. Ancient tribes often conquered each other out of fear, or competition for very scarce resources, but I don't believe that most thought it was a morally unproblematic thing to do. People today don't actually believe that factory farming is morally justified; that's why they get defensive when you bring it up. They think that factory farming is just the way of the world. If there were a very easy way for them to switch to non-factory farmed meat, they would do so.
I think that when we make moral judgements, we do so while assuming that certain descriptive facts are the case (I think this is right regardless of your metaethics). But in the actual world we often make many inferences about what the descriptive facts are, which may later turn out to be wrong--and if they turn out to be wrong, our previous judgements no longer hold any weight.
For example, if I see some kids burning a cat alive, I infer that there's a conscious creature there who's feeling a lot of pain, etc. Hence I make the judgement that they're doing something wrong. But if I then find out that it's a plushie they're burning, or that the cat somehow doesn't feel anything or enjoys it, or whatever, then my judgement changes. And importantly the fact that I just thought I was witnessing something very immoral should no longer matter for my judgement at all.
In hypotheticals, we can stipulate the descriptive facts to be some way, and so they should serve as the way to test our moral theories.
I am of the persuasion that the principles you have just outlined are not just applicable to philosophy in general, but are also applicable to jurisprudence, especially if one were to write a legal code himself.
I think your move to ground morality in God-given moral intuitions (and ultimately in God himself) affords you an enormous advantage over your Utilitarian interlocutors. I’ve seen a few comments trying to challenge your view, based on specific examples of a widely held moral beliefs being incorrect (like the permissibility of slavery), but I don’t think these have much force against your starting position—assuming your view is somewhat sympathetic to natural law theory (despite the Diet Coke issue) and would admit that cognitive errors and excess passion can surely lead to ill-formed moral reasoning. In other words, I don’t think examples of mistaken “moral intuitions” can do enough to negate your larger point that we can, to a high degree, trust that our most basic moral intuitions are reliable, as they are instilled in us by God, who is goodness itself. I think the point becomes more self-evident when we consider how easily we can identify a swath of universal goods that our morality sense seems oriented towards.
Defending and partially refining Simon's view as presented here, we actually see that the moral intuitions used to forbid slavery have always been there in human beings. In societies that permitted slavery, philosophers who wanted to uphold the institution have always given excuses as to why it is okay to enslave one's fellow man. These excuses are often to the effect that the fellow man really isn't fellow at all, but in some sense lesser. He is either of a lesser race or people (as evidenced by the fact that his people lost the war that caused them to become enslaved, or whatever other excuse may be given), or he is guilty of some personal or communal crime which justifies his enslavement, or he is simply not a person: his race is one of semi-intelligent beasts; he is not a man. These excuses betray the fact that they share the moral intuition that it's wrong to enslave one's fellow man. Whoever wants to enslave his fellow man (and feel morally justified in doing so) must either make him out to be not his fellow or not a man.
I have been a vegan for a very long time, and I think the world would be better if we granted animals more consideration on a societal level to the extent we can afford to. There's an obvious need for a reciprocal element to morality too though. While we may care a bit for shrimp and especially more intelligent animals, most of them care much less about us. It feels silly to point out this fact, but here we are.
I think this principle also extends to humans (though of course we should give more consideration to humans than other animals). If you are working with a human group who does not have a reciprocal moral relationship with your own, it's okay to acknowledge that. Simulations and history show that overly sacrificial cooperation with other groups can be counterproductive.
It is considered insane or evil to people on the furthest left side of the political spectrum for groups perceived as more powerful to have a bias towards their own group. I think it's perfectly rational to do so, and it prevents one-sided exploitation of the care principle. We should extend some level of care to other societies, and to a lesser degree other animals. We should practice virtue and set a leading example. We shouldn't have those things dictate our moral frames though. They are secondary.
That is all to say that: the "care" moral foundation is important, but pure utilitarianism is moral cuckoldry.
This was an interesting post, and I feel like I agree with the core of it. I would like to clarify two objections, though:
1) Your focus on the shrimp-welfare argument seems a tad uncharitable and ultimately, unnecessary. I imagine you focus on it because, from a rhetorical perspective, it is very effective: most ordinary people would consider fretting about it as an indication of weirdness + stupidity. But I think it is just simpler and clearer to start by a rejection of the moral premise that sustains it: ethics as reducible to pain=bad, pleasure=good and a moral law to impartially maximize utility amongst all sentient beings. There is no reason why one has to accept the premise: it both theoretically and practically conflicts with other, equally plausible premises, and it has no objective footing on which to force acceptance.
2) "The way that we get our moral principles is that we consult our intuition about particular moral questions". I don't think this is the best map for how we get our moral principles. I'd say a better one is that we are endowed by natural evolution with some altruistic and cooperative tendencies (a byproduct of ultimately egoistic and self-serving gene, individual and group strategies) that get reinforced by a history of cultural evolution that favors groups that create strong mechanisms for greater (in numbers and in degree) collaboration among human groups. Because humans are smart enough to reflect on the reasons for their actions and to develop mental schema, models and systems, we have built sundry moral systems and taken them as 'true', 'objective', and as guideposts for human behavior. And most of them converge a lot in practice because ultimately, their intersection is composed of sets of practices that help groups become stronger, more powerful, more prosperous both at individual and group level.
I think that the actual mental process by which people come to their moral intuitions is that they consult their intuitions and then they formulate general principles based on those intuitions.
This looks like a typo: "Key Takeaway #3: Common Sense moral judgments must broadly right because God would not equip us with wildly mistaken moral intuitions." I think you're missing "be" between "broadly" and "right."
I’ve argued that population utilitarianism systematically under-estimates the value of just being alive. Utilitarianism would not generate so many wildly counter-intuitive results if they just factored in a positive value to being a shrimp just being able to live, move around, grow, be part of a community of shrimp, experience life in whatever way shrimp experience life, and have agency to make decisions that are purely its own - versus non-existence. This is quite intuitive to my Christian understanding of life. I don’t know on what basis utilitarians reject this idea.
Or maybe the wildly counterintuitiveness is the point. A “new” theory that just affirms common sense Christian understanding would not generate the prestige and clicks, and would be more difficult to monetize.
Pure utilitarianism is deeply silly for the reason you note. I'd call it a religious impulse. People love to have simple rules handed down from above and then argue about whose harsh, nonsensical interpretation is most true to said from-on-high rules. A good example that doesn't require one to acknowledge the vast suffering humanity imposes on animals is from the EA community, where its strict utilitarian elements reject women's rights as important because they mean fewer babies, so less total (human) utility.
But your proposed alternative is equally blinded. Humans are very bad at thinking systemically. And we have understandable but very real biases in favor of the local and small bore. You can't just pretend those away with a different blanket rule that we can only induct moral or ethical frameworks from common sense, one at a time examples. That's just as blinkered as the utilitarians.
This remark: “if it really were the case that common sense had been completely wrong throughout most of human history, that would be an argument for moral nihilism” looks like it might imply people are moral realists and that if they were wrong about this we’d be left with moral nihilism. If so, then I think the claim itself is mistaken about what common sense consists in.
Common sense throughout history has not involved the “intuition” that “pain is bad.” People have disliked experiencing pain and not wanted at least some other people/living things to experience pain, but this has never required nor involved some sort of “intuition” nor has it involved any sort of commitment to the notion that “pain is bad” is true in any particular, metaphysically distinctive metaethical respect. People don’t typically like pain. But that doesn’t mean they’re e.g., moral realists. I'm not quite sure on what you meant by those remarks though so apologies if I've misunderstood.
I think this is almost there, but I would avoid conflating a moral intuition and a moral judgment. The judgment that it's wrong to kill Tom because he laughed at your bad haircut is a moral judgment, and it's built on a moral intuition. We don't receive the intuition from the judgment. We can see this because, if you asked someone why it's wrong to kill Tom, they'd answer something like "because it's murder" or "because killing is wrong," or, if they're a bit more thoughtful "because it's wrong to intentionally kill innocent people."
They are giving you their moral reasoning here implicitly.
1. It's wrong to kill innocent people (intuition)
2. Tom is an innocent person (implicit assumption)
3. Therefore, killing Tom is immoral (judgment)
Moral intuition is trustworthy for all the same reasons our other intuitions are trustworthy, as you outline here. If it's not trustworthy on a basic level, then doing ethics is pointless because it's impossible. You're spot on there.
But we do refine our basic moral intuitions by examining judgments. We don't form the intuition from the judgment. The intuition has to exist before our first judgment can be made. The basic moral intuition is something like "it's wrong to cause people pain." Then we observe that sometimes it's actually okay to cause people pain, for instance when we need to do surgery. Sometimes, pain is also necessary for justice. And, we see that it's better if we cause as little pain as possible whenever we need to cause it. So then we refine our intuition to say something like "it's wrong to cause people undue and/or undeserved pain."
You badly distort Catholic natural law theory when you say that it has the implication of saying that drinking Diet Coke is evil because it's a drink that dehydrates you.
In the first place, the principle you're criticizing, the perverted faculty argument, can be best stated as: "Whatever violates the natural end of a faculty is intrinsically evil." To give two examples:
1. The sexual faculty has a natural end in procreation and unity; to use it in a way that intentionally frustrates that (e.g. contraception) is deemed intrinsically disordered.
2. The faculty of speech has a natural end in communicating truth; thus, lying is intrinsically disordered.
Now, there is no "drinking" faculty. Drinking is an act that uses several faculties, primarily the nutritive faculty, which is aimed at bodily preservation and well-being.
Thus, by the argument's logic, to knowingly use it to consume something that cuts against your bodily preservation and well-being, such as poison, would be evil.
Now, Diet Coke is not a poison. Drinking a little bit of it will not undermine your health. Therefore, a Catholic ethicist would say that drinking a Diet Coke is, at most, a misuse (or imprudent use), not a perversion of the faculty. You are not actively trying to frustrate hydration in drinking Diet Coke; you are simply consuming something that may have other effects (perhaps it adds a bit of flavor to your meal). The act is not essentially disordered.
The only way drinking Diet Coke would be evil is if you either drank it to excess or you drank only Diet Coke to harm your body.
That argument doesn't work because an exactly analogous argument would hold that homosexual acts aren't a perversion of a natural faculty either. Using Diet Coke just because it tastes good is like doing homosexual acts just because they feel good (to the homosexual).
Drinking only Diet Coke frustrates the end of the nutritive faculty.
Drinking Diet Coke as part of a meal does not frustrate that faculty.
Doing homosexual acts only frustrates the reproductive faculty.
Doing homosexual acts in addition to heterosexual acts does not frustrate the reproductive faculty.
Or more realistically, when a couple does contracepted sexual acts in addition to non-contracepted sexual acts, their reproductive faculty is not frustrated.
Simon, your argument appears to be something like this:
(a) If a faculty is only sometimes used contrary to its natural end, but also used in ways consistent with that end, then its end is not frustrated as a whole.
(b) Therefore, the partial or occasional frustration of a faculty is morally permissible.
There are two problems with (a).
First, the Perverted Faculty Argument does not concern statistical averages but individual actions. Drinking a Diet Coke is not intrinsically disordered, because drinking is not per se ordered to nutrition. Drinking is not a "faculty" but a generic bodily act, and it is an action that can have numerous purposes. Not every faculty of drinking engages the nutritive act directly. By contrast, homosexual acts and contraceptive sexual acts are disordered in themselves, regardless of whether other kinds of sexual acts are also performed, because the particular act is essentially non-procreative and non-unitive in the marital sense.
Evil is not diluted by good. A good use of a faculty elsewhere does not redeem a disordered use.
Second, the analogy (a) depends on is faulty because drinking Diet Coke is not like having contraceptive sex. The act of drinking Diet Coke may not achieve the end of hydration, but it doesn't directly frustrate hydration. Therefore, it is not intrinsically opposed to the nutritive faculty's purpose. However, the act of contraceptive sex acts (and homosexual acts) deliberately excludes procreation, positively frustrating the end, so it is contrary by its very design to the natural end of the sexual faculty.
Drinking Diet Coke is a permissible but imperfect use, while contraceptive sex is a deliberate negation. For drinking a substance to be analogous to contraceptive sex acts, the liquid consumed (let's called Contra Cola) would need to be something that neutralizes nutrition by preventing the absorption of nutrients, allowing for someone to engage in the outward act of eating (presumably for the pleasure of the experience), knowing that it will not nourish him in the slightest. In that case, drinking Contra Cola would be like doing contraceptive acts: it would be immoral every time you did it, not just when you did so excessively.
The “Diet Coke” argument fails because natural law theory evaluates the moral structure of individual acts, not their nutritional outcomes. It only declares intrinsically wrong those acts that intentionally pervert a faculty’s natural end, not merely fail to fulfill it.
You may have buried the lede. If the position of all the major EA/LW and AI-alignment figures is that extermination of all life is preferable to our current world, that's a huge story. Most people, first being exposed to the concept of AI alignment, assume that it means "let's make sure the robots don't kill us all." If it instead means "let's make sure the robots become utility monsters before they kill us all" then they ought to be opposed. Perhaps I misunderstood the position, or perhaps the position is a minority one. But if not, this needs to be shouted from the rooftops.
> On a technicality, they may be wrong here because contrary to popular belief, Diet Coke does not actually dehydrate you. The caffeine it contains is not enough to dehydrate you by more than the amount of water in the Diet Coke. However, it’s possible in principle that there could be a beverage which dehydrates you, and it’s a devastating objection to the Catholic Natural Law theory that that theory says that it would be morally impermissible to drink that beverage, even if doing so would avert the destruction of the Earth.
Consider how contrived a scenario you have to concoct for your "Catholic Natural Law theory" example. That should be a sign that you're trying to abuse moral intuition.
You admit that Diet Coke doesn't actually have the properties you need for your example. A hypothetical drink the dehydrated you but you still wanted to drink would probably imply that it's seriously addictive so probably would be immoral to drink. Hence your need to add the implausible "avert the destruction of the Earth" condition.
I feel like you’re not understanding the argument that’s being made. It’s not that counterintuitive case judgments don’t serve as counter examples. It’s that you shouldn’t trust real world case judgments if they were formed without considering pertinent information (say, that nearly all suffering on earth is had by wild animals). Also, for the record, I don’t support wiping out life on earth and I don’t think the others do either, largely for trans humanist reasons.
Can you trust the real world case judgment that drinking diet coke is ok if you hadn't considered the pertinent information that it frustrates the natural faculty of hydration? If you CAN trust your diet coke judgment and reject natural law theory on that basis, how is that disanalogous to trusting your judgment that human civilization matters more than shrimp, and therefore concluding that either shrimp pain is not morally bad or utilitarianism is false?
In a previous post you said that it would be good to exterminate all wildlife on earth. I made a correction on this post to say that you do not support the extermination of humans. https://benthams.substack.com/p/an-open-letter-to-tree-huggers
Well no, then you should think about whether it seems wrong to drink diet coke after being aware that it has the functions that it does. When I do that, however, it still seems diet coke is fine to drink.
It's good to imagine hypothetical cases--e.g. that of drinking of some special diet coke that doesn't bring about natural faculties--and form judgments about it. But what you should not do is cling strongly to actual judgments formed about cases before one was informed of pertinent details.
As an analogy, you should not be wildly confident that drinking diet coke is fine if you learned that, say, drinking diet coke causes thousands of far-away aliens to die. You should not say "well everyone knows drinking diet coke is fine, and it's as obvious as any judgment." You've now learned something that makes it no longer obvious.
In short, intuitive judgments should not gain extra purchase from being about the actual world, and generally should gain less. So let me give you an instance of a fine intuition and then a not fine intuition.
Fine intuition: error theory implies it isn't wrong to torture babies if doing so provided no benefits. This is clearly absurd. So error theory is false.
Not fine intuition: according to the many worlds interpretation, random actions we do sometimes bring about the creation of staggering numbers of people. If this is right, then it means pro-natalism has utterly ridiculous implications. It implies that, e.g., driving to the store to purchase a murder weapon is really good so long as it increases the amount of universe splitting (after all, most pro-natalists aren't extreme deontologists, so they'd hold that one should kill to bring about a sufficient amount of good). Thus, the pro-natalist should think that it's much better if you kill someone and increase the amount of universe splitting than staying home! In fact, the pro-natalist thinks if many worlds is true, then our intuition that it's better to help suffering people than go to the beach is flatly wrong--it's better to go to the beach and increase universe splitting.
This is a bad argument because it appeals to intuitions formed before we learned some surprising factual claim as binding. But as the facts change, so should your moral judgments! If the facts about the actual world are surprising, then your judgments about the actual world should be as well.
So now let's consider, for instance, the judgment that shrimp suffering is a much bigger deal than human woes.
Now it's true, that considered on its own is counterintuitive. But it's also quite obvious we shouldn't simply trust our untutored intuitions about it! We're systematically biased against shrimp because they look ugly and weird--we don't empathize with them. We also display a bias called scope neglect where, because a million and a billion hit us intuitively the same way, we intuitively grasp the importance of mindboggling numbers
So to eliminate that bias, we should instead imagine that the shrimp look humanoid. Imagine creatures that look like people cry out in pain. To correct for scope neglect, rather than comparing 8 billion people to 10^18 https://reducing-suffering.org/how-many-wild-animals-are-there/#Terrestrial_arthropods marine arthropods, we should compare each human to some fixed number of arthropods.
When we do this, we conclude there are about 100 million marine arthropods for every human. So now imagine: if you were in a cave with 100 million humanoid creatures, that probably suffered, potentially quite intensely, would your interests be essentially all that matters? If you were in a cave with 100 million creatures that were essentially like super cognitively disabled people--meaning this would have about as many creatures as there are people in the United States--would it really be plausible that your interests are basically all that matters?
When one takes seriously the realization that suffering is bad because of how it feels--your headaches are bad because they hurt, not because you're smart--and realizes that animals in nature experience potentially millions of times more pain than we do each year, then wild animal suffering begins to look very serious!
"Fine intuition: error theory implies it isn't wrong to torture babies if doing so provided no benefits. This is clearly absurd. So error theory is false."
Whether something is "clearly absurd" or not can only be judged from one's personal perspective. I find nothing at all absurd about this, and I don't care if you do. Different people will think different things are absurd. There is no dialectical force to appealing to your personal judgment on the matter. Others simply don't have the same intuitions as you.
Nice to see you've unblocked me!
Yes I would agree that if a person doesn't share the intuition they should not be moved by it. But I'd be very confident that most people would reject the statement "torturing babies for fun isn't wrong."
I would reject that statement. Such a response is consistent with being an antirealist. Just about the only antirealist position that wouldn't do so is error theory, and most laypeople wouldn't understand error theory or the context in which the error theorist would say such a thing to understand why they'd respond that way. Expressivists, quasi-realists, constructivists, appraiser relativists, and proponents of view like mine would all reject that statement.
Appealing to the responses of untrained people regarding normative sentences that don't explicitly outline a specific metaethical interpretation is useless, as is attempting to figure out whether people are moral realists from the armchair. It's especially useless if your only reference point is one or a handful of languages (out of the 7000+ in the world) and one or a handful of cultures. Armchair analysis of English is not a good way to figure out how human beings think.
There are no good arguments that most people are moral realists. Even "linguistic intuitions" don't consistent do so; see e.g.,:
Franzén, N. (2024). The presumption of realism. Philosophical Studies, 181(5), 1191-1212.
And in any case intuitions are not a good way to determine empirical truths about what other people think.
I believe you and I have discussed before my stance on the "most people are moral realists," claim.
As you'll note, the claim you were criticizing was a counterexample to error theory. So it doesn't matter if it's a broader counterexample to moral anti-realism--I wasn't intending to provide such a thing.
Before the industrial takeoff, when people couldnt have predicted transhumanist futures, should they have wiped out life on earth?
I basically agree with what BB said: We *should* start with particular examples; but we should start with particular *hypothetical* cases rather than actual cases.
This I think what you added in the square brackets when you quoted my post is actually very important. It shouldn't say "particular actions" but "actual actions." The problem is that the world is "descriptively opaque," if you will--we don't have direct access to all the descriptive truths of the world. But our moral judgements require that descriptive facts are assumed to get off the ground, hence why we should start from hypotheticals, not actual cases.
As for the theism point, that's interesting, but I don't think it works. Surely you think there has been moral progress, but someone before slavery was abolished could have said the same thing as you do. Maybe it's just part of God's plan that we morally grow as a species. Whatever you say about previous moral development given theism will likeky apply here too.
I agree that we should not dismiss hypotheticals. Some judgments about particular cases do require a background of empirical premises. But I think the objection to shrimp-welfarism is different. I claim that even in principle, human civilization could never be outweighed by shrimp suffering.
Can you elaborate on what you meant by "but someone before slavery was abolished could have said the same thing as you do."?
Well, the argument you're objecting to in this post (at least the argument I was making in my post), was simply that you shouldn't use your judgements about the actual world--which you might not have made if you were considering a hypothetical--to moorean shift away from any moral theory or descriptive claim. So if yours is an in principle argument, I don't think it's really relevant.
On the second thing, I mean that there have been developments in our "moral circle" throughout time. At one point many considered it fine to have slaves, or good to brutally kill others, simply because they were from a different tribe. Imagine people started making arguments that actually it might not be so good to rape and kill the tribe next-door. To such arguments Limon Saird could have said: "Well God wouldn't have wanted us to have faulty moral intuitions, and so far we have thought it fine to kill others. Everything we have worked towards, every tribe we have sacrificed ourselves to conquer, would have been in vain and been deeply immoral. Surely God wouldn't have allowed this, so it can't be that it would be wrong to kill and rape others." From our POV this is clearly a bad argument, and so it serves as a counterexample to your idea that God wouldn't have given us wrong moral intuitions: It can't be very unlikely that God would allow us to have acted immorally all this time, as he has done so in the past many times.
I don't believe in "moral circle" theory for a couple of reasons.
1. I don't think that concern for others is the only dimension of morality. The crusade against personal alcohol use was considered a moral crusade, and that was about upright personal conduct as much as concern for others.
2. I don't believe that people actually thought that slavery, rape and conquest were morally good. People thought they were inevitable facts of life, but that's not the same as thinking they are good. Americans in the 1700s openly admitted that slavery was a "necessary evil" and many, including Thomas Jefferson, hoped that black Americans could soon return to Africa. Ancient tribes often conquered each other out of fear, or competition for very scarce resources, but I don't believe that most thought it was a morally unproblematic thing to do. People today don't actually believe that factory farming is morally justified; that's why they get defensive when you bring it up. They think that factory farming is just the way of the world. If there were a very easy way for them to switch to non-factory farmed meat, they would do so.
Why should we start with hypothetical cases?
I think that when we make moral judgements, we do so while assuming that certain descriptive facts are the case (I think this is right regardless of your metaethics). But in the actual world we often make many inferences about what the descriptive facts are, which may later turn out to be wrong--and if they turn out to be wrong, our previous judgements no longer hold any weight.
For example, if I see some kids burning a cat alive, I infer that there's a conscious creature there who's feeling a lot of pain, etc. Hence I make the judgement that they're doing something wrong. But if I then find out that it's a plushie they're burning, or that the cat somehow doesn't feel anything or enjoys it, or whatever, then my judgement changes. And importantly the fact that I just thought I was witnessing something very immoral should no longer matter for my judgement at all.
In hypotheticals, we can stipulate the descriptive facts to be some way, and so they should serve as the way to test our moral theories.
I thought the shrimp welfare thing was a joke when I heard about it the first time. I don’t even know what to say to these people.
I am of the persuasion that the principles you have just outlined are not just applicable to philosophy in general, but are also applicable to jurisprudence, especially if one were to write a legal code himself.
I think your move to ground morality in God-given moral intuitions (and ultimately in God himself) affords you an enormous advantage over your Utilitarian interlocutors. I’ve seen a few comments trying to challenge your view, based on specific examples of a widely held moral beliefs being incorrect (like the permissibility of slavery), but I don’t think these have much force against your starting position—assuming your view is somewhat sympathetic to natural law theory (despite the Diet Coke issue) and would admit that cognitive errors and excess passion can surely lead to ill-formed moral reasoning. In other words, I don’t think examples of mistaken “moral intuitions” can do enough to negate your larger point that we can, to a high degree, trust that our most basic moral intuitions are reliable, as they are instilled in us by God, who is goodness itself. I think the point becomes more self-evident when we consider how easily we can identify a swath of universal goods that our morality sense seems oriented towards.
Defending and partially refining Simon's view as presented here, we actually see that the moral intuitions used to forbid slavery have always been there in human beings. In societies that permitted slavery, philosophers who wanted to uphold the institution have always given excuses as to why it is okay to enslave one's fellow man. These excuses are often to the effect that the fellow man really isn't fellow at all, but in some sense lesser. He is either of a lesser race or people (as evidenced by the fact that his people lost the war that caused them to become enslaved, or whatever other excuse may be given), or he is guilty of some personal or communal crime which justifies his enslavement, or he is simply not a person: his race is one of semi-intelligent beasts; he is not a man. These excuses betray the fact that they share the moral intuition that it's wrong to enslave one's fellow man. Whoever wants to enslave his fellow man (and feel morally justified in doing so) must either make him out to be not his fellow or not a man.
Good point! Thank you for adding this!
I have been a vegan for a very long time, and I think the world would be better if we granted animals more consideration on a societal level to the extent we can afford to. There's an obvious need for a reciprocal element to morality too though. While we may care a bit for shrimp and especially more intelligent animals, most of them care much less about us. It feels silly to point out this fact, but here we are.
I think this principle also extends to humans (though of course we should give more consideration to humans than other animals). If you are working with a human group who does not have a reciprocal moral relationship with your own, it's okay to acknowledge that. Simulations and history show that overly sacrificial cooperation with other groups can be counterproductive.
It is considered insane or evil to people on the furthest left side of the political spectrum for groups perceived as more powerful to have a bias towards their own group. I think it's perfectly rational to do so, and it prevents one-sided exploitation of the care principle. We should extend some level of care to other societies, and to a lesser degree other animals. We should practice virtue and set a leading example. We shouldn't have those things dictate our moral frames though. They are secondary.
That is all to say that: the "care" moral foundation is important, but pure utilitarianism is moral cuckoldry.
I was a vegan for several years and I still avoid most animal products.
I agree with you. Pure utilitarianism is actually very short-sighted because it allows non-cooperators to inherit the future.
This was an interesting post, and I feel like I agree with the core of it. I would like to clarify two objections, though:
1) Your focus on the shrimp-welfare argument seems a tad uncharitable and ultimately, unnecessary. I imagine you focus on it because, from a rhetorical perspective, it is very effective: most ordinary people would consider fretting about it as an indication of weirdness + stupidity. But I think it is just simpler and clearer to start by a rejection of the moral premise that sustains it: ethics as reducible to pain=bad, pleasure=good and a moral law to impartially maximize utility amongst all sentient beings. There is no reason why one has to accept the premise: it both theoretically and practically conflicts with other, equally plausible premises, and it has no objective footing on which to force acceptance.
2) "The way that we get our moral principles is that we consult our intuition about particular moral questions". I don't think this is the best map for how we get our moral principles. I'd say a better one is that we are endowed by natural evolution with some altruistic and cooperative tendencies (a byproduct of ultimately egoistic and self-serving gene, individual and group strategies) that get reinforced by a history of cultural evolution that favors groups that create strong mechanisms for greater (in numbers and in degree) collaboration among human groups. Because humans are smart enough to reflect on the reasons for their actions and to develop mental schema, models and systems, we have built sundry moral systems and taken them as 'true', 'objective', and as guideposts for human behavior. And most of them converge a lot in practice because ultimately, their intersection is composed of sets of practices that help groups become stronger, more powerful, more prosperous both at individual and group level.
I think that the actual mental process by which people come to their moral intuitions is that they consult their intuitions and then they formulate general principles based on those intuitions.
This looks like a typo: "Key Takeaway #3: Common Sense moral judgments must broadly right because God would not equip us with wildly mistaken moral intuitions." I think you're missing "be" between "broadly" and "right."
I’ve argued that population utilitarianism systematically under-estimates the value of just being alive. Utilitarianism would not generate so many wildly counter-intuitive results if they just factored in a positive value to being a shrimp just being able to live, move around, grow, be part of a community of shrimp, experience life in whatever way shrimp experience life, and have agency to make decisions that are purely its own - versus non-existence. This is quite intuitive to my Christian understanding of life. I don’t know on what basis utilitarians reject this idea.
Or maybe the wildly counterintuitiveness is the point. A “new” theory that just affirms common sense Christian understanding would not generate the prestige and clicks, and would be more difficult to monetize.
Pure utilitarianism is deeply silly for the reason you note. I'd call it a religious impulse. People love to have simple rules handed down from above and then argue about whose harsh, nonsensical interpretation is most true to said from-on-high rules. A good example that doesn't require one to acknowledge the vast suffering humanity imposes on animals is from the EA community, where its strict utilitarian elements reject women's rights as important because they mean fewer babies, so less total (human) utility.
But your proposed alternative is equally blinded. Humans are very bad at thinking systemically. And we have understandable but very real biases in favor of the local and small bore. You can't just pretend those away with a different blanket rule that we can only induct moral or ethical frameworks from common sense, one at a time examples. That's just as blinkered as the utilitarians.
Why are you strawmanning Catholic natural law theory? No principle about it says drinking soda is bad.
This remark: “if it really were the case that common sense had been completely wrong throughout most of human history, that would be an argument for moral nihilism” looks like it might imply people are moral realists and that if they were wrong about this we’d be left with moral nihilism. If so, then I think the claim itself is mistaken about what common sense consists in.
Common sense throughout history has not involved the “intuition” that “pain is bad.” People have disliked experiencing pain and not wanted at least some other people/living things to experience pain, but this has never required nor involved some sort of “intuition” nor has it involved any sort of commitment to the notion that “pain is bad” is true in any particular, metaphysically distinctive metaethical respect. People don’t typically like pain. But that doesn’t mean they’re e.g., moral realists. I'm not quite sure on what you meant by those remarks though so apologies if I've misunderstood.
I think this is almost there, but I would avoid conflating a moral intuition and a moral judgment. The judgment that it's wrong to kill Tom because he laughed at your bad haircut is a moral judgment, and it's built on a moral intuition. We don't receive the intuition from the judgment. We can see this because, if you asked someone why it's wrong to kill Tom, they'd answer something like "because it's murder" or "because killing is wrong," or, if they're a bit more thoughtful "because it's wrong to intentionally kill innocent people."
They are giving you their moral reasoning here implicitly.
1. It's wrong to kill innocent people (intuition)
2. Tom is an innocent person (implicit assumption)
3. Therefore, killing Tom is immoral (judgment)
Moral intuition is trustworthy for all the same reasons our other intuitions are trustworthy, as you outline here. If it's not trustworthy on a basic level, then doing ethics is pointless because it's impossible. You're spot on there.
But we do refine our basic moral intuitions by examining judgments. We don't form the intuition from the judgment. The intuition has to exist before our first judgment can be made. The basic moral intuition is something like "it's wrong to cause people pain." Then we observe that sometimes it's actually okay to cause people pain, for instance when we need to do surgery. Sometimes, pain is also necessary for justice. And, we see that it's better if we cause as little pain as possible whenever we need to cause it. So then we refine our intuition to say something like "it's wrong to cause people undue and/or undeserved pain."
You badly distort Catholic natural law theory when you say that it has the implication of saying that drinking Diet Coke is evil because it's a drink that dehydrates you.
In the first place, the principle you're criticizing, the perverted faculty argument, can be best stated as: "Whatever violates the natural end of a faculty is intrinsically evil." To give two examples:
1. The sexual faculty has a natural end in procreation and unity; to use it in a way that intentionally frustrates that (e.g. contraception) is deemed intrinsically disordered.
2. The faculty of speech has a natural end in communicating truth; thus, lying is intrinsically disordered.
Now, there is no "drinking" faculty. Drinking is an act that uses several faculties, primarily the nutritive faculty, which is aimed at bodily preservation and well-being.
Thus, by the argument's logic, to knowingly use it to consume something that cuts against your bodily preservation and well-being, such as poison, would be evil.
Now, Diet Coke is not a poison. Drinking a little bit of it will not undermine your health. Therefore, a Catholic ethicist would say that drinking a Diet Coke is, at most, a misuse (or imprudent use), not a perversion of the faculty. You are not actively trying to frustrate hydration in drinking Diet Coke; you are simply consuming something that may have other effects (perhaps it adds a bit of flavor to your meal). The act is not essentially disordered.
The only way drinking Diet Coke would be evil is if you either drank it to excess or you drank only Diet Coke to harm your body.
That argument doesn't work because an exactly analogous argument would hold that homosexual acts aren't a perversion of a natural faculty either. Using Diet Coke just because it tastes good is like doing homosexual acts just because they feel good (to the homosexual).
https://www.youtube.com/watch?v=oltLyImVkd0
That is not the case, and you didn't actually read what I wrote.
Doing homosexual acts is less like drinking diet coke and more like drinking poison.
Diet Coke does not frustrate the ends of the nutritive faculty like homosexuality does.
Drinking only Diet Coke frustrates the end of the nutritive faculty.
Drinking Diet Coke as part of a meal does not frustrate that faculty.
Doing homosexual acts only frustrates the reproductive faculty.
Doing homosexual acts in addition to heterosexual acts does not frustrate the reproductive faculty.
Or more realistically, when a couple does contracepted sexual acts in addition to non-contracepted sexual acts, their reproductive faculty is not frustrated.
https://wollenblog.substack.com/p/diet-coke-contraception-and-natural
Simon, your argument appears to be something like this:
(a) If a faculty is only sometimes used contrary to its natural end, but also used in ways consistent with that end, then its end is not frustrated as a whole.
(b) Therefore, the partial or occasional frustration of a faculty is morally permissible.
There are two problems with (a).
First, the Perverted Faculty Argument does not concern statistical averages but individual actions. Drinking a Diet Coke is not intrinsically disordered, because drinking is not per se ordered to nutrition. Drinking is not a "faculty" but a generic bodily act, and it is an action that can have numerous purposes. Not every faculty of drinking engages the nutritive act directly. By contrast, homosexual acts and contraceptive sexual acts are disordered in themselves, regardless of whether other kinds of sexual acts are also performed, because the particular act is essentially non-procreative and non-unitive in the marital sense.
Evil is not diluted by good. A good use of a faculty elsewhere does not redeem a disordered use.
Second, the analogy (a) depends on is faulty because drinking Diet Coke is not like having contraceptive sex. The act of drinking Diet Coke may not achieve the end of hydration, but it doesn't directly frustrate hydration. Therefore, it is not intrinsically opposed to the nutritive faculty's purpose. However, the act of contraceptive sex acts (and homosexual acts) deliberately excludes procreation, positively frustrating the end, so it is contrary by its very design to the natural end of the sexual faculty.
Drinking Diet Coke is a permissible but imperfect use, while contraceptive sex is a deliberate negation. For drinking a substance to be analogous to contraceptive sex acts, the liquid consumed (let's called Contra Cola) would need to be something that neutralizes nutrition by preventing the absorption of nutrients, allowing for someone to engage in the outward act of eating (presumably for the pleasure of the experience), knowing that it will not nourish him in the slightest. In that case, drinking Contra Cola would be like doing contraceptive acts: it would be immoral every time you did it, not just when you did so excessively.
The “Diet Coke” argument fails because natural law theory evaluates the moral structure of individual acts, not their nutritional outcomes. It only declares intrinsically wrong those acts that intentionally pervert a faculty’s natural end, not merely fail to fulfill it.
You may have buried the lede. If the position of all the major EA/LW and AI-alignment figures is that extermination of all life is preferable to our current world, that's a huge story. Most people, first being exposed to the concept of AI alignment, assume that it means "let's make sure the robots don't kill us all." If it instead means "let's make sure the robots become utility monsters before they kill us all" then they ought to be opposed. Perhaps I misunderstood the position, or perhaps the position is a minority one. But if not, this needs to be shouted from the rooftops.
> On a technicality, they may be wrong here because contrary to popular belief, Diet Coke does not actually dehydrate you. The caffeine it contains is not enough to dehydrate you by more than the amount of water in the Diet Coke. However, it’s possible in principle that there could be a beverage which dehydrates you, and it’s a devastating objection to the Catholic Natural Law theory that that theory says that it would be morally impermissible to drink that beverage, even if doing so would avert the destruction of the Earth.
Consider how contrived a scenario you have to concoct for your "Catholic Natural Law theory" example. That should be a sign that you're trying to abuse moral intuition.
You admit that Diet Coke doesn't actually have the properties you need for your example. A hypothetical drink the dehydrated you but you still wanted to drink would probably imply that it's seriously addictive so probably would be immoral to drink. Hence your need to add the implausible "avert the destruction of the Earth" condition.
How about a Hard Diet Coke that contained as much alcohol as wine? Doesn't seem very contrived.