Higher-order descriptive properties are cheap. Sure, we can say there is an "objective" property of generosity, with explanatory power in some cultures and circumstances. But there is equally an "objective" property of *dishonor*, which equally "explains" why women are killed by their relatives in some cultures. We don't think the latter has any *normative* significance -- it isn't really a *moral* property -- so mere explanatory significance of this sort doesn't suffice to establish that what you're being a realist about is a genuinely normative property.
I had a whole section about this exact objection, actually, but it was getting unwieldy so I spun it out into a new post I'm working on now! Short answer, though: I think any maximally coherent theory about what moral properties do and don't exist would at least very plausibly conclude that either "dishonor" (to use your example) is not actually a moral property, or that the sorts of things that result in women being killed (sex outside of marriage, adopting western fashion, etc) don't actually instantiate it. In fact, honor killings seem like a really good example of a situation where the moral explanation fails the counterfactual test - if you believe that what a victim did was *not* dishonorable, but she was killed anyway, then you'd have to believe that, even if she wasn't dishonorable, they would have killed her anyway (because that's literally what did happen). Whereas, in the example of the donation, even someone who actively opposed generosity for some reason would still be likely to say that (in conjunction with a wide range of background conditions, like any explanation) the generosity really was what elicited the praise.
* Someone who thinks donating money is *not* generous (e.g. an anti-capitalist who views the money as "ill-gotten gains") will say the act elicited praise even though it wasn't generous.
* killing due to dishonor passes a counterfactual test in that even if the woman hadn't committed *this particular* dishonor, if she'd committed any other comparable dishonor she would have similarly been killed.
Oh, I would say there is no relevant disanalogy in a structural sense - both are totally legitimate assertions someone could make, but in both cases what matters is whether or not they have a theory of generosity or honor that's plausible enough to make us accept them.
The first statement is like an AIDS denialist who asserts that HIV is not deadly, so everyone who died while infected with HIV would have died even if they hadn't had it. There's nothing wrong with that argument *as an argument* - it's not logically fallacious, and if we accept the background assumptions, then it makes perfect sense. But there are, of course, other reasons we should think that the background assumption is false (just like, imo, we should say the same thing to the person who thinks donating money can't be generous under capitalism).
The second statement is a little more complicated, but it also basically reduces to a question of larger theories operating in the background. Maybe a good comparison would be a debate about the heritability of IQ - some people think academic achievement is primarily explained by a person's genetic background, while others would argue that it's primarily explained by the social factors that often correlate with that person's genetic background. So while it might be that a person's genetic background passes the counterfactual test, there are still good reasons "lurking in the background" to think the explanatorily relevant property is actually the corresponding social conditions (or at least, many people argue there are). So even if one particular thing passes the counterfactual test, there might be some other related thing that passes it just as well and also brings other virtues along with it when incorporated into a particular theory. I think that's also the case when it comes to "dishonor" as a fundamental moral explanation versus "violating particular sexed social standards" as a non-moral (or even morally virtuous) explanation.
But thank you! I definitely look forward to your thoughts once I publish it. I appreciate your feedback and willingness to share things you ultimately disagree with!
This shows that positing things like generosity and kindness has explanatory power, but don’t we need something further that shows that positing that generosity and kindness are GOOD has even more explanatory power?
I think the goodness of a particular act can sometimes be explanatorily powerful, but I also think the primary way we determine what constitutes goodness or badness with regards to particular objects or entities is very different - we can deduce that directly through study of what the object or entity is, rather than having to posit goodness and badness in terms of their indispensibility to explanations. But that's another blog post!
Do you mean that we can posit that generosity exists and we can analyse generosity and intuitively realise it’s good, and that’s how we are justified in positing moral facts?
Or is it through a moral naturalist lens (I don’t know what moral naturalists say with this kind of thing)
Well, I don't think it's that simple - our intuitive sense that it's good for people to be generous is part of the data that needs to be explained by a theory, but it's not all we can rely on. Because I do think goodness is a natural property, I think it's possible to come to know what traits are good for human beings through all sorts of different lines of inquiry - psychology, sociology, etc. I don't think anyone disagrees that botany can tell us what's good for an oak tree, so there's no obvious reason to me why other fields can't tell us what's good for human beings.
Cool article. I’m of the mind that morality explains how we can be blameworthy in various cases-though I think this requires the existence of non-natural moral properties as I see that as the only way to get categorical moral reasons in the picture (though maybe I’m wrong about that).
I definitely think there are other accounts of categorical moral reasons that are naturalistic, but they don't go "as far" as the accounts non-naturalists can give, so I understand why some people find them unsatisfying.
This is in fact what we do with labels like berries, or bugs, or fruits vs. vegetables, or the sky. Whales were fish until we formalized different litmus criteria. Pluto was a planet until we formalized different litmus criteria. Their properties "of being in this category" changed even though they swam and orbited along as they always did. All of these things are dependent upon lexical stances, which are in turn driven by care & concern stances (avoiding confusion, being consistent, keeping with tradition, and so forth -- some of which may be at odds). So it isn't so absurd to define object properties in this way. We call a raisin delicious, but when we pause and think, we remember that this deliciousness is a truncated, reified way of what is actually happening, which is that when I eat a raisin, I have yumminess sensations; it is a description of a relationship and an actual and/or hypothetical interaction.
Her friend commends her for her generous donation of $216. What is the best explanation for donation having the property "generous"? Not just its size, no; instead, her friend's stances as well, combined with any number of relevant (per her friend's stances) circumstantial factors. The non-stance facts combine with the evaluative stances to yield the judgment. We can get different judgments not just by turning our world-knob across non-stance facts, but we can also get different judgments by turning our world-knob across the friend's evaluative stances.
This same pair of categorical knobs shows up for other moral properties like cruelty, for taste properties like spiciness, and for anything else you please -- squareness, being a fish, blueness, flatness, roughness, bravery, callousness, messiness, recklessness, lushness, being a boulder, being a car, being a liar, on Earth vs. in space, being in tune, being good at drawing, being beautiful, being wise, being tall, being short, being black, being a Scotsman, and on and on.
The reason this seems semantic is because it is. If the question is whether X counts as Y, the answer depends not only on X, but on Y as well. "Is Die Hard a Christmas movie?" was always a roundabout way of probing folks' lexical stances on the litmus criteria for the "Christmas movie" category.
There are many views of moral irrealism / antirealism whereby morality is real in numerous important ways (despite the name). But we do not accept that categorical evaluations (Is this cruel? Is this brave? Is this generous? Is this warranted?) make sense without appealing to stances in some way, usually care & concern stances.
I think I would reply to this by saying yes, it's definitely possible to take this *general* approach to basically all terms - but for someone who does take a generally realist approach to the sciences and to our larger catalogue of concepts more broadly, I don't think there's a principled reason to suspend those assumptions in the specific case of moral properties. If you think what constitutes a boulder is entirely determined by our stances towards different objects, then it makes sense that you'd think the same about what constitutes cruelty. But if you think there really is some right or wrong answer about what (at least a paradigmatic) boulder would be, then I think you should also accept the same conclusion about cruelty.
But I also just think it's *false* to say that the ascription of generosity to the woman's donation depends on the stances of the friend - various stances the friend has might encourage her to withhold the ascription of generosity (because she doesn't like the Red Cross, maybe, or she thinks the friend is being reckless) or offer it more freely, but it just doesn't seem to that whether the woman was actually generous has anything to do with what the friend thinks. Would you also say that, for example, whether something is *dangerous* depends on our stances? That just can't be right to me - even if I think a grizzly bear isn't going to hurt me, that doesn't seem to be the sort of mental event that would be capable of determining a property possessed by the grizzly bear. If I was mauled by the grizzly bear afterwards, would I have been mauled by a non-dangerous grizzly? I would see the property of generosity as roughly analogous in that sense - it just seems like the sort of thing I could be wrong about!
Surely we agree that there is no right/wrong answer about a paradigmatic boulder (just as there isn't one about litmus criteria for 'Christmas movie')? The best we can do is some fuzzy average of current convention. "A pebble is not a boulder" is too easy, just as "kicking a baby just for fun is cruel" is too easy, both of them tricking us into thinking that these things have hard litmus boundaries defined impersonally. By contrast, it's the "sorites" cases that remind us what's at play.
Ingrid gives $200 to help fund a new technology program at the community school. Julie sees this and evaluates it as "stingy," because she knows Ingrid is wealthy and could afford to give a lot more (as all other well-off parents have done). Kara, by contrast, sees anything greater than the typical $20-$50 donation as increasingly generous, irrespective of the giver's financial status, and $200 is four times that, so "generous" it is.
Whether something is "dangerous" also depends on stances. The chance of dying on a motorcycle is only 0.01% per 400,000 miles. Larry doesn't consider that "dangerous" at all. But for Mike, the fact that this is 25x as deadly as driving a car has Mike calling motorcycles "dangerous."
The biggest mistake we can make in philosophy is to hope or assume that the above two puzzles have wholly stance-independent answers that we can suss out by navigating absurda & intuition alone. (This miscalibrated hope is in fact the root driver of some of the most perennial, un-landable discourses in philosophy broadly, with examples in epistemology, modal logic, agency phil, and so forth). Rather, the correct answer is that Ingrid and Kara are both right per the litmus criteria of "generous" into which they've opted ("bootstrapped"; what Putnam calls "optional"); similarly, Larry and Mike are both right per the litmus criteria of "dangerous" into which they've opted.
If we are careful to avoid the "easy cases" and focus on the "sorites cases," we can come up with a similar story for every single virtue and vice we commend and condemn.
I agree we shouldn't focus only on easy cases, but I also think it's equally dangerous to act as though easy cases don't matter - a good theory should be able to account for the most straightforward situations as well as the most difficult ones, and any theory that can't confirm plausible statements like "Someone who think a pebble is a boulder is wrong" is at a serious disadvantage compared to theories that can confirm them, even if that confirmation opens them up to difficult positions somewhere else down the road. I think this is a major issue with a lot of antirealist thought, actually - there's a sense that, because we all tend to agree on the easy cases without having to justify them discursively, there's no need for our accounts to justify them metaphysically. But I think that's almost exactly backwards, and that we should be testing our theories first and foremost on the most obvious propositions rather than the most difficult ones. It's not impossible for a large collection of difficult edge cases to force the revision or rejection of a category that works well on the easy ones, but it's certainly a high bar that can't be underestimated.
Otherwise, I would just say that we don't have to accept *either* that conceptions of danger or generosity are entirely fixed apart from human concerns *or* that they're entirely expressive with no objective content underneath. In both of the cases you mentioned, I think it's possible to go beyond the initial disagreement to look at the theories of generosity or danger involved and compare their relative merits - if Kara thinks generosity is unrelated to someone's particular financial situation, but Julie disagrees, then they don't have to position that as just some unbridgeable ethical chasm. They can simply ask each other, "Oh, why do you think that?" and offer reasons for what they believe. Maybe Julie offers a thought experiment about a trillionaire who donates $100 to some organization, and Kara says, "Hmm, I guess that wouldn't be generous, yeah. But maybe it's more to do with..." and they go on.
With that in mind, I guess I would just ask again about the grizzly bear case: Do you really think that someone who judges the grizzly bear to not be dangerous and then gets immediately mauled was not wrong? Let's say Frank knew there was a 90% chance he'd get mauled, but he didn't consider a 90% chance of death dangerous. It just seems to me that any theory incapable of saying "Frank was wrong about whether the bear was dangerous" is a non-starter; that's a *paradigmatic* example of wrongness! Mike and Larry might have two rival conceptions of danger that are equally justified on theoretical grounds - although it does seem at least plausible to me that they could discuss their conceptions together until they ultimately agreed on which was best or adopted another that combined the best features of both - but Larry is just *wrong.* Do you reject any boundary whatsoever on the validity of conceptions like these?
Indeed, I would criticize anyone suggesting that these conceptions have no important stance-independent content.
To be clear, the easy cases are already fully accounted-for under *both* stance-independence ("'Kicking a puppy for fun is cruel' is true apart from any cares & concerns") and stance-dependence ("'Kicking a puppy for fun is cruel' per our cares & concerns about puppies"). The advantage of stance-dependence is we can jog toward borderfuzz cases and the theory stays wholly intact, whereas stance-independence starts to fritz out. Is it cruel to swat a fly? Is it cruel to use ant poison designed to be carried back to destroy its colony? Is it cruel to go for a recreational Sunday drive and kill dozens of flying insects? Stance-independence says that within each of these questions is a labyrinth of considerations with an impersonal exit. It is hoodwinking us with this promise. Stance-dependence says that there are a few impersonal considerations at play, but ultimately, the evaluation will be *per* bootstrapped cares & concerns about these insects. It, by contrast, is being honest with us.
What happens is that when we bootstrap a set of cares & concerns that we broadly share & deeply feel, it leaves only the stance-independent questions to debate. If in fact Julie & Kara have compatible stances but Kara's stated position is a poor expression of those stances, Julie can use the kind of approach you describe to nudge Kara into being more consistent with her values.
However, it is of course possible that they really do have a lexical impasse, so that after Julie gives the trillionaire example, Kara says, "Listen, I know what you're saying, but firmly feel that is the wrong way to frame generosity; she didn't have to give anything -- it's a donation! -- and she certainly didn't have to give as much as she did." As they talk further, they realize that they have deep differences of opinion about what parents "owe" schools ("ought" means "owed"), leading to the disparity in how they are labeling the act (i.e., their disagreement about "whether the act has the property 'generous,'" to use reified language).
Let say Frank had a particular lexicon in which "dangerous" is applied only to certainly deadly prospects, and he uses "perilous" for 90% cases. Is he now wrong to deny that the grizzly bear is dangerous? No, he is not, per his lexicon. This may seem like dodging the issue, but if Frank treats the grizzly with a degree of apprehension equal to someone who gives "dangerous" a lower litmus criterion, then he will behave the same way around the grizzly even though he doesn't call the grizzly "dangerous," just "perilous." To truly focus on the underlying issue, we set all labels aside, and ask Frank how he responds to a situation in which he has a 9 in 10 chance of being mauled. Then he might ask us to say what's on the line. We might say, a million bucks if he makes it past. We are now burrowing through impersonal facts on our way to a final evaluative judgment (whether to risk the bear) that, at some point, must appeal to Frank's personal stances in some way. This is always the case for any evaluative judgment.
And that's what stance-dependence is all about: All true moral propositions are true *per* certain cares & concerns. When cares & concerns are widely held & deeply felt, their bootstrapping is "automatic," and all that's left is to debate what stance-independent facts are at play. When a lexicon is widely held (like the criteria of "dangerous"), it is bootstrapped as well, and treated as a stanceless fact. The result is a culture where these opt-in factors are treated as stanceless, and moral debate feels intuitively like a stanceless exercise.
You seem to think that moral realism would be vindicated if it could be shown to have the same problems as other forms of realism, on the assumption that a problem for everybody is a problem for nobody. But I’m afraid that a problem for everybody is a problem for everybody.
The mistake you are making is using an analogy between physical properties that provide causal explanations in the physical world with properties that provide causal explanations in the human world.
That is not valid because human behavior does not comport with the principle of relativity, which says that the laws of nature hold independently of reference frame. That is, the laws of nature operate the same in different times and places.
This is clearly not the case for moral understandings. People living in different times and places have different views on what sorts of behavior is acceptable (i.e. morally right). A classic example is slavery. Today it is seen as heinous, and yet Jesus of Nazareth, believed by millions to be God (definer of morality) did not seem to have a problem with slavery, nor did the God of the Old Testament who laid out some 600+ laws, none of which proscribed slavery.
The author writes "The nearest possible world where the first-level fact about her donation is different would be one in which she donates $217, or $215, or maybe the same $216 to Doctors Without Borders instead. But regardless, her friend still would have commended her. What actually passes the counterfactual test would be the fact that she demonstrated the higher-level property of generosity, because that’s the fact where, were it not the case, she would not have been commended."
No it doesn't. It is not a property of the act that evoked the friend's commendation. It the fact that the friend *approves* of the action. Suppose the friend is an Ayn Rand fan and did not approve. She would not commend her friend.
But a world in which the friend who did actually commend her has an entirely different set of values or beliefs around generosity isn't the closest possible world. So this is a little like saying "Yes, the baseball did break the window, but if the window had been made of bulletproof glass instead, it wouldn't have broken, so therefore the baseball hitting the window can't explain it breaking."
We currently live in a world in which most people would approve of the friends action. But in this same world there are some who would not approve. I know folks who would express approval and others who would say nothing. Not everyone shares the same notions of right and wrong.
Have you ever seen those Am I the Asshole type discussion sites? My wife reads me bits from these. Sometimes the respondents are pretty unanimous in their condemnation of some behavior. Other times there is diversity of opinion. So no, people reaction to actions with moral valence can be quite variable.
This doesn’t actually seem to imply that moral properties exist but just that people talk like they do which is obviously true. If someone was not generous (a made up concept that people understand), then she would not have been commended
Higher-order descriptive properties are cheap. Sure, we can say there is an "objective" property of generosity, with explanatory power in some cultures and circumstances. But there is equally an "objective" property of *dishonor*, which equally "explains" why women are killed by their relatives in some cultures. We don't think the latter has any *normative* significance -- it isn't really a *moral* property -- so mere explanatory significance of this sort doesn't suffice to establish that what you're being a realist about is a genuinely normative property.
I had a whole section about this exact objection, actually, but it was getting unwieldy so I spun it out into a new post I'm working on now! Short answer, though: I think any maximally coherent theory about what moral properties do and don't exist would at least very plausibly conclude that either "dishonor" (to use your example) is not actually a moral property, or that the sorts of things that result in women being killed (sex outside of marriage, adopting western fashion, etc) don't actually instantiate it. In fact, honor killings seem like a really good example of a situation where the moral explanation fails the counterfactual test - if you believe that what a victim did was *not* dishonorable, but she was killed anyway, then you'd have to believe that, even if she wasn't dishonorable, they would have killed her anyway (because that's literally what did happen). Whereas, in the example of the donation, even someone who actively opposed generosity for some reason would still be likely to say that (in conjunction with a wide range of background conditions, like any explanation) the generosity really was what elicited the praise.
I'm not seeing the disanalogy:
* Someone who thinks donating money is *not* generous (e.g. an anti-capitalist who views the money as "ill-gotten gains") will say the act elicited praise even though it wasn't generous.
* killing due to dishonor passes a counterfactual test in that even if the woman hadn't committed *this particular* dishonor, if she'd committed any other comparable dishonor she would have similarly been killed.
But I look forward to your next post!
Oh, I would say there is no relevant disanalogy in a structural sense - both are totally legitimate assertions someone could make, but in both cases what matters is whether or not they have a theory of generosity or honor that's plausible enough to make us accept them.
The first statement is like an AIDS denialist who asserts that HIV is not deadly, so everyone who died while infected with HIV would have died even if they hadn't had it. There's nothing wrong with that argument *as an argument* - it's not logically fallacious, and if we accept the background assumptions, then it makes perfect sense. But there are, of course, other reasons we should think that the background assumption is false (just like, imo, we should say the same thing to the person who thinks donating money can't be generous under capitalism).
The second statement is a little more complicated, but it also basically reduces to a question of larger theories operating in the background. Maybe a good comparison would be a debate about the heritability of IQ - some people think academic achievement is primarily explained by a person's genetic background, while others would argue that it's primarily explained by the social factors that often correlate with that person's genetic background. So while it might be that a person's genetic background passes the counterfactual test, there are still good reasons "lurking in the background" to think the explanatorily relevant property is actually the corresponding social conditions (or at least, many people argue there are). So even if one particular thing passes the counterfactual test, there might be some other related thing that passes it just as well and also brings other virtues along with it when incorporated into a particular theory. I think that's also the case when it comes to "dishonor" as a fundamental moral explanation versus "violating particular sexed social standards" as a non-moral (or even morally virtuous) explanation.
But thank you! I definitely look forward to your thoughts once I publish it. I appreciate your feedback and willingness to share things you ultimately disagree with!
This shows that positing things like generosity and kindness has explanatory power, but don’t we need something further that shows that positing that generosity and kindness are GOOD has even more explanatory power?
I think the goodness of a particular act can sometimes be explanatorily powerful, but I also think the primary way we determine what constitutes goodness or badness with regards to particular objects or entities is very different - we can deduce that directly through study of what the object or entity is, rather than having to posit goodness and badness in terms of their indispensibility to explanations. But that's another blog post!
Do you mean that we can posit that generosity exists and we can analyse generosity and intuitively realise it’s good, and that’s how we are justified in positing moral facts?
Or is it through a moral naturalist lens (I don’t know what moral naturalists say with this kind of thing)
Well, I don't think it's that simple - our intuitive sense that it's good for people to be generous is part of the data that needs to be explained by a theory, but it's not all we can rely on. Because I do think goodness is a natural property, I think it's possible to come to know what traits are good for human beings through all sorts of different lines of inquiry - psychology, sociology, etc. I don't think anyone disagrees that botany can tell us what's good for an oak tree, so there's no obvious reason to me why other fields can't tell us what's good for human beings.
Cool article. I’m of the mind that morality explains how we can be blameworthy in various cases-though I think this requires the existence of non-natural moral properties as I see that as the only way to get categorical moral reasons in the picture (though maybe I’m wrong about that).
I definitely think there are other accounts of categorical moral reasons that are naturalistic, but they don't go "as far" as the accounts non-naturalists can give, so I understand why some people find them unsatisfying.
This is in fact what we do with labels like berries, or bugs, or fruits vs. vegetables, or the sky. Whales were fish until we formalized different litmus criteria. Pluto was a planet until we formalized different litmus criteria. Their properties "of being in this category" changed even though they swam and orbited along as they always did. All of these things are dependent upon lexical stances, which are in turn driven by care & concern stances (avoiding confusion, being consistent, keeping with tradition, and so forth -- some of which may be at odds). So it isn't so absurd to define object properties in this way. We call a raisin delicious, but when we pause and think, we remember that this deliciousness is a truncated, reified way of what is actually happening, which is that when I eat a raisin, I have yumminess sensations; it is a description of a relationship and an actual and/or hypothetical interaction.
Her friend commends her for her generous donation of $216. What is the best explanation for donation having the property "generous"? Not just its size, no; instead, her friend's stances as well, combined with any number of relevant (per her friend's stances) circumstantial factors. The non-stance facts combine with the evaluative stances to yield the judgment. We can get different judgments not just by turning our world-knob across non-stance facts, but we can also get different judgments by turning our world-knob across the friend's evaluative stances.
This same pair of categorical knobs shows up for other moral properties like cruelty, for taste properties like spiciness, and for anything else you please -- squareness, being a fish, blueness, flatness, roughness, bravery, callousness, messiness, recklessness, lushness, being a boulder, being a car, being a liar, on Earth vs. in space, being in tune, being good at drawing, being beautiful, being wise, being tall, being short, being black, being a Scotsman, and on and on.
The reason this seems semantic is because it is. If the question is whether X counts as Y, the answer depends not only on X, but on Y as well. "Is Die Hard a Christmas movie?" was always a roundabout way of probing folks' lexical stances on the litmus criteria for the "Christmas movie" category.
There are many views of moral irrealism / antirealism whereby morality is real in numerous important ways (despite the name). But we do not accept that categorical evaluations (Is this cruel? Is this brave? Is this generous? Is this warranted?) make sense without appealing to stances in some way, usually care & concern stances.
I think I would reply to this by saying yes, it's definitely possible to take this *general* approach to basically all terms - but for someone who does take a generally realist approach to the sciences and to our larger catalogue of concepts more broadly, I don't think there's a principled reason to suspend those assumptions in the specific case of moral properties. If you think what constitutes a boulder is entirely determined by our stances towards different objects, then it makes sense that you'd think the same about what constitutes cruelty. But if you think there really is some right or wrong answer about what (at least a paradigmatic) boulder would be, then I think you should also accept the same conclusion about cruelty.
But I also just think it's *false* to say that the ascription of generosity to the woman's donation depends on the stances of the friend - various stances the friend has might encourage her to withhold the ascription of generosity (because she doesn't like the Red Cross, maybe, or she thinks the friend is being reckless) or offer it more freely, but it just doesn't seem to that whether the woman was actually generous has anything to do with what the friend thinks. Would you also say that, for example, whether something is *dangerous* depends on our stances? That just can't be right to me - even if I think a grizzly bear isn't going to hurt me, that doesn't seem to be the sort of mental event that would be capable of determining a property possessed by the grizzly bear. If I was mauled by the grizzly bear afterwards, would I have been mauled by a non-dangerous grizzly? I would see the property of generosity as roughly analogous in that sense - it just seems like the sort of thing I could be wrong about!
Surely we agree that there is no right/wrong answer about a paradigmatic boulder (just as there isn't one about litmus criteria for 'Christmas movie')? The best we can do is some fuzzy average of current convention. "A pebble is not a boulder" is too easy, just as "kicking a baby just for fun is cruel" is too easy, both of them tricking us into thinking that these things have hard litmus boundaries defined impersonally. By contrast, it's the "sorites" cases that remind us what's at play.
Ingrid gives $200 to help fund a new technology program at the community school. Julie sees this and evaluates it as "stingy," because she knows Ingrid is wealthy and could afford to give a lot more (as all other well-off parents have done). Kara, by contrast, sees anything greater than the typical $20-$50 donation as increasingly generous, irrespective of the giver's financial status, and $200 is four times that, so "generous" it is.
Whether something is "dangerous" also depends on stances. The chance of dying on a motorcycle is only 0.01% per 400,000 miles. Larry doesn't consider that "dangerous" at all. But for Mike, the fact that this is 25x as deadly as driving a car has Mike calling motorcycles "dangerous."
The biggest mistake we can make in philosophy is to hope or assume that the above two puzzles have wholly stance-independent answers that we can suss out by navigating absurda & intuition alone. (This miscalibrated hope is in fact the root driver of some of the most perennial, un-landable discourses in philosophy broadly, with examples in epistemology, modal logic, agency phil, and so forth). Rather, the correct answer is that Ingrid and Kara are both right per the litmus criteria of "generous" into which they've opted ("bootstrapped"; what Putnam calls "optional"); similarly, Larry and Mike are both right per the litmus criteria of "dangerous" into which they've opted.
If we are careful to avoid the "easy cases" and focus on the "sorites cases," we can come up with a similar story for every single virtue and vice we commend and condemn.
I agree we shouldn't focus only on easy cases, but I also think it's equally dangerous to act as though easy cases don't matter - a good theory should be able to account for the most straightforward situations as well as the most difficult ones, and any theory that can't confirm plausible statements like "Someone who think a pebble is a boulder is wrong" is at a serious disadvantage compared to theories that can confirm them, even if that confirmation opens them up to difficult positions somewhere else down the road. I think this is a major issue with a lot of antirealist thought, actually - there's a sense that, because we all tend to agree on the easy cases without having to justify them discursively, there's no need for our accounts to justify them metaphysically. But I think that's almost exactly backwards, and that we should be testing our theories first and foremost on the most obvious propositions rather than the most difficult ones. It's not impossible for a large collection of difficult edge cases to force the revision or rejection of a category that works well on the easy ones, but it's certainly a high bar that can't be underestimated.
Otherwise, I would just say that we don't have to accept *either* that conceptions of danger or generosity are entirely fixed apart from human concerns *or* that they're entirely expressive with no objective content underneath. In both of the cases you mentioned, I think it's possible to go beyond the initial disagreement to look at the theories of generosity or danger involved and compare their relative merits - if Kara thinks generosity is unrelated to someone's particular financial situation, but Julie disagrees, then they don't have to position that as just some unbridgeable ethical chasm. They can simply ask each other, "Oh, why do you think that?" and offer reasons for what they believe. Maybe Julie offers a thought experiment about a trillionaire who donates $100 to some organization, and Kara says, "Hmm, I guess that wouldn't be generous, yeah. But maybe it's more to do with..." and they go on.
With that in mind, I guess I would just ask again about the grizzly bear case: Do you really think that someone who judges the grizzly bear to not be dangerous and then gets immediately mauled was not wrong? Let's say Frank knew there was a 90% chance he'd get mauled, but he didn't consider a 90% chance of death dangerous. It just seems to me that any theory incapable of saying "Frank was wrong about whether the bear was dangerous" is a non-starter; that's a *paradigmatic* example of wrongness! Mike and Larry might have two rival conceptions of danger that are equally justified on theoretical grounds - although it does seem at least plausible to me that they could discuss their conceptions together until they ultimately agreed on which was best or adopted another that combined the best features of both - but Larry is just *wrong.* Do you reject any boundary whatsoever on the validity of conceptions like these?
Indeed, I would criticize anyone suggesting that these conceptions have no important stance-independent content.
To be clear, the easy cases are already fully accounted-for under *both* stance-independence ("'Kicking a puppy for fun is cruel' is true apart from any cares & concerns") and stance-dependence ("'Kicking a puppy for fun is cruel' per our cares & concerns about puppies"). The advantage of stance-dependence is we can jog toward borderfuzz cases and the theory stays wholly intact, whereas stance-independence starts to fritz out. Is it cruel to swat a fly? Is it cruel to use ant poison designed to be carried back to destroy its colony? Is it cruel to go for a recreational Sunday drive and kill dozens of flying insects? Stance-independence says that within each of these questions is a labyrinth of considerations with an impersonal exit. It is hoodwinking us with this promise. Stance-dependence says that there are a few impersonal considerations at play, but ultimately, the evaluation will be *per* bootstrapped cares & concerns about these insects. It, by contrast, is being honest with us.
What happens is that when we bootstrap a set of cares & concerns that we broadly share & deeply feel, it leaves only the stance-independent questions to debate. If in fact Julie & Kara have compatible stances but Kara's stated position is a poor expression of those stances, Julie can use the kind of approach you describe to nudge Kara into being more consistent with her values.
However, it is of course possible that they really do have a lexical impasse, so that after Julie gives the trillionaire example, Kara says, "Listen, I know what you're saying, but firmly feel that is the wrong way to frame generosity; she didn't have to give anything -- it's a donation! -- and she certainly didn't have to give as much as she did." As they talk further, they realize that they have deep differences of opinion about what parents "owe" schools ("ought" means "owed"), leading to the disparity in how they are labeling the act (i.e., their disagreement about "whether the act has the property 'generous,'" to use reified language).
Let say Frank had a particular lexicon in which "dangerous" is applied only to certainly deadly prospects, and he uses "perilous" for 90% cases. Is he now wrong to deny that the grizzly bear is dangerous? No, he is not, per his lexicon. This may seem like dodging the issue, but if Frank treats the grizzly with a degree of apprehension equal to someone who gives "dangerous" a lower litmus criterion, then he will behave the same way around the grizzly even though he doesn't call the grizzly "dangerous," just "perilous." To truly focus on the underlying issue, we set all labels aside, and ask Frank how he responds to a situation in which he has a 9 in 10 chance of being mauled. Then he might ask us to say what's on the line. We might say, a million bucks if he makes it past. We are now burrowing through impersonal facts on our way to a final evaluative judgment (whether to risk the bear) that, at some point, must appeal to Frank's personal stances in some way. This is always the case for any evaluative judgment.
And that's what stance-dependence is all about: All true moral propositions are true *per* certain cares & concerns. When cares & concerns are widely held & deeply felt, their bootstrapping is "automatic," and all that's left is to debate what stance-independent facts are at play. When a lexicon is widely held (like the criteria of "dangerous"), it is bootstrapped as well, and treated as a stanceless fact. The result is a culture where these opt-in factors are treated as stanceless, and moral debate feels intuitively like a stanceless exercise.
You seem to think that moral realism would be vindicated if it could be shown to have the same problems as other forms of realism, on the assumption that a problem for everybody is a problem for nobody. But I’m afraid that a problem for everybody is a problem for everybody.
The mistake you are making is using an analogy between physical properties that provide causal explanations in the physical world with properties that provide causal explanations in the human world.
That is not valid because human behavior does not comport with the principle of relativity, which says that the laws of nature hold independently of reference frame. That is, the laws of nature operate the same in different times and places.
This is clearly not the case for moral understandings. People living in different times and places have different views on what sorts of behavior is acceptable (i.e. morally right). A classic example is slavery. Today it is seen as heinous, and yet Jesus of Nazareth, believed by millions to be God (definer of morality) did not seem to have a problem with slavery, nor did the God of the Old Testament who laid out some 600+ laws, none of which proscribed slavery.
The author writes "The nearest possible world where the first-level fact about her donation is different would be one in which she donates $217, or $215, or maybe the same $216 to Doctors Without Borders instead. But regardless, her friend still would have commended her. What actually passes the counterfactual test would be the fact that she demonstrated the higher-level property of generosity, because that’s the fact where, were it not the case, she would not have been commended."
No it doesn't. It is not a property of the act that evoked the friend's commendation. It the fact that the friend *approves* of the action. Suppose the friend is an Ayn Rand fan and did not approve. She would not commend her friend.
But a world in which the friend who did actually commend her has an entirely different set of values or beliefs around generosity isn't the closest possible world. So this is a little like saying "Yes, the baseball did break the window, but if the window had been made of bulletproof glass instead, it wouldn't have broken, so therefore the baseball hitting the window can't explain it breaking."
We currently live in a world in which most people would approve of the friends action. But in this same world there are some who would not approve. I know folks who would express approval and others who would say nothing. Not everyone shares the same notions of right and wrong.
Have you ever seen those Am I the Asshole type discussion sites? My wife reads me bits from these. Sometimes the respondents are pretty unanimous in their condemnation of some behavior. Other times there is diversity of opinion. So no, people reaction to actions with moral valence can be quite variable.
This doesn’t actually seem to imply that moral properties exist but just that people talk like they do which is obviously true. If someone was not generous (a made up concept that people understand), then she would not have been commended