While I'm sympathetic to your points about error theory, I think your remarks on subjectivism come from a failure to account for the implicit indexicals that come up when parsing moral sentences on speaker relativism, or at the very least, a difference in understanding of how these sorts of indexicals should be understood.
You write:
> But this is not all the subjectivist is dedicated to. They're dedicated to saying that, for the person who doesn't care, *there is no moral obligation to save children from drowning in any sense whatsoever*.
But this isn't how a subjectivist (or, at least, the subjectivist writing this post) would understand the situation. It would be like saying:
> A subjectivist is saying that, for a person named William, my name is not, in any sense whatsoever, Talis.
It sounds kind of weird, doesn't it? That's because, on a subjectivist viewpoint, a person speaking a moral phrase, e.g." Murder is wrong", is speaking in a sort of linguistic shorthand that hides a first-person pronoun. A sentence like "Murder is okay for William" is trying to index to two things at once-- it's simply ungrammatical when interpreted metanormatively. Yes, we can easily interpret it as a normative sentence, something approaching "Murder is okay when William does it," but that's just agent relativism, and no appraiser relativist would be obligated to affirm it. A better option would be something like "If I were William, and I spoke the sentence 'Murder is okay', I would be telling the truth." This gets closer, but even this still has a reasonable normative interpretation that infringes on the metaethical: it could still seem like I'm *normatively judging* the situation of being William, speaking that moral sentence, and commiting murder, rather than me *playing the role of* William.
As a final addendum, I'm pretty sure there's some more complex philosophical work to be done here on the parsing of complex moral sentences on subjectivism. Much like there's been work on logics that distinguish asserting P from asserting belief in P, it may be worth investigating syntactic structures that only allow statements of belief, with no underlying truth beneath it. It may very well be the case that there is no coherent system of subjectivist semantics, or that there is no way to recursively construct sentences in such a system without asserting some level of belief in the sentence's constituent components. But that is something to be shown by logicians, or perhaps linguists, and in the absence of some proof of total implausibility of subjectivist semantics, I think your inferences about subjectivism are on shaky ground.
Sorry about the total wall of text. I hope this leads to interesting discussion, etc.!
Thanks for the long comment (happy to return the favour!). I feel my article addresses these very concerns with the "too narrow" criticism.
I understand why folks might want to interpret the consequent (and 2nd premise) in subjectivist language–after all, the antecedent claims that subjectivism is true, and part of that presumably implies a certain view about moral language. However, I do think this is too narrow an interpretation of the consequent because then it doesn't clarify that there would be no other way to derive the moral requirement in question. After all, it couldn't be that there is also be an objective moral fact that grounds such a moral obligation if subjectivism were right.
To perhaps make the point again, let's consider the argument against error theory. If the first premise is interpreted as...
1. If error theory is true, then there is nothing stance-independently wrong (or right) in a non-naturalist sense with genocide.
... then it makes it seem that there could still be wrongness in stance independent natural senses, and stance dependent senses. But of course, error theory denies these conceptions of morality too! If error theory is right, nothing is right or wrong in any sense, not just the stance independent non-natural sense.
Likewise for subjectivism, if subjectivism is right, then there is no sense in which the non-child rescuer has a moral obligation to rescue children if he lacks the prerequisite preferences, e.g. it's not as if there could also be a stance independent moral fact that could ground such a moral obligation.
I hope that helps.
As a second note, it's actually not clear to me whether the consequent (of the Rescue argument) is a moral claim. It's merely identifying a situation in which someone doesn't have the prerequisite desires to ground a moral obligation to save children (provided we assume that subjectivism is right). However, I don't think this is a moral claim, e.g. I'm not suggesting that subjectivists cannot say the guy is acting wrongfully (because they totally can say that). If the premise looked something more like:
1. If subjectivism is true, then it's not wrong for John to refrain from saving children if he doesn't care to.
2. But, it is wrong for John to refrain from saving children, even if he doesn't care to.
If the first premise looked like this, then I'd agree it was a moral claim. Though, it seems that various subjectivist views could consistently reject premise 1, (e.g. appraiser subjectivism), so I would reject this premise.
I mainly divulge what I take an accusation of normative entanglement to be in the intro.
I take normative entanglement (here) to refer to an argument (or premise) that relies on misinterpreting it to make a false point-especially regarding meta ethical views implying normative ethic conclusions.
For instance, if I was to argue that since error theorists think that there is nothing wrong with genocide, that they must think genocide is morally good, that would be entanglement. Alternatively, if a premise was worded to imply this conclusion (and falsely so) then it would also seem to be a sort of normative entanglement-though this is trickier given the intentions of the arguer. E.g. Someone might just be unfairly interpreting a decent argument.
That's kind of in the ballpark but still a bit imprecise. Have you read any articles specifically about normative entanglement, i.e., what it is, when it occurs, etc.?
I’ve mainly focused on what Travis seemed to have implied in their article. He didn’t exactly specify what it meant either, so I also tried to understand it based on your own article (though not fully). I’ve seen you include other things in normative entanglement (like issues with intelligibility of realist claims) but if that counts as normative entanglement then it’s not clear that it’s a problem that needs to be faced. This latter point was also not brought up by Travis too.
I ask because my I'm not sure from reading this post whether normative entanglement was understood or adequately addressed. I don't want to overwhelm you with my responses on the matter though so if you're interested in discussing realism/antirealism I'd rather hang back (at least here on Substack) and see how the interaction with Travis unfolds. You're welcome to join us on Discord though if you'd like to talk to us directly there in real time (in text or VC).
“If moral error theory is true, then there is nothing wrong (or right) with genocide.”
If moral error theory is true, nothing can entail a moral claim (especially error theory), because error theory concludes that all moral claims are false. Isn't “there is nothing wrong (or right) with genocide” a moral claim? Treating that proposition as potentially true contradicts error theory. Right and wrong don’t work as logical predicates if error theory is correct.
I was always a bit puzzled about how they deal with negation. (I haven’t actually studied error theory, just looked it up once.) I guess somehow the logical negation/denial of a moral claim need not be a moral claim? How would that work?
The non-cognitivists make more sense to me, saying that moral claims are not even propositions that can be true or false.
So another approach might be:
1) If Moral error theory is true, then all moral claims are false.
2) “there is nothing wrong (or right) with genocide.” Is a moral claim.
3) So if moral error theory is true, then “there is nothing wrong (or right) with genocide” is false.
4) If moral error theory is true, there is nothing wrong or right with genocide.
5) if moral error theory is true, then “there is nothing wrong (or right) with genocide” is both true and false.
The way Bart Streumer does it is he says that any statement that entails the existence of a moral *property* is false (he's actually talking about all normative properties, but that's not super important in this context). Some examples of claims that would entail the existence of normative properties:
- Burning cats is wrong
- There is decisive reason against torturing babies
- It's permissible not to vote.
- You ought to vote.
- There is some reason to donate to charity, but also some reason against.
- Pleasure is good
- Pain is good if someone wants it, but bad otherwise
Some examples of claims that would not entail the existence of a normative property:
- Nothing is either forbidden or permissible
- Nothing is either good or bad
- Nothing is either right or wrong
- It's not wrong to pet your dog in the evenings
- It's not right to double park
These last two might seem like they entail the existence of a normative property. But this is only because they *pragmatically implicate* a *different* moral claim which does (namely "it's permissible to pet your dog in the evenings" and "it's wrong to double park", respectively). But strictly speaking they don't imply the existence of a normative property.
So it's not that any sentence that contains a normative term is false. It's that any sentence that entails the existence of a normative property is false.
It still seems confusing, as people tend to think “right” and “wrong” are mutually exhaustive, so negating one entails the other. E.g. “it is not the case that burning cats is wrong” is not the same as “burning cats is right,” but people might interpret it that way. And I am not sure this is such a bad way of thinking. If I have no obligation preventing me from doing something, then I can do it or not depending on my own pragmatic evaluation of the situation, and that is pretty close to a reasonable interpretation of “right.” OTOH, this ignores the basic problem, but just pushes it back to “what obligations do we have and why.”
I’m not sure how this account differs from the noncognitivists', which would just say that containing “right” or “ wrong” as predicates makes a proposition neither true nor false. I don’t agree, but maybe I misunderstand them also.
And I guess by this account other forms of normativity all fall into the category of pragmatism? E.g., “one ought not chop off one's toe on a whim,” or “one ought not use fallacies when seeking to persuade,” or even “one may not move a pawn backwards when playing chess.” Those look like pragmatic or instrumental norms. Maybe the chess one is a constitutive norm, as there is no moral or practical reason not to move illegally, but then you're just not playing chess any more.
In any case, it's just as legitimate as my initial argument (which is to say it's fine). Though, I would have to update the conclusion to include "about what moral content there is".
So, error theorists think that nothing is right or wrong, since morality doesn’t really exist. It trivially follows from this that there is nothing wrong or right with genocide. The consequent is not prescribing wrongness or rightness to genocide. Precisely the opposite is being done.
While I'm sympathetic to your points about error theory, I think your remarks on subjectivism come from a failure to account for the implicit indexicals that come up when parsing moral sentences on speaker relativism, or at the very least, a difference in understanding of how these sorts of indexicals should be understood.
You write:
> But this is not all the subjectivist is dedicated to. They're dedicated to saying that, for the person who doesn't care, *there is no moral obligation to save children from drowning in any sense whatsoever*.
But this isn't how a subjectivist (or, at least, the subjectivist writing this post) would understand the situation. It would be like saying:
> A subjectivist is saying that, for a person named William, my name is not, in any sense whatsoever, Talis.
It sounds kind of weird, doesn't it? That's because, on a subjectivist viewpoint, a person speaking a moral phrase, e.g." Murder is wrong", is speaking in a sort of linguistic shorthand that hides a first-person pronoun. A sentence like "Murder is okay for William" is trying to index to two things at once-- it's simply ungrammatical when interpreted metanormatively. Yes, we can easily interpret it as a normative sentence, something approaching "Murder is okay when William does it," but that's just agent relativism, and no appraiser relativist would be obligated to affirm it. A better option would be something like "If I were William, and I spoke the sentence 'Murder is okay', I would be telling the truth." This gets closer, but even this still has a reasonable normative interpretation that infringes on the metaethical: it could still seem like I'm *normatively judging* the situation of being William, speaking that moral sentence, and commiting murder, rather than me *playing the role of* William.
As a final addendum, I'm pretty sure there's some more complex philosophical work to be done here on the parsing of complex moral sentences on subjectivism. Much like there's been work on logics that distinguish asserting P from asserting belief in P, it may be worth investigating syntactic structures that only allow statements of belief, with no underlying truth beneath it. It may very well be the case that there is no coherent system of subjectivist semantics, or that there is no way to recursively construct sentences in such a system without asserting some level of belief in the sentence's constituent components. But that is something to be shown by logicians, or perhaps linguists, and in the absence of some proof of total implausibility of subjectivist semantics, I think your inferences about subjectivism are on shaky ground.
Sorry about the total wall of text. I hope this leads to interesting discussion, etc.!
Thanks for the long comment (happy to return the favour!). I feel my article addresses these very concerns with the "too narrow" criticism.
I understand why folks might want to interpret the consequent (and 2nd premise) in subjectivist language–after all, the antecedent claims that subjectivism is true, and part of that presumably implies a certain view about moral language. However, I do think this is too narrow an interpretation of the consequent because then it doesn't clarify that there would be no other way to derive the moral requirement in question. After all, it couldn't be that there is also be an objective moral fact that grounds such a moral obligation if subjectivism were right.
To perhaps make the point again, let's consider the argument against error theory. If the first premise is interpreted as...
1. If error theory is true, then there is nothing stance-independently wrong (or right) in a non-naturalist sense with genocide.
... then it makes it seem that there could still be wrongness in stance independent natural senses, and stance dependent senses. But of course, error theory denies these conceptions of morality too! If error theory is right, nothing is right or wrong in any sense, not just the stance independent non-natural sense.
Likewise for subjectivism, if subjectivism is right, then there is no sense in which the non-child rescuer has a moral obligation to rescue children if he lacks the prerequisite preferences, e.g. it's not as if there could also be a stance independent moral fact that could ground such a moral obligation.
I hope that helps.
As a second note, it's actually not clear to me whether the consequent (of the Rescue argument) is a moral claim. It's merely identifying a situation in which someone doesn't have the prerequisite desires to ground a moral obligation to save children (provided we assume that subjectivism is right). However, I don't think this is a moral claim, e.g. I'm not suggesting that subjectivists cannot say the guy is acting wrongfully (because they totally can say that). If the premise looked something more like:
1. If subjectivism is true, then it's not wrong for John to refrain from saving children if he doesn't care to.
2. But, it is wrong for John to refrain from saving children, even if he doesn't care to.
If the first premise looked like this, then I'd agree it was a moral claim. Though, it seems that various subjectivist views could consistently reject premise 1, (e.g. appraiser subjectivism), so I would reject this premise.
Could you perhaps clarify what you take normative entanglement to be?
I mainly divulge what I take an accusation of normative entanglement to be in the intro.
I take normative entanglement (here) to refer to an argument (or premise) that relies on misinterpreting it to make a false point-especially regarding meta ethical views implying normative ethic conclusions.
For instance, if I was to argue that since error theorists think that there is nothing wrong with genocide, that they must think genocide is morally good, that would be entanglement. Alternatively, if a premise was worded to imply this conclusion (and falsely so) then it would also seem to be a sort of normative entanglement-though this is trickier given the intentions of the arguer. E.g. Someone might just be unfairly interpreting a decent argument.
That's kind of in the ballpark but still a bit imprecise. Have you read any articles specifically about normative entanglement, i.e., what it is, when it occurs, etc.?
I’ve mainly focused on what Travis seemed to have implied in their article. He didn’t exactly specify what it meant either, so I also tried to understand it based on your own article (though not fully). I’ve seen you include other things in normative entanglement (like issues with intelligibility of realist claims) but if that counts as normative entanglement then it’s not clear that it’s a problem that needs to be faced. This latter point was also not brought up by Travis too.
I ask because my I'm not sure from reading this post whether normative entanglement was understood or adequately addressed. I don't want to overwhelm you with my responses on the matter though so if you're interested in discussing realism/antirealism I'd rather hang back (at least here on Substack) and see how the interaction with Travis unfolds. You're welcome to join us on Discord though if you'd like to talk to us directly there in real time (in text or VC).
Btw, discord username is huskyhelianthus
If that’s so, then it would also seem that Travis was mistaken in referring to these criticisms as normative entanglement.
I’m down to chat about my and Travis’ articles.
What seems very important here is whether normative entanglement is indeed a criticism (or whether that is an open question).
I agree with Travis that normative entanglement is going on in your original arguments.
“If moral error theory is true, then there is nothing wrong (or right) with genocide.”
If moral error theory is true, nothing can entail a moral claim (especially error theory), because error theory concludes that all moral claims are false. Isn't “there is nothing wrong (or right) with genocide” a moral claim? Treating that proposition as potentially true contradicts error theory. Right and wrong don’t work as logical predicates if error theory is correct.
I was always a bit puzzled about how they deal with negation. (I haven’t actually studied error theory, just looked it up once.) I guess somehow the logical negation/denial of a moral claim need not be a moral claim? How would that work?
The non-cognitivists make more sense to me, saying that moral claims are not even propositions that can be true or false.
So another approach might be:
1) If Moral error theory is true, then all moral claims are false.
2) “there is nothing wrong (or right) with genocide.” Is a moral claim.
3) So if moral error theory is true, then “there is nothing wrong (or right) with genocide” is false.
4) If moral error theory is true, there is nothing wrong or right with genocide.
5) if moral error theory is true, then “there is nothing wrong (or right) with genocide” is both true and false.
6)So moral error theory is false.
A moral error theorist would need to deny 2 or 4.
The way Bart Streumer does it is he says that any statement that entails the existence of a moral *property* is false (he's actually talking about all normative properties, but that's not super important in this context). Some examples of claims that would entail the existence of normative properties:
- Burning cats is wrong
- There is decisive reason against torturing babies
- It's permissible not to vote.
- You ought to vote.
- There is some reason to donate to charity, but also some reason against.
- Pleasure is good
- Pain is good if someone wants it, but bad otherwise
Some examples of claims that would not entail the existence of a normative property:
- Nothing is either forbidden or permissible
- Nothing is either good or bad
- Nothing is either right or wrong
- It's not wrong to pet your dog in the evenings
- It's not right to double park
These last two might seem like they entail the existence of a normative property. But this is only because they *pragmatically implicate* a *different* moral claim which does (namely "it's permissible to pet your dog in the evenings" and "it's wrong to double park", respectively). But strictly speaking they don't imply the existence of a normative property.
So it's not that any sentence that contains a normative term is false. It's that any sentence that entails the existence of a normative property is false.
That seems like a much better account.
It still seems confusing, as people tend to think “right” and “wrong” are mutually exhaustive, so negating one entails the other. E.g. “it is not the case that burning cats is wrong” is not the same as “burning cats is right,” but people might interpret it that way. And I am not sure this is such a bad way of thinking. If I have no obligation preventing me from doing something, then I can do it or not depending on my own pragmatic evaluation of the situation, and that is pretty close to a reasonable interpretation of “right.” OTOH, this ignores the basic problem, but just pushes it back to “what obligations do we have and why.”
I’m not sure how this account differs from the noncognitivists', which would just say that containing “right” or “ wrong” as predicates makes a proposition neither true nor false. I don’t agree, but maybe I misunderstand them also.
And I guess by this account other forms of normativity all fall into the category of pragmatism? E.g., “one ought not chop off one's toe on a whim,” or “one ought not use fallacies when seeking to persuade,” or even “one may not move a pawn backwards when playing chess.” Those look like pragmatic or instrumental norms. Maybe the chess one is a constitutive norm, as there is no moral or practical reason not to move illegally, but then you're just not playing chess any more.
Though, if you want, you could also use an argument that only supposed that part of error theory is true. For example:
1. If error theory is right about what moral content there is (there’s nothing), then there is nothing right or wrong with genocide.
2. But there is something wrong with genocide.
3. So, error theory is false.
Somehow I think this misses the point.
In any case, it's just as legitimate as my initial argument (which is to say it's fine). Though, I would have to update the conclusion to include "about what moral content there is".
So, error theorists think that nothing is right or wrong, since morality doesn’t really exist. It trivially follows from this that there is nothing wrong or right with genocide. The consequent is not prescribing wrongness or rightness to genocide. Precisely the opposite is being done.