> Most anti-realists agree that if there are irrational desires, moral realism is true. This is true because morality is about what we have impartial reason to do, and impartiality is clearly coherent, so as long as there are desire independent reasons to do certain things and care about other things, moral realism is true.
I haven't read the Parfit book you cite, but this seems incomplete: some desires are irrational because they make it more difficult to do things that you desire more. For instance, if you want just about anything in life, also desiring to kill yourself would be irrational, both because having that desire makes it more likely that you'll act on it, & because (for most humans) having that desire is unpleasant, which is also undesirable. Or, to take a less extreme example, if you want to remain healthy, it is irrational to also want to eat lots of unhealthy food. However, if we exclude this exception, this argument seems correct.
> Here are a bunch of desires that aren’t irrational if anti-realism is true,
Most of these hypotheticals are desires that a human is very unlikely to have (e.g., wanting to eat a car, being indifferent to any amount of pain felt on Tuesdays), but that doesn't mean that it's impossible for any sort of mind to have these desires & be rational in pursuing them. (Likewise, destroying the Earth to turn it into paperclips is obviously irrational to a human, but that doesn't mean that a badly designed AI couldn't have desires that would make this action rational for it.) 4 & 7 are just illustrations of the fact that, for someone who discounts the future highly, acting rationally in the present often makes it harder to get what one will want in the future.
The criticism of noncognitivism you describe seems correct. However, if you think "that moral values are too odd to be part of the fabric of reality" (as I do, since as far as I know there is no evidence that morality is "part of the fabric of reality"), then the only possible conclusion is that there is no morality independent of people's conceptions of morality. Since people do not agree on morality, different people's or societies' ideas of morality must vary; therefore, to say that something is moral or immoral makes sense only within a particular moral worldview. Given this context, most of your supposed counterexamples to subjectivism seem reasonable; e.g., Nazi Germany acted morally by its standards but immorally by our standards. The only exceptions are "Some societal practices are immoral" (most societies have some practices that are immoral by their own standards; e.g., scams & racial stereotyping in modern American society) & "when I say “we should be utilitarians,” and Kant says “we shouldn’t be utilitarians,” we’re not actually disagreeing" (since these quotes clearly refer to objective & not subjective morality).
Is there a way of changing the commenting methods on your posts? It looks like you want to discontinue our exchange but in large part that could be because there's no good way of quoting people (at least that I'm aware of) to make for a clean and orderly discussion. But I'd like to have a written exchange, since that provides a different medium for conveying our ideas than just speaking.
Personally, for instance, I'm still very curious what you take the implications of moral relativism to be. I don't think relativism has *any* troubling implications, so I don't think people could be appropriately dissuaded from relativism by its alleged "implications."
If you really don't want to have this exchange here...which is a bit weird (since what else is a comment section for?) that's your prerogative, but if so, as you suggest, let's do another video chat. Either way I'd like these exchanges to made available to others!
//Most anti-realists agree that if there are irrational desires, moral realism is true.//
Maybe. This seems like an empirical question. Do you mean most philosophers who identify as moral antirealists, most lay antirealists, or something else?
In any case, I do not agree that if there are irrational desires that moral realism is true.
//This is true because morality is about what we have impartial reason to do//
I disagree. I do not think it’s true that morality is about what we have impartial reason to do. It is well within the purview of a moral theory that it permit or even obligate us to act in ways that are partial, e.g., one may be licensed to exercise a personal prerogative (see e.g., G. A. Cohen, who defends a finite personal prerogative to act in one’s own self-interest, which morally permits partial behavior), or ethical egoism, or moral standards whereby one has an obligation to act partially in one’s self-interest or in the interests of particular individuals or groups over other particular individuals or groups.
//Here are a bunch of desires that aren’t irrational if anti-realism is true//
Before denying whether any of these desires are “irrational,” I’d first have to know what you mean by “irrational.”
Before knowing what you mean, I am not in a position to affirm or deny this. Instead, I’d simply say I have no determinate position pending adequate clarification of what you mean.
//Very intuitive! //
What does this mean? Every one of these scenarios is underdescribed. It’s not that I find them “intuitive,” but that they are neither “intuitive” nor “unintuitive.” They lack the kind of information I’d use to judge whether the desire in question is “irrational” or not.
For instance, suppose I asked you this:
“Is it irrational to jump out of a plane?”
There’s not enough information. A person could be jumping out of a plane without a parachute because they mistakenly believe they can fly. Or a person could be jumping out of a plane with a parachute because they are skydiving. In the former case, the action may be “irrational,” while in the latter it may not be.
Whether I would consider a desire “rational” or not is going to turn on whether the desire is instrumental or non-instrumental. If it’s non-instrumental, then I think it’s a category mistake to label any of these desires “irrational,” for the same reason it makes no sense to tell someone whose favorite color is blue that they’re being irrational: one’s non-instrumental desires are just psychological facts about the agent that are no more subject to evaluation as “rational” or not as their age or hair color. If it’s instrumental, then whether it’s “rational” or not would depend on a variety of considerations, such as whether its conducive to their non-instrumental desires. If e.g., eating a car was conducive to someone’s non-instrumental desires, then why would it be irrational to eat a car? What’s irrational about simply wanting to eat a car? What would it even mean for it to be “irrational”?
//And then we come to moral problems. Moral anti-realism will either be subjectivism (according to which moral facts are true in virtue of what some people think of them, either the speaker or society), error theory (according to which moral claims are semantically intended to be realist, but all moral claims are false), and non-cognitivism, according to which moral claims don’t have a truth value.//
These are not the only options. That’s a claim Huemer made in one of his books, but it’s not true.
Don Loeb endorses moral incoherentism; I endorse a non-semantic generalized antirealism which denies the existence of stance-independent moral facts but isn’t committed to any particular account of the meaning of ordinary terms; this position is coupled with moral quietism and metaethical indeterminacy/variability, which are explicitly expressed in such a way that they involve a rejection of conventional uniform and determinate accounts of the meaning of ordinary moral claims. See here:
Gill, M. B. (2009). Indeterminacy and variability in meta-ethics. Philosophical studies, 145(2), 215-234.
Gill, M. B. (2008). Metaethical variability, incoherence, and error. In W. Sinnott-Armstrong (Ed.), Moral psychology: The cognitive science of morality (Vol. 2) (pp. 387-402). Cambridge, MA: The MIT Press.
Loeb, D. (2008). Moral incoherentism: How to pull a metaphysical rabbit out of a semantic hat. In W. Sinnott-Armstrong (Ed.), Moral psychology: The cognitive science of morality (Vol. 2) (pp. 355-386). Cambridge, MA: The MIT Press.
These may not be common positions, but a position does not have to be common to represent a legitimate alternative. Incoherentism, the IV thesis, and moral quietism (my view) are all legitimate alternatives to the conventional three metaethical positions.
//(a) Evaluative statements take the form of declarative sentences, rather than, say, imperatives, questions, or interjections. //
Imperatives in English are often expressed in declarative form. For instance, if a teenager is caught trying to sneak out of the house, their mother or father may say:
“You will not go to that party.”
This is phrased as a declarative sentence, yet it is best understood (and the parent might confirm this) to be an imperative, not a declarative sentence. Declarative sentences can borrow the rhetorical force of a declarative sentence as a form of emphasis. Sentences that appear declarative can and are used for non-declarative purposes.
Whether or not, and how often, people use moral claims without intending to express a propositional claim is an empirical question, and not one that can be settled by a priori reasoning. Philosophers, Huemer included, who do not engage with or conduct the proper empirical research are simply not employing the appropriate methods for determining what people mean when they say things like “Murder is wrong.” This is a question to be settled by psychologists, linguists, anthropologists, etc., or philosophers competent in the use of the appropriate empirical methods.
"/Most anti-realists agree that if there are irrational desires, moral realism is true.//
Maybe. This seems like an empirical question. Do you mean most philosophers who identify as moral antirealists, most lay antirealists, or something else?"
I have yet to see any philosopher deny this parity. If there are desire independent reasons, that seems to be sufficient for moral reasons.
"//This is true because morality is about what we have impartial reason to do//
I disagree. I do not think it’s true that morality is about what we have impartial reason to do. It is well within the purview of a moral theory that it permit or even obligate us to act in ways that are partial, e.g., one may be licensed to exercise a personal prerogative (see e.g., G. A. Cohen, who defends a finite personal prerogative to act in one’s own self-interest, which morally permits partial behavior), or ethical egoism, or moral standards whereby one has an obligation to act partially in one’s self-interest or in the interests of particular individuals or groups over other particular individuals or groups."
Well, in that case, for that to be a moral theory, I think it would have to defend that if you were rational and impartial you'd sometimes act in self interested ways. If not, then I wouldn't classify that as a moral theory. I think, for example, that Ayn Rand would be espousing a moral theory because she tried to deduce from first principles why reason alone would be sufficient to turn someone into an egoist (she was also ofc ludicrously wrong).
"Before denying whether any of these desires are “irrational,” I’d first have to know what you mean by “irrational.”
Before knowing what you mean, I am not in a position to affirm or deny this. Instead, I’d simply say I have no determinate position pending adequate clarification of what you mean."
Well, irrational would be roughly synonymous to foolish, stupid, unwise, idiotic, and out of accord with reason. I'd be using a definition that roughly tracks ordinary language. When someone says "you're being irrational," I mean roughly what I think they mean.
"“Is it irrational to jump out of a plane?”
There’s not enough information. A person could be jumping out of a plane without a parachute because they mistakenly believe they can fly. Or a person could be jumping out of a plane with a parachute because they are skydiving. In the former case, the action may be “irrational,” while in the latter it may not be.
Whether I would consider a desire “rational” or not is going to turn on whether the desire is instrumental or non-instrumental. If it’s non-instrumental, then I think it’s a category mistake to label any of these desires “irrational,” for the same reason it makes no sense to tell someone whose favorite color is blue that they’re being irrational: one’s non-instrumental desires are just psychological facts about the agent that are no more subject to evaluation as “rational” or not as their age or hair color. If it’s instrumental, then whether it’s “rational” or not would depend on a variety of considerations, such as whether its conducive to their non-instrumental desires. If e.g., eating a car was conducive to someone’s non-instrumental desires, then why would it be irrational to eat a car? What’s irrational about simply wanting to eat a car? What would it even mean for it to be “irrational”?"
Well, there's nothing irrational about having the thought that it would be fun to eat a car. What would be irrational would be aiming to eat a car if one derived no happiness from it. I think it's useful to distinguish desires and deeper aims. I'm referring to deeper aims here. So I'd like to be able to say that deontologists are making some sort of error, as are those who want to set themselves on fire, despite not finding doing so to be at all pleasant.
"Don Loeb endorses moral incoherentism; I endorse a non-semantic generalized antirealism which denies the existence of stance-independent moral facts but isn’t committed to any particular account of the meaning of ordinary terms; this position is coupled with moral quietism and metaethical indeterminacy/variability"
I'll look into the Loeb view. However, if your view is that moral statements express one of those three views, depending on context and the speaker, then you'd have to hold that the reductios to some of those views would apply. So when I utter the statement "it's immoral to torture infants," what I'm saying is false, given that I think I'm appealing to an external standard.
"These may not be common positions, but a position does not have to be common to represent a legitimate alternative. Incoherentism, the IV thesis, and moral quietism (my view) are all legitimate alternatives to the conventional three metaethical positions."
Moral quietism doesn't seem to be much of a view, it just seems to be the notion that we don't need an account of what people mean by it.
"//(a) Evaluative statements take the form of declarative sentences, rather than, say, imperatives, questions, or interjections. //
Imperatives in English are often expressed in declarative form. For instance, if a teenager is caught trying to sneak out of the house, their mother or father may say:
“You will not go to that party.”
This is phrased as a declarative sentence, yet it is best understood (and the parent might confirm this) to be an imperative, not a declarative sentence. Declarative sentences can borrow the rhetorical force of a declarative sentence as a form of emphasis. Sentences that appear declarative can and are used for non-declarative purposes."
That's true, but these are rare exceptions. Most of the time such statements are declarative. It can't be settled by a priori reasoning, but a priori reasoning can reveal its implausible implications.
Your response style makes it hard to tell where my quotes end and where your responses begin. Is there a way to change the way you’re formatting responses to make it easier to tell where your responses begin? Maybe enable some kind of quoting thing that brackets off quotes?
//I have yet to see any philosopher deny this parity. //
I’m not clear on the context here. What parity?
As far as I know, Joyce accepts categorical epistemic reasons but not categorical moral reasons. So it seems there’s at least one person - arguably the most prominent living error theorist - who may hold that there are desire independent reasons, but that none of them are moral. I’m not sure, and it’s worth double-checking.
However, even if Joyce doesn’t endorse this view now (or didn’t before), I outright deny that if there are desire independent reasons that this is sufficient for “moral” reasons. You’d still have to offer an account of what makes some particular subset of desire independent reasons distinctively moral. There are other normative domains; not all desire independent reasons are necessarily moral ones: it is logically possible to be a normative realist without being a moral realist, and it does not follow that if normative realism is true that moral realism is true.
//Well, in that case, for that to be a moral theory, I think it would have to defend that if you were rational and impartial you'd sometimes act in self interested ways.//
That’s irrelevant. My point is that it isn’t part of the definition of moral theories that they concern what we impartially ought to do. It isn’t simply part of the concept of morality that it be impartial; people can and do incorporate and advocate for partiality being morally permissible or even required. Whether they can convincingly defend such views is irrelevant. This is just the same as if someone were to presume that a “moral theory” is necessarily a theory about consequences, or rules; a deontologist or consequentialist could rightly object that any attempt to define “moral theory” in such a way question-beggingly presumes that their own normative standards are constitutive of what it even means to be a moral theory. I’m claiming you’re doing the same, only with impartiality. I do not agree that moral theories just are theories about what we impartially have reason to do. For that matter, one could be a reasons eliminativist and still endorse normative moral theories, so not only do you not need the impartiality component, you don’t need the “reasons” component, either.
You’re baking your own metaethical and normative stances into your notion of what is required for something to be a moral theory.
//Well, irrational would be roughly synonymous to foolish, stupid, unwise, idiotic, and out of accord with reason.//
That’s insufficient. Each one of these terms could be unpacked in ways consistent with instrumentalist and non-instrumentalist accounts of rationality. I certainly think people can be foolish, stupid, unwise, etc., but my use of all these terms is *also* likely different from your conceptions of these terms, if they are intended to be similar to or synonyms with “rationality” in such a way that they bake in realist or categorical presuppositions.
//I'd be using a definition that roughly tracks ordinary language.//
Do you have empirical evidence on how these terms are used in ordinary language? If not, on what grounds do you claim your use of these terms resembles ordinary language?
//When someone says "you're being irrational," I mean roughly what I think they mean. //
What do you think they mean?
//What would be irrational would be aiming to eat a car if one derived no happiness from it.//
Why? I don’t think it’s constitutive of rationality that it’s not rational to do something you desire to do if you wouldn’t be made happy by it. Although I care about happiness, I don’t think it’s a requirement of a rational agent that they care about happiness and that their actions are directed towards becoming more happy.
I suspect you’re just projecting your own preference for happiness onto others, and declaring them “irrational” if they don’t care about what you care about. What you’re suggesting here strikes me as just as silly as if someone went around saying that anyone who doesn’t like the same food or music as them is “irrational.”
//However, if your view is that moral statements express one of those three views, depending on context and the speaker, then you'd have to hold that the reductios to some of those views would apply. //
I don’t. I endorse metaethical indeterminacy: I don’t think ordinary moral claims have any determinate meaning with respect to the realism/antirealism divide, for the same reason I don’t think ordinary causal claims determinately commit ordinary people to the Copenhagen, Many Worlds, or Hidden Variable accounts of quantum mechanics. People’s causal claims don’t require explicit theories about quantum mechanics, nor implicit commitments to them. Likewise, I don’t believe ordinary moral claims require explicit stances towards or implicit commitments to metaethical theories, and that generally speaking I think most people don’t have metaethical views, and wouldn’t understand metaethical positions without significant training, just like quantum mechanics. There are almost certainly occasional outliers who do adopt determinate metaethical stances, so there’s a dash of metaethical variability/pluralism in there, for the same reason a few laypeople have looked into it and favor views in quantum mechanics: in both cases, the fact that such people exist does not mean that there is some kind of public or common meaning to claims like “murder is wrong” and “it is going to rain tomorrow” that have built-in metaethical and quantum mechanical presuppositions, respectively.
//Moral quietism doesn't seem to be much of a view, it just seems to be the notion that we don't need an account of what people mean by it. //
That’s not what my position on moral quietism is.
//That's true, but these are rare exceptions. //
How could you possibly know that without conducting empirical research?
//. Most of the time such statements are declarative//
Now you’re making statements about the proportion of seemingly-declarative statements that are not declarative. This is genuinely bizarre: how would you know without conducting research?
//It can't be settled by a priori reasoning, but a priori reasoning can reveal its implausible implications. //
I don’t even agree with that, for some of the aforementioned reasons. A priori reasoning conducted by a linguistically, culturally, and educationally homogenous and insular community is in an extremely poor position to make judgments about how people outside that community speak and think. Philosophers are quite literally taught to speak and think in highly idiosyncratic and formal ways; if we’re going to get a priori here, my a priorio reasoning tells me that philosophers’ a priori reasoning is not sufficient to do much of anything interesting when it comes to speculating about how ordinary people speak or think.
What I find especially weird about this is…(a) if it could, the only way to show that would, ironically, be to provide corroborating empirical evidence revealing that philosophers’ judgments were reliable all along. I don’t think the research consistently bears this out (b) even if you think a priori philosophical methodology can be helpful, *why not* advocate for or even conduct the relevant empirical work? You seem to read and draw on a bunch of armchair philosophy even when the topics cross over into or would be better-informed by engagement with empirical research.
//One can issue imperatives and emotional expressions directed at things that are characterized morally. If non-cognitivism is true, what do these mean: 'Do the right thing.' 'Hurray for virtue!'//
“Do the right thing” could mean “Doing what we both agree would be the ‘good’ thing to? Yay.” or “Act in a way consistent with an intersubjective set of normative standards we both regard as the ideal standard to emulate” or any number of nonpropositional things. Nothing puzzling about this, and it’s easy for a noncognitivist to handle. “Hurray for virtue” is even easier to handle. Firstly, this is probably not something most people would say or think. But second, even if they did, what’s the issue? Such a person could be conveying a positive attitude for some generalized notion of being courageous or generous or whatever. What’s the problem? No disrespect to Huemer, but I don’t think he’s making much of an attempt to stretch his imagination on these utterances. These can easily be understood by a noncognitivist.
///Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it/
Again, Huemer doesn’t seem to be exercising his interpretative imagination here.
When someone says “We shouldn’t be doing this,” they could be acknowledging descriptive norms, such as norms about what society approves or disapproves of. And these norms may conflict with their desires or emotions. So this sentence could mean something like
“This action would be met with the disapproval of our community if they knew about it, but I don’t care, let’s do it anyway.” Normative language such as “shouldn’t” can be used to reference knowledge of descriptive norms rather than being, itself, an expression of the speaker’s normative standards.
//And yet moral realism is the weird one?!//
Yes. I’ll be blunt and say that I don’t think you’ve presented *anything* to make me even slightly reconsider antirealism.
Virtually every argument here falls back on brute appeals to realist intuitions, or what strikes me as an embarrassingly inadequate method for evaluating the meaning of ordinary moral claims: armchair / a priori philosophy, which, while not useless, comes nowhere close to capably establishing that cognitivism or noncognitivism is true.
The meaning of ordinary moral utterances is an empirical question. The methods of standard analytic philosophy are not the proper methods for establishing what people mean when they make moral claims.
I don’t have realist intuitions, and I don’t think antirealists have any problems handling moral language, nor does moral language strike me as supporting realism, anyway. The best and most recent empirical research on how ordinary people respond to questions about metaethics suggests that most, at least in the United States, favor antirealist positions. Polzler and Wright found that relativism and other antirealist positions were consistently favored over realist positions across a variety of paradigms, while Davis found that, out of all the positions made available to participants, noncognitivism was the most common.
Pölzler, T., & Wright, J. C. (2020). Anti-realist pluralism: A new approach to folk metaethics. Review of Philosophy and Psychology, 11(1), 53-82.
Davis, T. (2021). Beyond objectivism: new methods for studying metaethical intuitions. Philosophical Psychology, 34(1), 125-153.
"“Do the right thing” could mean “Doing what we both agree would be the ‘good’ thing to? Yay.” or “Act in a way consistent with an intersubjective set of normative standards we both regard as the ideal standard to emulate” or any number of nonpropositional things. Nothing puzzling about this, and it’s easy for a noncognitivist to handle. “Hurray for virtue” is even easier to handle. Firstly, this is probably not something most people would say or think. But second, even if they did, what’s the issue? Such a person could be conveying a positive attitude for some generalized notion of being courageous or generous or whatever. What’s the problem? No disrespect to Huemer, but I don’t think he’s making much of an attempt to stretch his imagination on these utterances. These can easily be understood by a noncognitivist."
If do the right thing means "do the thing that we'd agree upon under x conditions," then x being right is propositional--it describes what we'd do under x conditions.
"///Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it/
Again, Huemer doesn’t seem to be exercising his interpretative imagination here.
When someone says “We shouldn’t be doing this,” they could be acknowledging descriptive norms, such as norms about what society approves or disapproves of. And these norms may conflict with their desires or emotions. So this sentence could mean something like
“This action would be met with the disapproval of our community if they knew about it, but I don’t care, let’s do it anyway.” Normative language such as “shouldn’t” can be used to reference knowledge of descriptive norms rather than being, itself, an expression of the speaker’s normative standards."
But that would be propositional. That wouldn't be non cognitivist. In this case, x is good means "x is the thing that would be approved of by our community"
"Virtually every argument here falls back on brute appeals to realist intuitions, or what strikes me as an embarrassingly inadequate method for evaluating the meaning of ordinary moral claims: armchair / a priori philosophy, which, while not useless, comes nowhere close to capably establishing that cognitivism or noncognitivism is true."
Well, they are appeals to intuition. But the point of the article was that anti-realism is unintuitive.
"I don’t have realist intuitions, and I don’t think antirealists have any problems handling moral language, nor does moral language strike me as supporting realism, anyway. The best and most recent empirical research on how ordinary people respond to questions about metaethics suggests that most, at least in the United States, favor antirealist positions. Polzler and Wright found that relativism and other antirealist positions were consistently favored over realist positions across a variety of paradigms, while Davis found that, out of all the positions made available to participants, noncognitivism was the most common."
But I think people aren't great at telling the words that they're using. I think people often claim to be relativists at first, until they hear the implications of it. I don't think that studies that you agree are shoddy are necessarily a trump card to robust a priori arguments.
I know you don't, but the claim here is that many do. Ordinary moral language renders the implications of error theory, subjectivism, and non-cognitivism implausible.
//If do the right thing means "do the thing that we'd agree upon under x conditions," then x being right is propositional--it describes what we'd do under x conditions.//
I agree, but that’s not what I said. I said ““Doing what we both agree would be the ‘good’ thing to? Yay.” An emotional attitude *about* a proposition is not, itself, a proposition. Saying “yay” in response to considering the “2+2=4” is not a proposition. Likewise, responding to a consideration with “yay” is not a proposition.
//But that would be propositional. //
It sure would, but there’s no inconsistency or unintelligibility with being a moral noncognitivist and making claims about descriptive norms. Here’s what Huemer initially said:
“Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it.”
The noncognitivist is not committed to interpreting “we shouldn’t do this” in noncognitivist terms, but they also don’t have to interpret it in line with cognitivist theories. This is because cognitivist theories are theories about the meaning of *normative* moral claims, not descriptive moral claims. It’s a propositional claim to say e.g., “People in our society generally disapprove of lying,” but such claims don’t commit one to cognitivist about first-order normative claims. One can make propositional claims about the norms, institutions, and attitudes of individuals or communities without thereby taking one’s own or other’s moral utterances to consist of propositional claims or to refer to moral properties.
Second, people can use normative language to reference descriptive claims. Consider “It’s wrong to steal here, but I don’t care.” In such cases, “It’s wrong to steal here” can mean something like “it’s illegal to steal here and people who get caught doing so will get punished,” or “members of this community disapprove of stealing.” When people mean this with such a sentence, it isn’t subject to analysis as either cognitivist or noncognitivist, because it’s making a descriptive claim, not a normative one.
As such, my point is that one can interpret the first part of the sentence, “we shouldn’t be doing this,” as a descriptive claim, which doesn’t have anything to do with “reasons” and isn’t even attempting to express a moral propositional claim about what we should or shouldn’t do, despite superficial appearances.
You can be a noncognitivist and say things like “it is a violation of the local law to speed” without this rendering your commitment to noncognitivism intelligible. Huemer is making an error in thinking that the remark, taken as a whole, requires the noncognitivist to interpret superficially normative utterances as normative claims, rather than some such sentences being shorthand for conveying claims about descriptive norms.
Another way to put this is that the mistake Huemer is making is that, even if it were true that the noncognitivist analysis of the remark would be unintelligible “if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it,” the problem is that the noncognitivist doesn’t have to interpret “We shouldn’t be doing this” in noncognitivist terms, but not doing so wouldn’t commit them to cognitivism. Huemer has made the mistake of failing to recognize that “We shouldn’t be doing this” can be interpreted in a way that is not, technically, the appropriate type of moral claim for analyzing along cognitive/noncognitive lines.
Finally, it’s not at all obvious why it would be unintelligible in the first place. Substitute in the suggestions he proposes: “I have a negative feeling towards this, but I don't care; let's do it anyway'.” I’ve personally experienced the compulsion to engage in actions I emotionally disapproved of. So I quite literally have thought things like:
“This? UGH, BOO! But I don’t care, I’ll do it anyway!”
I don’t see what’s unintelligible about this. One can feel a compulsion to do something they have a negative attitude towards.
//In this case, x is good means "x is the thing that would be approved of by our community"//
No, it doesn’t. Someone can be a noncognitivist about normative moral claims and also make propositional claims about descriptive norms. I’m saying that “You shouldn’t do X” can be interpreted not as a normative moral claim but a descriptive claim about what other people think, feel, etc. (or what you think they think or feel).
As such, “we shouldn’t be doing this,” it’s ambiguous between a normative and a descriptive claim. In fact, it’s not even clearly a moral claim. For instance, someone might be talking about wearing pajamas to a work meeting. People often say things like “That’s so wrong” with a positive attitude. For instance, a group of coworkers might plan a prank on another coworker and set up an elaborate trap to hurl a pie in the person’s face. And they might say
“Hahaha, we shouldn’t do this, but who cares! Let’s do it anyway.”
By “we shouldn’t do this,” they don’t have to be making a moral claim. They could just mean something like “This is an outrageous action that is wildly out of accord with what is considered professional workplace behavior” or something.
My point is, that statements like “we shouldn’t do this,” can mean a variety of things, depending on the context, and a noncognitivist isn’t required to interpret all instances of “we shouldn’t do this” in such a way that they convey e.g., emotions or imperatives. They could convey descriptive norms. And, now that I think about it, such remarks don’t even have to be about morality.
//But the point of the article was that anti-realism is unintuitive//
To you.
//I think people often claim to be relativists at first, until they hear the implications of it//
You’re welcome to provide evidence of that. We may not even agree on the implications of relativist. Personally, I don’t think there are much in the way of bad implications, and I don’t trust realists to impartially “inform” people about what the alleged implications of relativism are. Personally, I suspect that most people brought into the realist fold after initially finding relativism appealing do so due primarily to normative entanglement and other rhetorical, emotional, and social appeals, not because there are any bad implications to relativism, nor because arguments for realism are good.
FWIW, I’ve taught courses on ethics and moral psychology, and almost all my students were some kind of moral antirealist (usually relativists), and very few would change under pressure. Granted, I’m not a sincere realist, so I may not be good at putting pressure on them, but I don’t trust realists to put the appropriate kind of pressure on people, either.
//I don't think that studies that you agree are shoddy are necessarily a trump card to robust a priori arguments. //
I agree. But I don’t think you or Huemer have robust a priori arguments in the first place.
//Ordinary moral language renders the implications of error theory, subjectivism, and non-cognitivism implausible. //
It really doesn’t. And you haven’t presented any good arguments that it does. Neither has Huemer.
//Moral predicates can be transformed into abstract nouns, suggesting that they are intended to refer to properties; we talk about 'goodness', 'rightness', and so on, as in 'I am not questioning the act's prudence, but its rightness'. //
You can do the same with emotions, e.g., sadness. This doesn’t suggest much. In any case, the fact that they can be transformed into abstract nouns does not entail that ordinary people speak or think as cognitivists. Merely because one can do so doesn’t mean that people in fact do so, or that a commitment to doing so is implicit in the way they speak or think. Again, this is an empirical question that cannot be resolved from the armchair.
//We ascribe to evaluations the same sort of properties as other propositions. You can say, 'It is true that I have done some wrong things in the past', 'It is false that contraception is murder', and 'It is possible that abortion is wrong'. 'True', 'false', and 'possible' are predicates that we apply only to propositions. //
Same problem as above. Words and concepts can be employed in different ways in different contexts. Simply because people *can* embed moral properties in ways that are suitable for propositional claims does not entail that the primary or central function of moral utterances is to make propositional claims. One simply could use them to make non-propositional utterances in their primary or central usages, or at least some of those uses, and not do so in other instances. Language is flexible and dynamic, and moral terms and concepts can be used, in principle, *both* to convey emotions without conveying propositional content, *and* to convey propositional content, with or without conveying emotions or other non-propositional content. This is why “You will not go to the party” *could* be used as a declarative sentence (e.g., a prediction) or an imperative (e.g., “don’t go to the party”).
The terms and concepts we employ in everyday discourse don’t have fixed, rigid meanings. Their meanings unfold in the relevant contexts of deployment. When philosophers sit around thinking about moral language, they notice that moral language is frequently conveyed using the structure of a declarative sentence. On reflection, nothing seems odd about talking about the properties of “rightness” or “wrongness” … to philosophers, who’ve read Plato and Aristotle and are used to talking about “properties.” But how do they know, and why would they presume, that the way *they* think about these topics, while sitting in the grad student lounge or in a classroom, is reflective of the way most ordinary people out in the world use moral language in everyday contexts? Such people are in different social circumstances, and reflect a vast array of distinct cultural, linguistic, and educational backgrounds, among other factors. Philosophers, meanwhile, systematically diverge from the rest of the populations they come from and from humanity as a whole. Attempting to draw inferences about what people mean, and what they are doing, when they make moral judgments in ordinary life by consulting how an idiosyncratic group of people *outside* those contexts think about these situations is a terrible method: it reflects little more than extremely unsystematic armchair psychology.
If you want to know what people mean when they say things like “murder is wrong,” why on earth would you opt to use the methods of armchair philosophy? Why not use the methods of the social sciences, which are specifically intended for the purpose of gathering systematic data about exactly these sorts of questions.
The same issue applies to (d), (e), and so on: it all relies on the presumption that sentences like “murder is wrong” have fixed meanings, and that philosophical reflection on the characteristics of language is sufficient to draw inferences about how ordinary people use moral language in the real world. I see no reason to think such inferences are warranted.
Just like the “you will not go to the party” inference, such a statement could be unpacked, with its component parts, in such a way so as to show that *one* way one could mean that sentence is, decidedly, to make a propositional claim. Yet it is a simple fact that it can be used as an imperative. All Huemer’s examples illustrate is that moral language can coherently be used cognitivistically. It does *not* show that the central, primary, or exclusive function of moral claims is to express propositional claims. Only that this is one option available to us.
Again, back to the “you will not go to the party” sentence: whether or not sentences of this kind are typically used as imperatives or propositions will be an empirical question: one is simply not in a position to know how typical either usage is without conducting the appropriate research.
Just the same, it could turn out that people use moral claims to express imperatives or emotions 85% of the time, and to make propositional claims 15% of the time, and this is all that would be necessary for (a) - (g) to make sense, yet still not demonstrate that the central, primary, or exclusive function of moral utterances is to make propositional claims.
"You can do the same with emotions, e.g., sadness. This doesn’t suggest much. In any case, the fact that they can be transformed into abstract nouns does not entail that ordinary people speak or think as cognitivists. Merely because one can do so doesn’t mean that people in fact do so, or that a commitment to doing so is implicit in the way they speak or think. Again, this is an empirical question that cannot be resolved from the armchair."
Right, and I think saying "that movie was sad" is propositional, likely subjectivist. But if most people use words like goodness and badness then that shows that they think of it as expressing propositional content. People don't have words like shutthedoorness.
"
Same problem as above. Words and concepts can be employed in different ways in different contexts. Simply because people *can* embed moral properties in ways that are suitable for propositional claims does not entail that the primary or central function of moral utterances is to make propositional claims. One simply could use them to make non-propositional utterances in their primary or central usages, or at least some of those uses, and not do so in other instances. Language is flexible and dynamic, and moral terms and concepts can be used, in principle, *both* to convey emotions without conveying propositional content, *and* to convey propositional content, with or without conveying emotions or other non-propositional content. This is why “You will not go to the party” *could* be used as a declarative sentence (e.g., a prediction) or an imperative (e.g., “don’t go to the party”)."
But if it seems semantically propositional, that seems like good evidence that it's thought to be propositional. This wouldn't be good evidence that if you asked people if moral statements were propositional, after explaining the meaning of terms, they'd say yes. Instead, it's good evidence that based on actual usage, people treat it as propositional. If people didn't treat it as propositional, it's unclear how they would intelligibly ask things like "is this moral?"
"The terms and concepts we employ in everyday discourse don’t have fixed, rigid meanings. Their meanings unfold in the relevant contexts of deployment. When philosophers sit around thinking about moral language, they notice that moral language is frequently conveyed using the structure of a declarative sentence. On reflection, nothing seems odd about talking about the properties of “rightness” or “wrongness” … to philosophers, who’ve read Plato and Aristotle and are used to talking about “properties.” But how do they know, and why would they presume, that the way *they* think about these topics, while sitting in the grad student lounge or in a classroom, is reflective of the way most ordinary people out in the world use moral language in everyday contexts?"
I hear people describe acts rightness--they don't use it that much but it's a thing people say, even lay people.
"If you want to know what people mean when they say things like “murder is wrong,” why on earth would you opt to use the methods of armchair philosophy? Why not use the methods of the social sciences, which are specifically intended for the purpose of gathering systematic data about exactly these sorts of questions."
We should use both. However, I think a priori reasoning can figure out what people mean even if they don't realize they mean it. People will often say they're cultural relativists, until they realize its implications. Most people don't have intelligible fleshed out views about how they use particular moral language.
"Just the same, it could turn out that people use moral claims to express imperatives or emotions 85% of the time, and to make propositional claims 15% of the time, and this is all that would be necessary for (a) - (g) to make sense, yet still not demonstrate that the central, primary, or exclusive function of moral utterances is to make propositional claims."
But it seems like in most uses of moral language that are intended to be moral (So this would rule out sentences like "this is a good ruler," or "that cake tasted good,") if you ask "is that good," it seems people would find that intelligible. If a person says something like "abortion is moral," and you say "do you really think it's moral," they would understand the question (though perhaps being annoyed by your repetitivenesss).
//I hear people describe acts rightness--they don't use it that much but it's a thing people say, even lay people.//
I sometimes talk about tastiness. I’m not a cognitivist about taste. I don’t think the fact that people sometimes talk in property-ish terms entails they “think” or are committed to the existence of properties. People depend heavily on metaphor and analogy, and the way they speak does not require deep or substantive metaphysical commitments.
//However, I think a priori reasoning can figure out what people mean even if they don't realize they mean it. //
How would you know what people mean who don’t realize that’s what they mean using a priori methods alone? Would there be any independent means of confirming this? It seems like you’d be stuck with supposition.
//People will often say they're cultural relativists, until they realize its implications//
Another empirical claim. What do you think the implications of relativism are? We probably disagree about what they are.
//But it seems like in most uses of moral language that are intended to be moral (So this would rule out sentences like "this is a good ruler," or "that cake tasted good,") if you ask "is that good," it seems people would find that intelligible.//
Yes, but that’s only a problem if someone is committed to a uniform and determinate notion of folk noncognitivism. If you think people use moral language in ways that have little to do with metaethics then such remarks may be intelligible for reasons that have little to no implications on whether moral language is cognitivist/noncognitivist.
The intelligibility of the sorts of examples you present is simply not probative of cognitivism or noncognitivism if that intelligibility wouldn’t be cashed out in a way that entailed one or the other.
//If a person says something like "abortion is moral," and you say "do you really think it's moral," they would understand the question (though perhaps being annoyed by your repetitivenesss). //
Even a committed noncognitivist would understand this. I don’t see the problem. It seems a lot of people studying metaethics seem to think that if you’re a noncognitivist, that you would find anything other than the exclusive expression of emotions to be utterly mystifying. This isn’t how noncognitivists think. I don’t know if you’ve spoken to many serious noncognitivists but the reactions to noncognitivists seem to me to caricature the position and treat its adherents like they have an utterly idiotic and completely unsophisticiated way of dealing with moral language that appears inconsistent with their views. We’re not dealing with the kinds of crude emotivism from the 1930s. It’s 2022. Contemporary expressivist accounts are sufficiently sophisticated that they can thread many of the needles you and apparently Huemer seem to think pose insurmountable problems; insurmountable problems to old, flat-footed emotivism? Maybe. But that’s the effectively like kicking a corpse.
//If people didn't treat it as propositional, it's unclear how they would intelligibly ask things like "is this moral?" //
It’s not unclear to me.
I believe philosophers operate under the mistaken impression that moral discourse, and discourse in general, especially discourse surrounding argument, persuasion, and reasoning, primarily serves the function of facilitating the acquisition of true beliefs. I don’t think this is the case. Instead, I believe people argue and reason like lawyers, not philosophers. They start with their conclusions, and then recruit arguments, reasons, etc. so as to achieve their goals. As such, one of the central functions of reasoning is to facilitate argumentation and persuasion.
Let’s extend this to moral discourse, from a noncognitivist framework.
Suppose I have a positive attitude towards certain things, and a negative attitude towards others. I want the world to conform to my attitudes: I want more of the good, less of the bad.
When I interact with others, I recognize that not everyone shares the same emotional reactions to things as me. They want different things than me, they have different emotional attitudes, different degrees of approval and disapproval.
One way of navigating our conflicting desires is to attempt to persuade people to act in accordance with my desires and to not act in ways inconsistent with my desires. Cue the language of persuasion, coalition formation, etc. All superficially propositional language could appear and even function in propositional terms for the purposes of discourse, even if, psychologically speaking, my and other people’s moral positions ultimately boiled down to conflicting emotional attitudes. In such circumstances, I may draw on something approximating syllogistic logic, make claims that appear propositional, etc.; all to borrow the rhetorical force seeming to be speaking in propositional terms, when the propositional discourse is in fact a kind of overlay that is, in fact, only serving the function of attempting to persuade or coerce compliance with my desires. In such circumstances, there would be a type of fundamental insincerity or superficiality in whatever propositional account one offered of the relevant moral discourse; the background mental states of the speakers are, in fact, simply competing emotions. That is, on some level, “murder is wrong” may be offered as a propositional claim for persuasive or rhetorical purposes, even if, in truth, the person making such a claim merely disapproves of murder, and has no substantive or genuine commitment to the existence of moral properties.
In such cases, we might analyze some instances of such language in a way that best fits an analysis that seems to accord with cognitivism; however, this does not mean that the central or primary function of moral claims is to express propositions, or that the person making this claim believes or is actually committed to the existence of moral properties. In truth, there are simply employing declarative sentences to parasitize their rhetorical oomph.
This goes back to that initial example I gave. A mother says to the teen trying to sneak out to go to the party:
“You will NOT go to that party.”
This is a declarative sentence, but it is in fact intended to convey an imperative.
How common are such claims? I have no idea. The point, however, is that the meaning of this claim isn’t determined by the semantics of the sentence, but by the intentions, or mental states, of the mother. That is, the meaning is cashed out not merely by analyzing the content of the sentence, but the mental states of the speaker.
This is how I assess meaning. Call it Lance’s Law if you like: “Words don’t mean things, people mean things.”
I think it makes no sense whatsoever to even ask whether moral language “is propositional.” It isn’t, nor could it be, propositional or nonpropositional in some straightforward way that could be determined merely by armchair analysis of how the sentences appear to operate outside their contexts of actual usage. Outside such contexts, sentences have no meaning at all. Sentences don’t mean anything. Only people using sentences mean things when using those sentences.
The more proper way to pose the question would be, “When ordinary people make normative moral claims, are they intending to make propositional claims or are they intending to do something else?” That is, the question is, “What are speakers trying to do when they make moral claims?”
And if the answer turns out to frequently be something like “convince people to do what they want,” then it’s not at all clear why I should take someone who is doing such a thing, and using propositional language to do so, to be a “cognitivist.” Such a person need not have any few about whether there is any such thing as a first-order moral claim, to think that all such claims share a common semantic analysis, and that that analysis is that they convey propositions. They need not implicitly use language in a way that reliably conforms to this, either.
Instead, they may use language in ways that is flexible and adapts to the context. We might predict, for instance, that a person would use seemingly-realist language when it suited them, and seemingly-antirealist language when it suited them, and that the best prediction of whether they use language in a way that fits a particular metaethical account is whether it is useful for them to do so in that context. If so, such a person would have no principled and uniform commitment to using moral language in a realist or antirealist way; instead, their use would transient and a feature of their argumentative or other social goals.
In other words, there would be no principled implicit shared meaning to such a person’s use of moral language. Their use would be governed by socio-functional goals, not adherence to the principles and norms that motivate academic philosophers, e.g., logical consistency.
Philosophers seem to operate under the mistaken presumption that the way ordinary people speak and think somehow implicitly conforms to the norms and concerns that philosophers care about. I see no reason to think this is the case, and quite a lot of reason to think otherwise. As a result, in sum, I think the answer to questions about the meaning of moral claims will turn on facts about the intentions of the speaker and the context of the utterance, that the answer will be “it varies from speaker to speaker, and context to context,” and that there is no such thing as a single, central meaning to moral claims.
The theory I’ve been drawing on is Sperber and Mercier’s argumentative theory of reasoning. They had a BBS article on it some time ago. See here:
They also put out a book a few years ago that covers the theory: Mercier, H., & Sperber, D. (2017). The enigma of reason. Harvard University Press.
Generally speaking, I think moral philosophers have mistaken notions about how language and moral psychology work, and that moral psychologists also have mistaken notions about how moral psychology works, and that in general everyone is fumbling around as a result of projecting the theories and concerns that motivated 20th century analytic philosophers onto ordinary people, and mistakenly thinking that all the theories and concerns that occupy philosophers are buried and secreted away in the minds and words of the folk. I don’t think t hey are, and that most of the theories philosophers come up with are made up nonsense that have little or even nothing to do with what ordinary people think or how they speak
///But if it seems semantically propositional, that seems like good evidence that it's thought to be propositional.//
I don’t share the same views about language as you do. So I think that things seem “semantically propositional” to you because you probably have weird and mistaken notions about how languages work. I take the meaning of language to be highly context-specific and heavily dependent on background assumptions and pragmatics. So very few things seem “semantically propositional” to me…or, more accurately, I’m not even sure what you mean, and I probably think nothing “seems semantically propositional” - I take the meaning of utterances to depend on the intentions of the speaker; an entire body of discourse can appear superficially propositional but simply not be. It’s always going to end up being an empirical question. More generally, I think a semantics-centric approach to language is, itself, fundamentally misguided and a product of a bizarre conception of how terms and concepts work that’s largely a byproduct of the poor methods taught to people with training in analytic philosophy.
//This wouldn't be good evidence that if you asked people if moral statements were propositional, after explaining the meaning of terms, they'd say yes. Instead, it's good evidence that based on actual usage, people treat it as propositional. //
Without studying *actual* actual usage of moral claims, you’re not even entitled to such claims. All Huemer tends to do is deal with toy moral sentences, e.g., “murder is wrong.”
This is no more a real moral sentence than saying “Bob was murdered” is an actual instance of a murder.
If you want to show that the primary, exclusive, or central usage of moral terms and concepts is propositional, you’re going to need to conduct empirical research. That is, you’re going to need to study actual actual usage, not fake, made-up “actual” usage of the kind Huemer and armchair philosophers generally deal with. Toy sentences aren’t real sentences.
> Most anti-realists agree that if there are irrational desires, moral realism is true. This is true because morality is about what we have impartial reason to do, and impartiality is clearly coherent, so as long as there are desire independent reasons to do certain things and care about other things, moral realism is true.
I haven't read the Parfit book you cite, but this seems incomplete: some desires are irrational because they make it more difficult to do things that you desire more. For instance, if you want just about anything in life, also desiring to kill yourself would be irrational, both because having that desire makes it more likely that you'll act on it, & because (for most humans) having that desire is unpleasant, which is also undesirable. Or, to take a less extreme example, if you want to remain healthy, it is irrational to also want to eat lots of unhealthy food. However, if we exclude this exception, this argument seems correct.
> Here are a bunch of desires that aren’t irrational if anti-realism is true,
Most of these hypotheticals are desires that a human is very unlikely to have (e.g., wanting to eat a car, being indifferent to any amount of pain felt on Tuesdays), but that doesn't mean that it's impossible for any sort of mind to have these desires & be rational in pursuing them. (Likewise, destroying the Earth to turn it into paperclips is obviously irrational to a human, but that doesn't mean that a badly designed AI couldn't have desires that would make this action rational for it.) 4 & 7 are just illustrations of the fact that, for someone who discounts the future highly, acting rationally in the present often makes it harder to get what one will want in the future.
The criticism of noncognitivism you describe seems correct. However, if you think "that moral values are too odd to be part of the fabric of reality" (as I do, since as far as I know there is no evidence that morality is "part of the fabric of reality"), then the only possible conclusion is that there is no morality independent of people's conceptions of morality. Since people do not agree on morality, different people's or societies' ideas of morality must vary; therefore, to say that something is moral or immoral makes sense only within a particular moral worldview. Given this context, most of your supposed counterexamples to subjectivism seem reasonable; e.g., Nazi Germany acted morally by its standards but immorally by our standards. The only exceptions are "Some societal practices are immoral" (most societies have some practices that are immoral by their own standards; e.g., scams & racial stereotyping in modern American society) & "when I say “we should be utilitarians,” and Kant says “we shouldn’t be utilitarians,” we’re not actually disagreeing" (since these quotes clearly refer to objective & not subjective morality).
Is there a way of changing the commenting methods on your posts? It looks like you want to discontinue our exchange but in large part that could be because there's no good way of quoting people (at least that I'm aware of) to make for a clean and orderly discussion. But I'd like to have a written exchange, since that provides a different medium for conveying our ideas than just speaking.
Personally, for instance, I'm still very curious what you take the implications of moral relativism to be. I don't think relativism has *any* troubling implications, so I don't think people could be appropriately dissuaded from relativism by its alleged "implications."
If you really don't want to have this exchange here...which is a bit weird (since what else is a comment section for?) that's your prerogative, but if so, as you suggest, let's do another video chat. Either way I'd like these exchanges to made available to others!
Why don't you make a substack and write up a response, post the link there, and then I can write an article responding.
Part 1:
//Most anti-realists agree that if there are irrational desires, moral realism is true.//
Maybe. This seems like an empirical question. Do you mean most philosophers who identify as moral antirealists, most lay antirealists, or something else?
In any case, I do not agree that if there are irrational desires that moral realism is true.
//This is true because morality is about what we have impartial reason to do//
I disagree. I do not think it’s true that morality is about what we have impartial reason to do. It is well within the purview of a moral theory that it permit or even obligate us to act in ways that are partial, e.g., one may be licensed to exercise a personal prerogative (see e.g., G. A. Cohen, who defends a finite personal prerogative to act in one’s own self-interest, which morally permits partial behavior), or ethical egoism, or moral standards whereby one has an obligation to act partially in one’s self-interest or in the interests of particular individuals or groups over other particular individuals or groups.
//Here are a bunch of desires that aren’t irrational if anti-realism is true//
Before denying whether any of these desires are “irrational,” I’d first have to know what you mean by “irrational.”
Before knowing what you mean, I am not in a position to affirm or deny this. Instead, I’d simply say I have no determinate position pending adequate clarification of what you mean.
//Very intuitive! //
What does this mean? Every one of these scenarios is underdescribed. It’s not that I find them “intuitive,” but that they are neither “intuitive” nor “unintuitive.” They lack the kind of information I’d use to judge whether the desire in question is “irrational” or not.
For instance, suppose I asked you this:
“Is it irrational to jump out of a plane?”
There’s not enough information. A person could be jumping out of a plane without a parachute because they mistakenly believe they can fly. Or a person could be jumping out of a plane with a parachute because they are skydiving. In the former case, the action may be “irrational,” while in the latter it may not be.
Whether I would consider a desire “rational” or not is going to turn on whether the desire is instrumental or non-instrumental. If it’s non-instrumental, then I think it’s a category mistake to label any of these desires “irrational,” for the same reason it makes no sense to tell someone whose favorite color is blue that they’re being irrational: one’s non-instrumental desires are just psychological facts about the agent that are no more subject to evaluation as “rational” or not as their age or hair color. If it’s instrumental, then whether it’s “rational” or not would depend on a variety of considerations, such as whether its conducive to their non-instrumental desires. If e.g., eating a car was conducive to someone’s non-instrumental desires, then why would it be irrational to eat a car? What’s irrational about simply wanting to eat a car? What would it even mean for it to be “irrational”?
//And then we come to moral problems. Moral anti-realism will either be subjectivism (according to which moral facts are true in virtue of what some people think of them, either the speaker or society), error theory (according to which moral claims are semantically intended to be realist, but all moral claims are false), and non-cognitivism, according to which moral claims don’t have a truth value.//
These are not the only options. That’s a claim Huemer made in one of his books, but it’s not true.
Don Loeb endorses moral incoherentism; I endorse a non-semantic generalized antirealism which denies the existence of stance-independent moral facts but isn’t committed to any particular account of the meaning of ordinary terms; this position is coupled with moral quietism and metaethical indeterminacy/variability, which are explicitly expressed in such a way that they involve a rejection of conventional uniform and determinate accounts of the meaning of ordinary moral claims. See here:
Gill, M. B. (2009). Indeterminacy and variability in meta-ethics. Philosophical studies, 145(2), 215-234.
Gill, M. B. (2008). Metaethical variability, incoherence, and error. In W. Sinnott-Armstrong (Ed.), Moral psychology: The cognitive science of morality (Vol. 2) (pp. 387-402). Cambridge, MA: The MIT Press.
Loeb, D. (2008). Moral incoherentism: How to pull a metaphysical rabbit out of a semantic hat. In W. Sinnott-Armstrong (Ed.), Moral psychology: The cognitive science of morality (Vol. 2) (pp. 355-386). Cambridge, MA: The MIT Press.
These may not be common positions, but a position does not have to be common to represent a legitimate alternative. Incoherentism, the IV thesis, and moral quietism (my view) are all legitimate alternatives to the conventional three metaethical positions.
//(a) Evaluative statements take the form of declarative sentences, rather than, say, imperatives, questions, or interjections. //
Imperatives in English are often expressed in declarative form. For instance, if a teenager is caught trying to sneak out of the house, their mother or father may say:
“You will not go to that party.”
This is phrased as a declarative sentence, yet it is best understood (and the parent might confirm this) to be an imperative, not a declarative sentence. Declarative sentences can borrow the rhetorical force of a declarative sentence as a form of emphasis. Sentences that appear declarative can and are used for non-declarative purposes.
Whether or not, and how often, people use moral claims without intending to express a propositional claim is an empirical question, and not one that can be settled by a priori reasoning. Philosophers, Huemer included, who do not engage with or conduct the proper empirical research are simply not employing the appropriate methods for determining what people mean when they say things like “Murder is wrong.” This is a question to be settled by psychologists, linguists, anthropologists, etc., or philosophers competent in the use of the appropriate empirical methods.
Thanks for the reply.
"/Most anti-realists agree that if there are irrational desires, moral realism is true.//
Maybe. This seems like an empirical question. Do you mean most philosophers who identify as moral antirealists, most lay antirealists, or something else?"
I have yet to see any philosopher deny this parity. If there are desire independent reasons, that seems to be sufficient for moral reasons.
"//This is true because morality is about what we have impartial reason to do//
I disagree. I do not think it’s true that morality is about what we have impartial reason to do. It is well within the purview of a moral theory that it permit or even obligate us to act in ways that are partial, e.g., one may be licensed to exercise a personal prerogative (see e.g., G. A. Cohen, who defends a finite personal prerogative to act in one’s own self-interest, which morally permits partial behavior), or ethical egoism, or moral standards whereby one has an obligation to act partially in one’s self-interest or in the interests of particular individuals or groups over other particular individuals or groups."
Well, in that case, for that to be a moral theory, I think it would have to defend that if you were rational and impartial you'd sometimes act in self interested ways. If not, then I wouldn't classify that as a moral theory. I think, for example, that Ayn Rand would be espousing a moral theory because she tried to deduce from first principles why reason alone would be sufficient to turn someone into an egoist (she was also ofc ludicrously wrong).
"Before denying whether any of these desires are “irrational,” I’d first have to know what you mean by “irrational.”
Before knowing what you mean, I am not in a position to affirm or deny this. Instead, I’d simply say I have no determinate position pending adequate clarification of what you mean."
Well, irrational would be roughly synonymous to foolish, stupid, unwise, idiotic, and out of accord with reason. I'd be using a definition that roughly tracks ordinary language. When someone says "you're being irrational," I mean roughly what I think they mean.
"“Is it irrational to jump out of a plane?”
There’s not enough information. A person could be jumping out of a plane without a parachute because they mistakenly believe they can fly. Or a person could be jumping out of a plane with a parachute because they are skydiving. In the former case, the action may be “irrational,” while in the latter it may not be.
Whether I would consider a desire “rational” or not is going to turn on whether the desire is instrumental or non-instrumental. If it’s non-instrumental, then I think it’s a category mistake to label any of these desires “irrational,” for the same reason it makes no sense to tell someone whose favorite color is blue that they’re being irrational: one’s non-instrumental desires are just psychological facts about the agent that are no more subject to evaluation as “rational” or not as their age or hair color. If it’s instrumental, then whether it’s “rational” or not would depend on a variety of considerations, such as whether its conducive to their non-instrumental desires. If e.g., eating a car was conducive to someone’s non-instrumental desires, then why would it be irrational to eat a car? What’s irrational about simply wanting to eat a car? What would it even mean for it to be “irrational”?"
Well, there's nothing irrational about having the thought that it would be fun to eat a car. What would be irrational would be aiming to eat a car if one derived no happiness from it. I think it's useful to distinguish desires and deeper aims. I'm referring to deeper aims here. So I'd like to be able to say that deontologists are making some sort of error, as are those who want to set themselves on fire, despite not finding doing so to be at all pleasant.
"Don Loeb endorses moral incoherentism; I endorse a non-semantic generalized antirealism which denies the existence of stance-independent moral facts but isn’t committed to any particular account of the meaning of ordinary terms; this position is coupled with moral quietism and metaethical indeterminacy/variability"
I'll look into the Loeb view. However, if your view is that moral statements express one of those three views, depending on context and the speaker, then you'd have to hold that the reductios to some of those views would apply. So when I utter the statement "it's immoral to torture infants," what I'm saying is false, given that I think I'm appealing to an external standard.
"These may not be common positions, but a position does not have to be common to represent a legitimate alternative. Incoherentism, the IV thesis, and moral quietism (my view) are all legitimate alternatives to the conventional three metaethical positions."
Moral quietism doesn't seem to be much of a view, it just seems to be the notion that we don't need an account of what people mean by it.
"//(a) Evaluative statements take the form of declarative sentences, rather than, say, imperatives, questions, or interjections. //
Imperatives in English are often expressed in declarative form. For instance, if a teenager is caught trying to sneak out of the house, their mother or father may say:
“You will not go to that party.”
This is phrased as a declarative sentence, yet it is best understood (and the parent might confirm this) to be an imperative, not a declarative sentence. Declarative sentences can borrow the rhetorical force of a declarative sentence as a form of emphasis. Sentences that appear declarative can and are used for non-declarative purposes."
That's true, but these are rare exceptions. Most of the time such statements are declarative. It can't be settled by a priori reasoning, but a priori reasoning can reveal its implausible implications.
Your response style makes it hard to tell where my quotes end and where your responses begin. Is there a way to change the way you’re formatting responses to make it easier to tell where your responses begin? Maybe enable some kind of quoting thing that brackets off quotes?
//I have yet to see any philosopher deny this parity. //
I’m not clear on the context here. What parity?
As far as I know, Joyce accepts categorical epistemic reasons but not categorical moral reasons. So it seems there’s at least one person - arguably the most prominent living error theorist - who may hold that there are desire independent reasons, but that none of them are moral. I’m not sure, and it’s worth double-checking.
However, even if Joyce doesn’t endorse this view now (or didn’t before), I outright deny that if there are desire independent reasons that this is sufficient for “moral” reasons. You’d still have to offer an account of what makes some particular subset of desire independent reasons distinctively moral. There are other normative domains; not all desire independent reasons are necessarily moral ones: it is logically possible to be a normative realist without being a moral realist, and it does not follow that if normative realism is true that moral realism is true.
//Well, in that case, for that to be a moral theory, I think it would have to defend that if you were rational and impartial you'd sometimes act in self interested ways.//
That’s irrelevant. My point is that it isn’t part of the definition of moral theories that they concern what we impartially ought to do. It isn’t simply part of the concept of morality that it be impartial; people can and do incorporate and advocate for partiality being morally permissible or even required. Whether they can convincingly defend such views is irrelevant. This is just the same as if someone were to presume that a “moral theory” is necessarily a theory about consequences, or rules; a deontologist or consequentialist could rightly object that any attempt to define “moral theory” in such a way question-beggingly presumes that their own normative standards are constitutive of what it even means to be a moral theory. I’m claiming you’re doing the same, only with impartiality. I do not agree that moral theories just are theories about what we impartially have reason to do. For that matter, one could be a reasons eliminativist and still endorse normative moral theories, so not only do you not need the impartiality component, you don’t need the “reasons” component, either.
You’re baking your own metaethical and normative stances into your notion of what is required for something to be a moral theory.
//Well, irrational would be roughly synonymous to foolish, stupid, unwise, idiotic, and out of accord with reason.//
That’s insufficient. Each one of these terms could be unpacked in ways consistent with instrumentalist and non-instrumentalist accounts of rationality. I certainly think people can be foolish, stupid, unwise, etc., but my use of all these terms is *also* likely different from your conceptions of these terms, if they are intended to be similar to or synonyms with “rationality” in such a way that they bake in realist or categorical presuppositions.
//I'd be using a definition that roughly tracks ordinary language.//
Do you have empirical evidence on how these terms are used in ordinary language? If not, on what grounds do you claim your use of these terms resembles ordinary language?
//When someone says "you're being irrational," I mean roughly what I think they mean. //
What do you think they mean?
//What would be irrational would be aiming to eat a car if one derived no happiness from it.//
Why? I don’t think it’s constitutive of rationality that it’s not rational to do something you desire to do if you wouldn’t be made happy by it. Although I care about happiness, I don’t think it’s a requirement of a rational agent that they care about happiness and that their actions are directed towards becoming more happy.
I suspect you’re just projecting your own preference for happiness onto others, and declaring them “irrational” if they don’t care about what you care about. What you’re suggesting here strikes me as just as silly as if someone went around saying that anyone who doesn’t like the same food or music as them is “irrational.”
//However, if your view is that moral statements express one of those three views, depending on context and the speaker, then you'd have to hold that the reductios to some of those views would apply. //
I don’t. I endorse metaethical indeterminacy: I don’t think ordinary moral claims have any determinate meaning with respect to the realism/antirealism divide, for the same reason I don’t think ordinary causal claims determinately commit ordinary people to the Copenhagen, Many Worlds, or Hidden Variable accounts of quantum mechanics. People’s causal claims don’t require explicit theories about quantum mechanics, nor implicit commitments to them. Likewise, I don’t believe ordinary moral claims require explicit stances towards or implicit commitments to metaethical theories, and that generally speaking I think most people don’t have metaethical views, and wouldn’t understand metaethical positions without significant training, just like quantum mechanics. There are almost certainly occasional outliers who do adopt determinate metaethical stances, so there’s a dash of metaethical variability/pluralism in there, for the same reason a few laypeople have looked into it and favor views in quantum mechanics: in both cases, the fact that such people exist does not mean that there is some kind of public or common meaning to claims like “murder is wrong” and “it is going to rain tomorrow” that have built-in metaethical and quantum mechanical presuppositions, respectively.
//Moral quietism doesn't seem to be much of a view, it just seems to be the notion that we don't need an account of what people mean by it. //
That’s not what my position on moral quietism is.
//That's true, but these are rare exceptions. //
How could you possibly know that without conducting empirical research?
//. Most of the time such statements are declarative//
Now you’re making statements about the proportion of seemingly-declarative statements that are not declarative. This is genuinely bizarre: how would you know without conducting research?
//It can't be settled by a priori reasoning, but a priori reasoning can reveal its implausible implications. //
I don’t even agree with that, for some of the aforementioned reasons. A priori reasoning conducted by a linguistically, culturally, and educationally homogenous and insular community is in an extremely poor position to make judgments about how people outside that community speak and think. Philosophers are quite literally taught to speak and think in highly idiosyncratic and formal ways; if we’re going to get a priori here, my a priorio reasoning tells me that philosophers’ a priori reasoning is not sufficient to do much of anything interesting when it comes to speculating about how ordinary people speak or think.
What I find especially weird about this is…(a) if it could, the only way to show that would, ironically, be to provide corroborating empirical evidence revealing that philosophers’ judgments were reliable all along. I don’t think the research consistently bears this out (b) even if you think a priori philosophical methodology can be helpful, *why not* advocate for or even conduct the relevant empirical work? You seem to read and draw on a bunch of armchair philosophy even when the topics cross over into or would be better-informed by engagement with empirical research.
Yeah, I'll reformat it for future. Let's just discuss over call at some point, this is getting very jumbled and drawn out.
I love jumbled and drawn out discussions. That's kind of my thing.
Part 3:
//One can issue imperatives and emotional expressions directed at things that are characterized morally. If non-cognitivism is true, what do these mean: 'Do the right thing.' 'Hurray for virtue!'//
“Do the right thing” could mean “Doing what we both agree would be the ‘good’ thing to? Yay.” or “Act in a way consistent with an intersubjective set of normative standards we both regard as the ideal standard to emulate” or any number of nonpropositional things. Nothing puzzling about this, and it’s easy for a noncognitivist to handle. “Hurray for virtue” is even easier to handle. Firstly, this is probably not something most people would say or think. But second, even if they did, what’s the issue? Such a person could be conveying a positive attitude for some generalized notion of being courageous or generous or whatever. What’s the problem? No disrespect to Huemer, but I don’t think he’s making much of an attempt to stretch his imagination on these utterances. These can easily be understood by a noncognitivist.
///Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it/
Again, Huemer doesn’t seem to be exercising his interpretative imagination here.
When someone says “We shouldn’t be doing this,” they could be acknowledging descriptive norms, such as norms about what society approves or disapproves of. And these norms may conflict with their desires or emotions. So this sentence could mean something like
“This action would be met with the disapproval of our community if they knew about it, but I don’t care, let’s do it anyway.” Normative language such as “shouldn’t” can be used to reference knowledge of descriptive norms rather than being, itself, an expression of the speaker’s normative standards.
//And yet moral realism is the weird one?!//
Yes. I’ll be blunt and say that I don’t think you’ve presented *anything* to make me even slightly reconsider antirealism.
Virtually every argument here falls back on brute appeals to realist intuitions, or what strikes me as an embarrassingly inadequate method for evaluating the meaning of ordinary moral claims: armchair / a priori philosophy, which, while not useless, comes nowhere close to capably establishing that cognitivism or noncognitivism is true.
The meaning of ordinary moral utterances is an empirical question. The methods of standard analytic philosophy are not the proper methods for establishing what people mean when they make moral claims.
I don’t have realist intuitions, and I don’t think antirealists have any problems handling moral language, nor does moral language strike me as supporting realism, anyway. The best and most recent empirical research on how ordinary people respond to questions about metaethics suggests that most, at least in the United States, favor antirealist positions. Polzler and Wright found that relativism and other antirealist positions were consistently favored over realist positions across a variety of paradigms, while Davis found that, out of all the positions made available to participants, noncognitivism was the most common.
Pölzler, T., & Wright, J. C. (2020). Anti-realist pluralism: A new approach to folk metaethics. Review of Philosophy and Psychology, 11(1), 53-82.
Davis, T. (2021). Beyond objectivism: new methods for studying metaethical intuitions. Philosophical Psychology, 34(1), 125-153.
"“Do the right thing” could mean “Doing what we both agree would be the ‘good’ thing to? Yay.” or “Act in a way consistent with an intersubjective set of normative standards we both regard as the ideal standard to emulate” or any number of nonpropositional things. Nothing puzzling about this, and it’s easy for a noncognitivist to handle. “Hurray for virtue” is even easier to handle. Firstly, this is probably not something most people would say or think. But second, even if they did, what’s the issue? Such a person could be conveying a positive attitude for some generalized notion of being courageous or generous or whatever. What’s the problem? No disrespect to Huemer, but I don’t think he’s making much of an attempt to stretch his imagination on these utterances. These can easily be understood by a noncognitivist."
If do the right thing means "do the thing that we'd agree upon under x conditions," then x being right is propositional--it describes what we'd do under x conditions.
"///Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it/
Again, Huemer doesn’t seem to be exercising his interpretative imagination here.
When someone says “We shouldn’t be doing this,” they could be acknowledging descriptive norms, such as norms about what society approves or disapproves of. And these norms may conflict with their desires or emotions. So this sentence could mean something like
“This action would be met with the disapproval of our community if they knew about it, but I don’t care, let’s do it anyway.” Normative language such as “shouldn’t” can be used to reference knowledge of descriptive norms rather than being, itself, an expression of the speaker’s normative standards."
But that would be propositional. That wouldn't be non cognitivist. In this case, x is good means "x is the thing that would be approved of by our community"
"Virtually every argument here falls back on brute appeals to realist intuitions, or what strikes me as an embarrassingly inadequate method for evaluating the meaning of ordinary moral claims: armchair / a priori philosophy, which, while not useless, comes nowhere close to capably establishing that cognitivism or noncognitivism is true."
Well, they are appeals to intuition. But the point of the article was that anti-realism is unintuitive.
"I don’t have realist intuitions, and I don’t think antirealists have any problems handling moral language, nor does moral language strike me as supporting realism, anyway. The best and most recent empirical research on how ordinary people respond to questions about metaethics suggests that most, at least in the United States, favor antirealist positions. Polzler and Wright found that relativism and other antirealist positions were consistently favored over realist positions across a variety of paradigms, while Davis found that, out of all the positions made available to participants, noncognitivism was the most common."
But I think people aren't great at telling the words that they're using. I think people often claim to be relativists at first, until they hear the implications of it. I don't think that studies that you agree are shoddy are necessarily a trump card to robust a priori arguments.
I know you don't, but the claim here is that many do. Ordinary moral language renders the implications of error theory, subjectivism, and non-cognitivism implausible.
//If do the right thing means "do the thing that we'd agree upon under x conditions," then x being right is propositional--it describes what we'd do under x conditions.//
I agree, but that’s not what I said. I said ““Doing what we both agree would be the ‘good’ thing to? Yay.” An emotional attitude *about* a proposition is not, itself, a proposition. Saying “yay” in response to considering the “2+2=4” is not a proposition. Likewise, responding to a consideration with “yay” is not a proposition.
//But that would be propositional. //
It sure would, but there’s no inconsistency or unintelligibility with being a moral noncognitivist and making claims about descriptive norms. Here’s what Huemer initially said:
“Even more puzzlingly for the non-cognitivist, you can imagine appropriate contexts for such remarks as, 'We shouldn't be doing this, but I don't care; let's do it anyway'. This is perfectly intelligible, but it would be unintelligible if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it.”
The noncognitivist is not committed to interpreting “we shouldn’t do this” in noncognitivist terms, but they also don’t have to interpret it in line with cognitivist theories. This is because cognitivist theories are theories about the meaning of *normative* moral claims, not descriptive moral claims. It’s a propositional claim to say e.g., “People in our society generally disapprove of lying,” but such claims don’t commit one to cognitivist about first-order normative claims. One can make propositional claims about the norms, institutions, and attitudes of individuals or communities without thereby taking one’s own or other’s moral utterances to consist of propositional claims or to refer to moral properties.
Second, people can use normative language to reference descriptive claims. Consider “It’s wrong to steal here, but I don’t care.” In such cases, “It’s wrong to steal here” can mean something like “it’s illegal to steal here and people who get caught doing so will get punished,” or “members of this community disapprove of stealing.” When people mean this with such a sentence, it isn’t subject to analysis as either cognitivist or noncognitivist, because it’s making a descriptive claim, not a normative one.
As such, my point is that one can interpret the first part of the sentence, “we shouldn’t be doing this,” as a descriptive claim, which doesn’t have anything to do with “reasons” and isn’t even attempting to express a moral propositional claim about what we should or shouldn’t do, despite superficial appearances.
You can be a noncognitivist and say things like “it is a violation of the local law to speed” without this rendering your commitment to noncognitivism intelligible. Huemer is making an error in thinking that the remark, taken as a whole, requires the noncognitivist to interpret superficially normative utterances as normative claims, rather than some such sentences being shorthand for conveying claims about descriptive norms.
Another way to put this is that the mistake Huemer is making is that, even if it were true that the noncognitivist analysis of the remark would be unintelligible “if 'We shouldn't be doing this' either expressed an aversive emotion towards the proposed action or issued an imperative not to do it,” the problem is that the noncognitivist doesn’t have to interpret “We shouldn’t be doing this” in noncognitivist terms, but not doing so wouldn’t commit them to cognitivism. Huemer has made the mistake of failing to recognize that “We shouldn’t be doing this” can be interpreted in a way that is not, technically, the appropriate type of moral claim for analyzing along cognitive/noncognitive lines.
Finally, it’s not at all obvious why it would be unintelligible in the first place. Substitute in the suggestions he proposes: “I have a negative feeling towards this, but I don't care; let's do it anyway'.” I’ve personally experienced the compulsion to engage in actions I emotionally disapproved of. So I quite literally have thought things like:
“This? UGH, BOO! But I don’t care, I’ll do it anyway!”
I don’t see what’s unintelligible about this. One can feel a compulsion to do something they have a negative attitude towards.
//In this case, x is good means "x is the thing that would be approved of by our community"//
No, it doesn’t. Someone can be a noncognitivist about normative moral claims and also make propositional claims about descriptive norms. I’m saying that “You shouldn’t do X” can be interpreted not as a normative moral claim but a descriptive claim about what other people think, feel, etc. (or what you think they think or feel).
As such, “we shouldn’t be doing this,” it’s ambiguous between a normative and a descriptive claim. In fact, it’s not even clearly a moral claim. For instance, someone might be talking about wearing pajamas to a work meeting. People often say things like “That’s so wrong” with a positive attitude. For instance, a group of coworkers might plan a prank on another coworker and set up an elaborate trap to hurl a pie in the person’s face. And they might say
“Hahaha, we shouldn’t do this, but who cares! Let’s do it anyway.”
By “we shouldn’t do this,” they don’t have to be making a moral claim. They could just mean something like “This is an outrageous action that is wildly out of accord with what is considered professional workplace behavior” or something.
My point is, that statements like “we shouldn’t do this,” can mean a variety of things, depending on the context, and a noncognitivist isn’t required to interpret all instances of “we shouldn’t do this” in such a way that they convey e.g., emotions or imperatives. They could convey descriptive norms. And, now that I think about it, such remarks don’t even have to be about morality.
//But the point of the article was that anti-realism is unintuitive//
To you.
//I think people often claim to be relativists at first, until they hear the implications of it//
You’re welcome to provide evidence of that. We may not even agree on the implications of relativist. Personally, I don’t think there are much in the way of bad implications, and I don’t trust realists to impartially “inform” people about what the alleged implications of relativism are. Personally, I suspect that most people brought into the realist fold after initially finding relativism appealing do so due primarily to normative entanglement and other rhetorical, emotional, and social appeals, not because there are any bad implications to relativism, nor because arguments for realism are good.
FWIW, I’ve taught courses on ethics and moral psychology, and almost all my students were some kind of moral antirealist (usually relativists), and very few would change under pressure. Granted, I’m not a sincere realist, so I may not be good at putting pressure on them, but I don’t trust realists to put the appropriate kind of pressure on people, either.
//I don't think that studies that you agree are shoddy are necessarily a trump card to robust a priori arguments. //
I agree. But I don’t think you or Huemer have robust a priori arguments in the first place.
//Ordinary moral language renders the implications of error theory, subjectivism, and non-cognitivism implausible. //
It really doesn’t. And you haven’t presented any good arguments that it does. Neither has Huemer.
Part 2:
//Moral predicates can be transformed into abstract nouns, suggesting that they are intended to refer to properties; we talk about 'goodness', 'rightness', and so on, as in 'I am not questioning the act's prudence, but its rightness'. //
You can do the same with emotions, e.g., sadness. This doesn’t suggest much. In any case, the fact that they can be transformed into abstract nouns does not entail that ordinary people speak or think as cognitivists. Merely because one can do so doesn’t mean that people in fact do so, or that a commitment to doing so is implicit in the way they speak or think. Again, this is an empirical question that cannot be resolved from the armchair.
//We ascribe to evaluations the same sort of properties as other propositions. You can say, 'It is true that I have done some wrong things in the past', 'It is false that contraception is murder', and 'It is possible that abortion is wrong'. 'True', 'false', and 'possible' are predicates that we apply only to propositions. //
Same problem as above. Words and concepts can be employed in different ways in different contexts. Simply because people *can* embed moral properties in ways that are suitable for propositional claims does not entail that the primary or central function of moral utterances is to make propositional claims. One simply could use them to make non-propositional utterances in their primary or central usages, or at least some of those uses, and not do so in other instances. Language is flexible and dynamic, and moral terms and concepts can be used, in principle, *both* to convey emotions without conveying propositional content, *and* to convey propositional content, with or without conveying emotions or other non-propositional content. This is why “You will not go to the party” *could* be used as a declarative sentence (e.g., a prediction) or an imperative (e.g., “don’t go to the party”).
The terms and concepts we employ in everyday discourse don’t have fixed, rigid meanings. Their meanings unfold in the relevant contexts of deployment. When philosophers sit around thinking about moral language, they notice that moral language is frequently conveyed using the structure of a declarative sentence. On reflection, nothing seems odd about talking about the properties of “rightness” or “wrongness” … to philosophers, who’ve read Plato and Aristotle and are used to talking about “properties.” But how do they know, and why would they presume, that the way *they* think about these topics, while sitting in the grad student lounge or in a classroom, is reflective of the way most ordinary people out in the world use moral language in everyday contexts? Such people are in different social circumstances, and reflect a vast array of distinct cultural, linguistic, and educational backgrounds, among other factors. Philosophers, meanwhile, systematically diverge from the rest of the populations they come from and from humanity as a whole. Attempting to draw inferences about what people mean, and what they are doing, when they make moral judgments in ordinary life by consulting how an idiosyncratic group of people *outside* those contexts think about these situations is a terrible method: it reflects little more than extremely unsystematic armchair psychology.
If you want to know what people mean when they say things like “murder is wrong,” why on earth would you opt to use the methods of armchair philosophy? Why not use the methods of the social sciences, which are specifically intended for the purpose of gathering systematic data about exactly these sorts of questions.
The same issue applies to (d), (e), and so on: it all relies on the presumption that sentences like “murder is wrong” have fixed meanings, and that philosophical reflection on the characteristics of language is sufficient to draw inferences about how ordinary people use moral language in the real world. I see no reason to think such inferences are warranted.
Just like the “you will not go to the party” inference, such a statement could be unpacked, with its component parts, in such a way so as to show that *one* way one could mean that sentence is, decidedly, to make a propositional claim. Yet it is a simple fact that it can be used as an imperative. All Huemer’s examples illustrate is that moral language can coherently be used cognitivistically. It does *not* show that the central, primary, or exclusive function of moral claims is to express propositional claims. Only that this is one option available to us.
Again, back to the “you will not go to the party” sentence: whether or not sentences of this kind are typically used as imperatives or propositions will be an empirical question: one is simply not in a position to know how typical either usage is without conducting the appropriate research.
Just the same, it could turn out that people use moral claims to express imperatives or emotions 85% of the time, and to make propositional claims 15% of the time, and this is all that would be necessary for (a) - (g) to make sense, yet still not demonstrate that the central, primary, or exclusive function of moral utterances is to make propositional claims.
"You can do the same with emotions, e.g., sadness. This doesn’t suggest much. In any case, the fact that they can be transformed into abstract nouns does not entail that ordinary people speak or think as cognitivists. Merely because one can do so doesn’t mean that people in fact do so, or that a commitment to doing so is implicit in the way they speak or think. Again, this is an empirical question that cannot be resolved from the armchair."
Right, and I think saying "that movie was sad" is propositional, likely subjectivist. But if most people use words like goodness and badness then that shows that they think of it as expressing propositional content. People don't have words like shutthedoorness.
"
Same problem as above. Words and concepts can be employed in different ways in different contexts. Simply because people *can* embed moral properties in ways that are suitable for propositional claims does not entail that the primary or central function of moral utterances is to make propositional claims. One simply could use them to make non-propositional utterances in their primary or central usages, or at least some of those uses, and not do so in other instances. Language is flexible and dynamic, and moral terms and concepts can be used, in principle, *both* to convey emotions without conveying propositional content, *and* to convey propositional content, with or without conveying emotions or other non-propositional content. This is why “You will not go to the party” *could* be used as a declarative sentence (e.g., a prediction) or an imperative (e.g., “don’t go to the party”)."
But if it seems semantically propositional, that seems like good evidence that it's thought to be propositional. This wouldn't be good evidence that if you asked people if moral statements were propositional, after explaining the meaning of terms, they'd say yes. Instead, it's good evidence that based on actual usage, people treat it as propositional. If people didn't treat it as propositional, it's unclear how they would intelligibly ask things like "is this moral?"
"The terms and concepts we employ in everyday discourse don’t have fixed, rigid meanings. Their meanings unfold in the relevant contexts of deployment. When philosophers sit around thinking about moral language, they notice that moral language is frequently conveyed using the structure of a declarative sentence. On reflection, nothing seems odd about talking about the properties of “rightness” or “wrongness” … to philosophers, who’ve read Plato and Aristotle and are used to talking about “properties.” But how do they know, and why would they presume, that the way *they* think about these topics, while sitting in the grad student lounge or in a classroom, is reflective of the way most ordinary people out in the world use moral language in everyday contexts?"
I hear people describe acts rightness--they don't use it that much but it's a thing people say, even lay people.
"If you want to know what people mean when they say things like “murder is wrong,” why on earth would you opt to use the methods of armchair philosophy? Why not use the methods of the social sciences, which are specifically intended for the purpose of gathering systematic data about exactly these sorts of questions."
We should use both. However, I think a priori reasoning can figure out what people mean even if they don't realize they mean it. People will often say they're cultural relativists, until they realize its implications. Most people don't have intelligible fleshed out views about how they use particular moral language.
"Just the same, it could turn out that people use moral claims to express imperatives or emotions 85% of the time, and to make propositional claims 15% of the time, and this is all that would be necessary for (a) - (g) to make sense, yet still not demonstrate that the central, primary, or exclusive function of moral utterances is to make propositional claims."
But it seems like in most uses of moral language that are intended to be moral (So this would rule out sentences like "this is a good ruler," or "that cake tasted good,") if you ask "is that good," it seems people would find that intelligible. If a person says something like "abortion is moral," and you say "do you really think it's moral," they would understand the question (though perhaps being annoyed by your repetitivenesss).
Reply part 3:
//I hear people describe acts rightness--they don't use it that much but it's a thing people say, even lay people.//
I sometimes talk about tastiness. I’m not a cognitivist about taste. I don’t think the fact that people sometimes talk in property-ish terms entails they “think” or are committed to the existence of properties. People depend heavily on metaphor and analogy, and the way they speak does not require deep or substantive metaphysical commitments.
//However, I think a priori reasoning can figure out what people mean even if they don't realize they mean it. //
How would you know what people mean who don’t realize that’s what they mean using a priori methods alone? Would there be any independent means of confirming this? It seems like you’d be stuck with supposition.
//People will often say they're cultural relativists, until they realize its implications//
Another empirical claim. What do you think the implications of relativism are? We probably disagree about what they are.
//But it seems like in most uses of moral language that are intended to be moral (So this would rule out sentences like "this is a good ruler," or "that cake tasted good,") if you ask "is that good," it seems people would find that intelligible.//
Yes, but that’s only a problem if someone is committed to a uniform and determinate notion of folk noncognitivism. If you think people use moral language in ways that have little to do with metaethics then such remarks may be intelligible for reasons that have little to no implications on whether moral language is cognitivist/noncognitivist.
The intelligibility of the sorts of examples you present is simply not probative of cognitivism or noncognitivism if that intelligibility wouldn’t be cashed out in a way that entailed one or the other.
//If a person says something like "abortion is moral," and you say "do you really think it's moral," they would understand the question (though perhaps being annoyed by your repetitivenesss). //
Even a committed noncognitivist would understand this. I don’t see the problem. It seems a lot of people studying metaethics seem to think that if you’re a noncognitivist, that you would find anything other than the exclusive expression of emotions to be utterly mystifying. This isn’t how noncognitivists think. I don’t know if you’ve spoken to many serious noncognitivists but the reactions to noncognitivists seem to me to caricature the position and treat its adherents like they have an utterly idiotic and completely unsophisticiated way of dealing with moral language that appears inconsistent with their views. We’re not dealing with the kinds of crude emotivism from the 1930s. It’s 2022. Contemporary expressivist accounts are sufficiently sophisticated that they can thread many of the needles you and apparently Huemer seem to think pose insurmountable problems; insurmountable problems to old, flat-footed emotivism? Maybe. But that’s the effectively like kicking a corpse.
Reply part 2:
//If people didn't treat it as propositional, it's unclear how they would intelligibly ask things like "is this moral?" //
It’s not unclear to me.
I believe philosophers operate under the mistaken impression that moral discourse, and discourse in general, especially discourse surrounding argument, persuasion, and reasoning, primarily serves the function of facilitating the acquisition of true beliefs. I don’t think this is the case. Instead, I believe people argue and reason like lawyers, not philosophers. They start with their conclusions, and then recruit arguments, reasons, etc. so as to achieve their goals. As such, one of the central functions of reasoning is to facilitate argumentation and persuasion.
Let’s extend this to moral discourse, from a noncognitivist framework.
Suppose I have a positive attitude towards certain things, and a negative attitude towards others. I want the world to conform to my attitudes: I want more of the good, less of the bad.
When I interact with others, I recognize that not everyone shares the same emotional reactions to things as me. They want different things than me, they have different emotional attitudes, different degrees of approval and disapproval.
One way of navigating our conflicting desires is to attempt to persuade people to act in accordance with my desires and to not act in ways inconsistent with my desires. Cue the language of persuasion, coalition formation, etc. All superficially propositional language could appear and even function in propositional terms for the purposes of discourse, even if, psychologically speaking, my and other people’s moral positions ultimately boiled down to conflicting emotional attitudes. In such circumstances, I may draw on something approximating syllogistic logic, make claims that appear propositional, etc.; all to borrow the rhetorical force seeming to be speaking in propositional terms, when the propositional discourse is in fact a kind of overlay that is, in fact, only serving the function of attempting to persuade or coerce compliance with my desires. In such circumstances, there would be a type of fundamental insincerity or superficiality in whatever propositional account one offered of the relevant moral discourse; the background mental states of the speakers are, in fact, simply competing emotions. That is, on some level, “murder is wrong” may be offered as a propositional claim for persuasive or rhetorical purposes, even if, in truth, the person making such a claim merely disapproves of murder, and has no substantive or genuine commitment to the existence of moral properties.
In such cases, we might analyze some instances of such language in a way that best fits an analysis that seems to accord with cognitivism; however, this does not mean that the central or primary function of moral claims is to express propositions, or that the person making this claim believes or is actually committed to the existence of moral properties. In truth, there are simply employing declarative sentences to parasitize their rhetorical oomph.
This goes back to that initial example I gave. A mother says to the teen trying to sneak out to go to the party:
“You will NOT go to that party.”
This is a declarative sentence, but it is in fact intended to convey an imperative.
How common are such claims? I have no idea. The point, however, is that the meaning of this claim isn’t determined by the semantics of the sentence, but by the intentions, or mental states, of the mother. That is, the meaning is cashed out not merely by analyzing the content of the sentence, but the mental states of the speaker.
This is how I assess meaning. Call it Lance’s Law if you like: “Words don’t mean things, people mean things.”
I think it makes no sense whatsoever to even ask whether moral language “is propositional.” It isn’t, nor could it be, propositional or nonpropositional in some straightforward way that could be determined merely by armchair analysis of how the sentences appear to operate outside their contexts of actual usage. Outside such contexts, sentences have no meaning at all. Sentences don’t mean anything. Only people using sentences mean things when using those sentences.
The more proper way to pose the question would be, “When ordinary people make normative moral claims, are they intending to make propositional claims or are they intending to do something else?” That is, the question is, “What are speakers trying to do when they make moral claims?”
And if the answer turns out to frequently be something like “convince people to do what they want,” then it’s not at all clear why I should take someone who is doing such a thing, and using propositional language to do so, to be a “cognitivist.” Such a person need not have any few about whether there is any such thing as a first-order moral claim, to think that all such claims share a common semantic analysis, and that that analysis is that they convey propositions. They need not implicitly use language in a way that reliably conforms to this, either.
Instead, they may use language in ways that is flexible and adapts to the context. We might predict, for instance, that a person would use seemingly-realist language when it suited them, and seemingly-antirealist language when it suited them, and that the best prediction of whether they use language in a way that fits a particular metaethical account is whether it is useful for them to do so in that context. If so, such a person would have no principled and uniform commitment to using moral language in a realist or antirealist way; instead, their use would transient and a feature of their argumentative or other social goals.
In other words, there would be no principled implicit shared meaning to such a person’s use of moral language. Their use would be governed by socio-functional goals, not adherence to the principles and norms that motivate academic philosophers, e.g., logical consistency.
Philosophers seem to operate under the mistaken presumption that the way ordinary people speak and think somehow implicitly conforms to the norms and concerns that philosophers care about. I see no reason to think this is the case, and quite a lot of reason to think otherwise. As a result, in sum, I think the answer to questions about the meaning of moral claims will turn on facts about the intentions of the speaker and the context of the utterance, that the answer will be “it varies from speaker to speaker, and context to context,” and that there is no such thing as a single, central meaning to moral claims.
The theory I’ve been drawing on is Sperber and Mercier’s argumentative theory of reasoning. They had a BBS article on it some time ago. See here:
https://www.cambridge.org/core/journals/behavioral-and-brain-sciences/article/abs/why-do-humans-reason-arguments-for-an-argumentative-theory/53E3F3180014E80E8BE9FB7A2DD44049
They also put out a book a few years ago that covers the theory: Mercier, H., & Sperber, D. (2017). The enigma of reason. Harvard University Press.
Generally speaking, I think moral philosophers have mistaken notions about how language and moral psychology work, and that moral psychologists also have mistaken notions about how moral psychology works, and that in general everyone is fumbling around as a result of projecting the theories and concerns that motivated 20th century analytic philosophers onto ordinary people, and mistakenly thinking that all the theories and concerns that occupy philosophers are buried and secreted away in the minds and words of the folk. I don’t think t hey are, and that most of the theories philosophers come up with are made up nonsense that have little or even nothing to do with what ordinary people think or how they speak
Reply Part 1:
///But if it seems semantically propositional, that seems like good evidence that it's thought to be propositional.//
I don’t share the same views about language as you do. So I think that things seem “semantically propositional” to you because you probably have weird and mistaken notions about how languages work. I take the meaning of language to be highly context-specific and heavily dependent on background assumptions and pragmatics. So very few things seem “semantically propositional” to me…or, more accurately, I’m not even sure what you mean, and I probably think nothing “seems semantically propositional” - I take the meaning of utterances to depend on the intentions of the speaker; an entire body of discourse can appear superficially propositional but simply not be. It’s always going to end up being an empirical question. More generally, I think a semantics-centric approach to language is, itself, fundamentally misguided and a product of a bizarre conception of how terms and concepts work that’s largely a byproduct of the poor methods taught to people with training in analytic philosophy.
//This wouldn't be good evidence that if you asked people if moral statements were propositional, after explaining the meaning of terms, they'd say yes. Instead, it's good evidence that based on actual usage, people treat it as propositional. //
Without studying *actual* actual usage of moral claims, you’re not even entitled to such claims. All Huemer tends to do is deal with toy moral sentences, e.g., “murder is wrong.”
This is no more a real moral sentence than saying “Bob was murdered” is an actual instance of a murder.
If you want to show that the primary, exclusive, or central usage of moral terms and concepts is propositional, you’re going to need to conduct empirical research. That is, you’re going to need to study actual actual usage, not fake, made-up “actual” usage of the kind Huemer and armchair philosophers generally deal with. Toy sentences aren’t real sentences.