I think a lot of these discourse features can be explained as a consequence of the role that morality plays in human society. Because we are a social creature we need to agree a moral code in order the function as a group. If you think it's fine to graze your animals in the communal fields and I think it's not, it is absolutely required that we come to an agreement on the matter or otherwise we end up in conflict. Morality has this feature in common with empirical facts. If I believe you *did* graze your livestock on the field but you deny that, we are once again in conflict. But if we disagree about the tastiness of licorice, no such problems arise. That's why people don't nearly feel the same compulsion to impose their aesthetic judgements on others. But even there you sometimes hear discourse that seems to imply that there is an objectively best or worst movie. I reckon instincts like those stem from communities needing to agree on which crops to plant or where to set up camp.
I think this is a reasonable response for sure, but the problem for me is that the more complex our justifications get for the divide, the more likely it is that they'll begin to include plausibly objective features that make it hard for things like social cohesion or conflict resolution to clearly count as counterexamples. It's very easy for a realist like me to just take the opposite approach of maintaining the objective/subjective divide and seeing what you're bringing up as evidence for the objectivity of social cohesion's goodness or the badness of conflict.
Sorry, that was a little unclear. All I mean is that, if you're a moral realist, then you almost certainly believe that a huge portion of the objective moral facts out there are facts that relate to human development and flourishing. So that means you'd think conflicts over grazing or setting up camp or otherwise engaging in the sorts of things that "keep society running" would be questions with objective answers: It's objectively true that Grog shouldn't graze his cattle on the shared lands too much, or that Lucy should do her fair share to set up camp, or whatever. And so if you accept the objective/subjective divide I laid out, and you're a moral realist, then you should believe that disputes with the ability to really impact human lives will be discussed in the way we're talking about, and that disputes without any such an impact will be talked about differently. In other words, while you're right that the divergence in how we engage in discourse could be explained as "things that impact people vs. things that don't impact people," a moral realist should just think that division is coextensive with "disputes over objective facts versus disputes over subjective facts," since impacting people automatically gets your involved moral facts that are objective. So I'm just saying that, even if the division you're giving in your original comment is a totally valid way of conceptualizing things, a moral realist can accept that division and just say it's fundamentally wrapped up with the other division I'm giving, rather than being a separate explanation that's actually better. Does that make sense?
Tell me if I'm interpenetrating correctly. Your contention seems to be that the fact that morality and descriptive fact share these pragmatic features is not a coincident. They both "bite back" because they are both objectively real, while aesthetic facts does not "bite back" and is subjective.
For example, if society became convinced that stealing is ethical it would fall apart. This counts as evidence that the badness of theft is objective. The logic is similar to the categorical imperative.
But I don't think that quite works. The commonalities between moral discourse and empirical discourse were more front and center in my original comment, because it were those that I meant to explain. But there are differences too. When people disagree about descriptive fact it's essential not only that we agree on a common reality, it is usually also essential that we agree on the one reality that is objectively correct. If your society agrees that a poisonous mushroom is in fact save to eat, you will run into problems.
But in the case of morality it is possible to build a successful society on any number of mutually exclusive moral frameworks. A large portion of societies in history have functioned and perpetuated themselves while keeping slaves in bondage. If there is an objective morality to bite back on moral falsehoods, you'd expect it to bite back on these societies.
I foresee the objection that many of the same societies that were wrong on slavery were also wrong about a lot of factual things, and reality did not bite back on them either. One example of that is geocentrism, which was near universally believed until the renaissance. But geocentrism concerns itself with the nature of outer space, which these societies never interacted with (apart from receiving sunlight, which they could model more or less correctly). While slavery is something deeply intertwined into their societies. A second example is medicine. Where I think false believes did bite back on them, but since nobody knew the correct answer, everyone was equally bitten. Once some societies figured out the truth, they rapidly experienced a lot of success relative to their peers.
Well, it's not really an argument about pragmatic consequences - it's true that a culture where stealing is condoned would fall apart, but that doesn't prove stealing is objectively wrong, because it could be that a society falling apart isn't objectively bad. Rather, my point is just this:
Start with the assumption that disputes regarding issues that meaningfully impact human beings are moral disputes, and the assumption that moral disputes relate to objective facts (i.e. the moral realist assumption). Then accept the subjective/objective distinction I've put forward to explain why our discursive practices diverge. Those three beliefs in tandem predict that we will use "objective-style discourse" when we engage on questions that meaningfully impact human beings, because those are disputes relating to objective facts. So therefore, the fact that we use these discursive practices to speak about things that meaningfully impact human beings isn't a challenge to the subjective/objective explanation.
As a rough analogy, imagine I explain why we let some people sit in the front row of a concert but not others by saying it depends on whether or not they bought a special ticket. And then someone else comes along and says "I don't think you need to rely on the ticket to explain things - I think it's just that some of them are really big fans of the musician and others aren't." That distinction between big fans and just regular fans might work as a solid heuristic for determining whether they'll sit in the front row, but that doesn't mean the "bought the special ticket or just a regular one" isn't the actual deciding factor (because the big fans buy the special ticket and the regular fans don't). Similarly, even if it's the case that we can map our discursive practices onto an "impacts people vs doesn't impact people" standard, that doesn't mean the "objective vs subjective" standard is wrong (because stuff that impacts people is automatically objective and vice versa).
But also, I would say reality did "bite back on" slave societies in the relevant way we'd expect it to - those societies really did have meaningfully lower levels of human flourishing, which is what the moral facts in question relate to. I don't think we should necessarily expect a transgression of those moral facts to have any other result.
>>The second preliminary point I’d like to make is that the firm line between practical ethics and metaethics that antirealists often take for granted isn’t actually obvious, or somehow guaranteed by more general facts about how different levels of discourse interact. In a purely a priori sense, we have absolutely no reason to think that radical changes to our second-order understanding of the terms and concepts involved in a particular discourse will have no impact whatsoever on whether our first-order discursive practices are justified [...]
On this point, I'd draw a distinction between whether our second-order views logically entail any particular consequences, and whether they tend, in practice, to result in changes in our first-order normative moral standards, as well as our sentiments, attitudes, dispositions, and so on. The former may be a more formal question about the conceptual relation between metaethics and normative ethics, while the latter may be more a question of human psychology that cannot readily be settled without empirical evidence.
Another feature of moral discourse that only seems grounded if morality is in fact objective is moral theorizing. People come up with theories of what makes something moral or immoral, or at least general rules of thumb, and decide what particular actions are right or wrong based on those theories. This is a regular feature of objective discourses - we make predictions about the results of an experiment based on our physical theories or about the outcome of an event based on rules of thumb we've learned from previous similar events. And it's obviously grounded in the objective nature of the discourse - if morality is objective, then there should be some description of what it is, and that description will likely have features like simplicity that theories of objectively existing things tend to have. But if morality is stance-dependent, then we don't have to worry about theoretical virtues - it can be whatever our stance says it is. Furthermore, we would never have any reason to base our judgement of a particular case on a general theory, just as you wouldn't change your stance on whether gummy bears taste good even if you had learned a general rule of thumb that implied that they should or shouldn't taste good.
Another feature is completeness - this goes nicely with consistency as mentioned in the article. When discussing moral matters, we assume that there is some fact of the matter about any particular case, even one we haven't thought about before. We assume, for example, that either it is wrong to lie in order to make someone happy (in a specific case that specifies all morally relevant details), or it is not wrong. But if morality is stance-dependent, it could be that our stance is simply silent on the matter. Our stance neither judges it to be wrong, nor for it to be non-wrong because it is an ambiguous stance. In fact, for almost all accounts of stance-dependent moral facts, this will be the case for most moral dilemmas. But then why debate moral dilemmas in the first place?
Yeah, one thing I regret is that I didn't do more to spell out exactly what I meant by "consistency," because it's basically an analogue to what you're saying here. It's hard for me to understand how antirealism can ground universal claims like "Torture is wrong" solely by reference to individual experiences of negative feelings towards torture - we expect that kind of consistency in situations where we're discussing objective properties, but there's nothing weird about the idea that you might disapprove of several instances of torture but semi-randomly approve of another.
As an error theory adherent of sorts (local error theory i.e. categoricity-only rejection) I'm not sure if our moral discourse needs justification in a reflective sense. That is, the whole fact that we have a moral discourse. But I also don't say that serious revisions to whatever implicit second-order beliefs people happen to hold would have no bearing on actual moral behavior. Seems like it could. An adherent to error theory could also advise a noble lie, or to let sleeping dogs lie, or have a whole other attempt at making moral behavior intelligible:
In any case, I don't perceive myself to need to believe anything like categoricity to operate in everyday life irt descriptive language. So, this part
"The essential features of moral discourse are best justified in other non-moral domains by reference to the objectivity of the facts involved, and this gives us good reason to believe that objectivity is what justifies those essential features in moral discourse as well."
doesn't strike me as decisive even if we grant it, because objectivity is not enough for any strong moral realism. You need categoricity too.
This is a good example of a well-made philosophical argument in that once you start agreeing with it you find it very difficult to stop (and I don't just mean that it's deductively valid). I would, of course, disagree with or make modifications to basically every premise and assertion in this article (lol) but I can feel why it would be compelling to someone who is more on the fence than I am. It kind of speaks to a belief in "mere" moral discourse that must either be wholly disregarded or inevitably accepted as indicative of realism. Also, thanks for mentioning error theorist fictionalist 🥂.
Very interesting. I'm not terribly familiar with the debate between realism and non-realism, but I think I have a counter example of a domain that's not objective, where we engage in similar ways.
Take for example a workplace, where the boss sets the expectation that staff will follow a certain dress code, and complete certain tasks by certain times. The boss's expectations here are not objective facts, but something she wilfully sets. But she will still assign out praise and blame based on those expectations, still be expected by others for her expectations to be consistent, and in all likelihood she will be intransigent and largely unwilling to compromise on her expectations.
We might look at other instances of non-objective expectations/standards as well, such as standards of spelling, grammar, or good manners. We absolutely try to impose these expectations/standards on others, assigning praise and blame, expecting consistency, and being intransigent about them.
Yes, this is definitely a complicating factor that you're right to bring up. My take on it would be, briefly, that the assignation of praise and blame (and the presence of related features) is justified only to the degree that the subjective standards are themselves justified on the basis of objective second-order standards. So, for example, it might be justified on an objective basis to have *some* sort of dress code, because it prevents conflicts and improves the image of the business. And once that objective need has been established, then it would be derivatively justified to adopt whatever subjective standards are needed to fulfill that goal. Etiquette sorta works like this too - the behaviors it polices are subjective, in the sense that there's nothing objectively bad about picking your nose at dinner or something, but *that there be some set of social standards that are enforced* could still be objectively justified in a way that makes the specific subjective standards we pick justified too.
I think that would work, but is it implausible that they might just be unjustified and not really need justification? The boss might make up rules just because, and while they're unjustified, she has the power and can wield it how she likes.
Perhaps morality likewise is nothing more than us setting expectations and standards on each other, and the only grounding for it is our will/desires. The justification would then be just a question of whether we can convince (or coerce) others to uphold the standards.
The oddity with the liquorice case is perhaps just, why would anyone care what sweets others like? But then, people do, strangely, care about such things, like opposing pineapple on pizza, or worrying about others' sex lives. It seems your argument requires not condemning pineapple on pizza!
I just wrote an absurdly long response to this, but I was immensely and especially puzzled by this remark:
>>Antirealists might also be tempted to adopt a more radical approach that denies the need to justify these features of discourse entirely. They can just say these sorts of discursive practices are appropriate in one domain but not another, and that’s just the way it is. But a move like this imposes a major theoretical cost, in that it reduces all questions of justification to mere social convention or else renders the entire project incomprehensible.
Why should I think there's anything "radical" about this? What does that mean? It never even occurred to me consider whether the features of moral discourse were "justified." What does that that mean, too? Justified according to what standard? I'm not even sure it makes sense to say they're "appropriate in one domain but not another."
People engage in moral discourse. Why would I need to think doing so was "justified," and that I have to explain why it's justified? And why would I think I'd be paying any sort of theoretical cost if I didn't think it was justified? What cost would that be?
I don't even know what notion of "justification" you're working with. And perhaps the projects you are engaged in simply are incomprehensible. I certainly think a lot of analytic philosophy is pointless or nonsensical. I don't think I'd be paying any sort of cost if I thought the same of a bit more of what philosophers do.
What I suppose might help is if you say a bit more about what you think justification is, why you think moral discourse is justified, and why someone would think they'd be paying some kind of cost if they denied this.
Honestly, I'm not exactly sure how to answer this question. If you're asking me for a complete theory of normative justification, then I wouldn't say I have one. But if you're just asking what I mean when I make claims about certain beliefs, attitudes, actions, etc being justified in the context of first-order discourse, do we really disagree on that? Using the example in the piece, would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears? As I said in the piece, I'm not attaching any particular normative framework or analysis at the start. But would you just, in daily life, say that was unjustified or no?
I don't need a complete theory. I simply don't know what you have in mind by "justification." Like at all.
>>But if you're just asking what I mean when I make claims about certain beliefs, attitudes, actions, etc being justified in the context of first-order discourse, do we really disagree on that?
I think we probably do, yes.
>>Using the example in the piece, would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears?
I don't know, because I don't know what you're asking me. I think it'd be mean for someone to do that, and I wouldn't like if they did it. I might even think they were insane. But what do you have in mind when you suggest they would be "unjustified" in doing this?
>>But would you just, in daily life, say that was unjustified or no?
Probably not. If I did, I'd mean something extremely reductive like that it was rude or I didn't approve of them doing it. Is that what you mean by justification? I imagine not.
Are you saying, then, that you never think of actions or beliefs you encounter in your own life as justified or unjustified? If you were in a meeting at work and made a suggestion, and your boss said "I don't think that would be justified right now," you wouldn't know what they were trying to say or what you would need to do in order to dispute their judgment? Or if I asked you what would justify leaving a restaurant without paying, you wouldn't be able to conceive of a coherent answer?
I certainly don't think explicitly in terms of them being "justified" or "unjustified." I really mean it when I say I don't know what you mean by that. I use the *term* "justified," but this is just another way of saying something about my expectations or attitudes in most contexts. It might mean something different in some other context. I don't think there's some shared, underlying notion of "justification" that drives how I speak or think (nor do I think this is the case for pretty much anyone else, except maybe philosophers).
>>If you were in a meeting at work and made a suggestion, and your boss said "I don't think that would be justified right now," you wouldn't know what they were trying to say or what you would need to do in order to dispute their judgment?
I'd infer their meaning from the context. And what people mean by the term "justified" would vary from one context to another. I don't have enough information about the context here to have a sharper sense of what my boss meant, but I can imagine a context where this would make perfect sense. For instance, suppose the company is having serious financial problems and I suggest we file for bankruptcy. If they said they didn't think that'd be justified, I'd interpret them to mean something like "that's a drastic action, and our finances aren't so bad that I'd consider that the best course of option just yet." The use of "just now" would imply something like that they're open to it being an option in the future, though.
My boss could have just as easily have phrased the remark without any reference to "justification." I don't think this kind of remark calls for any distinctive conception of "justification" to do the relevant work it was doing in some real world context.
So when you start talking about justification in a philosophical context, this is some other context where the relevant cues I think I and nonphilosophers typically use to infer someone's meaning are no longer present. In other words, whatever it is you are doing with the term "justification" when you write about philosophy, I do not take you to mean the same thing as what I take my boss to mean. I've written about this quite a bit: I think what people mean by the words and phrases they use only emerges in the relevant contexts of usage. While I am fully capable of inferring the meaning of a term in a non-philosophical context, I don't grant that we can transpose meaning in those contexts over to philosophical contexts. I think philosophers thus have a responsibility to explain what they mean in the relevant (philosophical) contexts of usage.
So if your question is intended to suggest I struggle to understand everyday use of the term "justification" No. I don't. But I am explicitly denying that such usages necessary match your usage outside of those contexts.
The same goes for this remark:
>> Or if I asked you what would justify leaving a restaurant without paying, you wouldn't be able to conceive of a coherent answer?
It would be easy to give a coherent answer. Here are some possible "justifications": I told them I have a food allergy and they served food loaded in whatever it is I am allergic to, I couldn't eat it, I got sick, and they insisted I pay. Or the waiter physically assaulted me. Or the restaurant is on fire. Or if I own the restaurant and don't have to pay. And so on.
Note that if you think these are perfectly fine justifications, I can revisit the gummy bear situation:
>> would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears?
If I thought that people who prefer licorice over gummy bears were evil and should not be accepted in polite society, I'd not only think it was justified to scream at them, I might think it was mandatory to do so. Screaming at them might not even be going far enough. It might be mandatory to consign the person to life in prison. And I don't think I'd be making any errors in doing so, and I don't think anyone could provide a single decent reason for me to think someone who felt this way was making any kind of error (and I would very strongly object to any suggestion that such a stance involves "admitting defeat".)
FWIW: I think you and I might have very different views about language and meaning, and that his may influence our downstream positions on these matters.
Okay, this helps! It seems to me that you do understand what people mean when they talk about justification in a general first-order sense (or at least what I mean when I do), because you were able to give coherent and meaningful responses to everyday first-order justificatory claims. So I guess I'm not sure what the issue is? I don't mean that in a snarky way at all - I'm legitimately unsure of what else it is you're asking. Like I said, I don't have a fully worked out theory for the exact conceptual boundaries of justification, how it works, etc - but ironically, now it seems like I'm the one saying that shouldn't matter for the related general first-order discourse! All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about. In an entirely first-order sense, do you disagree with that? Your response above doesn't really answer the question, because you've amended a bizarre belief you don't hold that alters the situation as described. I'm looking for answers like the ones you gave for the "leaving the restaurant" questions - the actual answers you'd give as an agent in the situation.
The issue is that I don't know what you mean by the term in your blog post. As I said, I think terms acquire meaning in their contexts of usage. While I can infer meaning in ordinary contexts of usage, you're not using the term "justification" in an ordinary context of usage. You're using it in some kind of philosophical context of usage. So I don't know what you mean.
I am very puzzled by all of this. Can you just tell me what you mean by “justification”? I’m not asking for a theory. I’m just trying to understand what you mean by the term.
>>All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about.
I don't think that's all your argument needs. People mean different things in different contexts when they say things are "unjustified" and it's an empirical question what anyone means in any given context. I think it's probably going to vary, and many even vary in ways that are mutually inconsistent.
>>All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about. In an entirely first-order sense, do you disagree with that?
Yes. I disagree with you. Some people would consider it justified and some wouldn’t. So if you’re asking a descriptive question about how other people think, lots of people are, I believe, going to think that some people’s subjective preferences are morally wrong. People moralize getting tattoos, sexual orientation, what music people listen to, and so on. They have historically, and they continue to do so today.
Also, I think if people’s subjective preferences had practical implications for others they’d be much more likely to not consider such actions unjustified. I do not think that people’s judgments about what is justified divide along objective/subjective grounds. I think they divide along what the person values vs. what the person doesn’t value. I think people’s subjective preferences and values are doing the work here, not their beliefs about what’s objectively right or wrong.
>>I'm looking for answers like the ones you gave for the "leaving the restaurant" questions - the actual answers you'd give as an agent in the situation.
I might say it’s unjustified. And most people might. But all this tells me is something about the kinds of words people are likely to use in English in these situations, and this would be conditioned on contingent facts about those people’s psychology. I don’t think any answer I gave is going to do the kind of philosophical work you put “justification” to in your post.
Paraphrasing the gist of the core of the argument as understood by me:
1. How we discuss and engage with claims in various domains diverge in ways that correlate with whether or not there must be exactly one correct position for claims made in that domain.
2. When there is exactly one correct position, the correlation occurs as a consequence of the nature of that characteristic.
3. The correlation holds true for the moral domain.
4. It is sensible to think that it holds true for the same reason.
I'd say there is a more exact correlation with how people perceive the severity of disagreement. That mostly comes down to how much of a difference there is between alternative positions, the stakes involved, and any factors that aggravate the emotions involved.
There are certainly scenarios where the three essential features mentioned in the article won't correlate with objective/subjective but do correlate with severe/not-severe disagreement.
But there is a lot of apparent overlap, for a couple of main reasons.
In a dispute between a proven position and a disproven position, it's hard to disagree by only a little bit, unless the stakes are low. So disputes about established objective facts are prone to being severe disagreements.
The second reason is that pretty much any subjective question that has high stakes will have some moral or morality-adjacent issue attached to it. If you exclude all of those, then you're mostly just left with the low-stakes stuff.
I stress severity in my response to BSB, but it isn't just about severity, I argue, but about scope and practical relevance more generally.
When it comes to taste preferences, my standards typically concern my own conduct. They have a personal scope: I want to eat food I like, and avoid food I dislike. This simply doesn't apply to other people's conduct.
When it comes to moral preferences, my standards concern everyone's conduct. They are universal in scope: I don't want anyone to do anything immoral. This clearly does apply to other people's conduct.
In the examples BSB provides, the cases of snack preferences and hobbies, the moment those actions interfered with our values or threatened to harm us, we would moralize them and care about them. As such, the practical relevance of other people's actions can account for when and why we'd be concerned with condemning those actions.
I don't think an implicit conception of moral concerns as objective and nonmoral concerns as subjective does much work in explaining any of this. On the contrary, I think accounting for our attitudes in terms of our subjective goals and values does. If I really cared about everyone loving gummy bears, I'd be outraged at someone hating them. That most readers would think this is insane and maybe "unjustified" can be accounted for by their shared subjective values; no need to posit a commitment to objectivity.
I think a lot of these discourse features can be explained as a consequence of the role that morality plays in human society. Because we are a social creature we need to agree a moral code in order the function as a group. If you think it's fine to graze your animals in the communal fields and I think it's not, it is absolutely required that we come to an agreement on the matter or otherwise we end up in conflict. Morality has this feature in common with empirical facts. If I believe you *did* graze your livestock on the field but you deny that, we are once again in conflict. But if we disagree about the tastiness of licorice, no such problems arise. That's why people don't nearly feel the same compulsion to impose their aesthetic judgements on others. But even there you sometimes hear discourse that seems to imply that there is an objectively best or worst movie. I reckon instincts like those stem from communities needing to agree on which crops to plant or where to set up camp.
I think this is a reasonable response for sure, but the problem for me is that the more complex our justifications get for the divide, the more likely it is that they'll begin to include plausibly objective features that make it hard for things like social cohesion or conflict resolution to clearly count as counterexamples. It's very easy for a realist like me to just take the opposite approach of maintaining the objective/subjective divide and seeing what you're bringing up as evidence for the objectivity of social cohesion's goodness or the badness of conflict.
I don't quite understand the move you're making in that last sentence. Can you rephrase it?
Sorry, that was a little unclear. All I mean is that, if you're a moral realist, then you almost certainly believe that a huge portion of the objective moral facts out there are facts that relate to human development and flourishing. So that means you'd think conflicts over grazing or setting up camp or otherwise engaging in the sorts of things that "keep society running" would be questions with objective answers: It's objectively true that Grog shouldn't graze his cattle on the shared lands too much, or that Lucy should do her fair share to set up camp, or whatever. And so if you accept the objective/subjective divide I laid out, and you're a moral realist, then you should believe that disputes with the ability to really impact human lives will be discussed in the way we're talking about, and that disputes without any such an impact will be talked about differently. In other words, while you're right that the divergence in how we engage in discourse could be explained as "things that impact people vs. things that don't impact people," a moral realist should just think that division is coextensive with "disputes over objective facts versus disputes over subjective facts," since impacting people automatically gets your involved moral facts that are objective. So I'm just saying that, even if the division you're giving in your original comment is a totally valid way of conceptualizing things, a moral realist can accept that division and just say it's fundamentally wrapped up with the other division I'm giving, rather than being a separate explanation that's actually better. Does that make sense?
Tell me if I'm interpenetrating correctly. Your contention seems to be that the fact that morality and descriptive fact share these pragmatic features is not a coincident. They both "bite back" because they are both objectively real, while aesthetic facts does not "bite back" and is subjective.
For example, if society became convinced that stealing is ethical it would fall apart. This counts as evidence that the badness of theft is objective. The logic is similar to the categorical imperative.
But I don't think that quite works. The commonalities between moral discourse and empirical discourse were more front and center in my original comment, because it were those that I meant to explain. But there are differences too. When people disagree about descriptive fact it's essential not only that we agree on a common reality, it is usually also essential that we agree on the one reality that is objectively correct. If your society agrees that a poisonous mushroom is in fact save to eat, you will run into problems.
But in the case of morality it is possible to build a successful society on any number of mutually exclusive moral frameworks. A large portion of societies in history have functioned and perpetuated themselves while keeping slaves in bondage. If there is an objective morality to bite back on moral falsehoods, you'd expect it to bite back on these societies.
I foresee the objection that many of the same societies that were wrong on slavery were also wrong about a lot of factual things, and reality did not bite back on them either. One example of that is geocentrism, which was near universally believed until the renaissance. But geocentrism concerns itself with the nature of outer space, which these societies never interacted with (apart from receiving sunlight, which they could model more or less correctly). While slavery is something deeply intertwined into their societies. A second example is medicine. Where I think false believes did bite back on them, but since nobody knew the correct answer, everyone was equally bitten. Once some societies figured out the truth, they rapidly experienced a lot of success relative to their peers.
Well, it's not really an argument about pragmatic consequences - it's true that a culture where stealing is condoned would fall apart, but that doesn't prove stealing is objectively wrong, because it could be that a society falling apart isn't objectively bad. Rather, my point is just this:
Start with the assumption that disputes regarding issues that meaningfully impact human beings are moral disputes, and the assumption that moral disputes relate to objective facts (i.e. the moral realist assumption). Then accept the subjective/objective distinction I've put forward to explain why our discursive practices diverge. Those three beliefs in tandem predict that we will use "objective-style discourse" when we engage on questions that meaningfully impact human beings, because those are disputes relating to objective facts. So therefore, the fact that we use these discursive practices to speak about things that meaningfully impact human beings isn't a challenge to the subjective/objective explanation.
As a rough analogy, imagine I explain why we let some people sit in the front row of a concert but not others by saying it depends on whether or not they bought a special ticket. And then someone else comes along and says "I don't think you need to rely on the ticket to explain things - I think it's just that some of them are really big fans of the musician and others aren't." That distinction between big fans and just regular fans might work as a solid heuristic for determining whether they'll sit in the front row, but that doesn't mean the "bought the special ticket or just a regular one" isn't the actual deciding factor (because the big fans buy the special ticket and the regular fans don't). Similarly, even if it's the case that we can map our discursive practices onto an "impacts people vs doesn't impact people" standard, that doesn't mean the "objective vs subjective" standard is wrong (because stuff that impacts people is automatically objective and vice versa).
But also, I would say reality did "bite back on" slave societies in the relevant way we'd expect it to - those societies really did have meaningfully lower levels of human flourishing, which is what the moral facts in question relate to. I don't think we should necessarily expect a transgression of those moral facts to have any other result.
>>The second preliminary point I’d like to make is that the firm line between practical ethics and metaethics that antirealists often take for granted isn’t actually obvious, or somehow guaranteed by more general facts about how different levels of discourse interact. In a purely a priori sense, we have absolutely no reason to think that radical changes to our second-order understanding of the terms and concepts involved in a particular discourse will have no impact whatsoever on whether our first-order discursive practices are justified [...]
On this point, I'd draw a distinction between whether our second-order views logically entail any particular consequences, and whether they tend, in practice, to result in changes in our first-order normative moral standards, as well as our sentiments, attitudes, dispositions, and so on. The former may be a more formal question about the conceptual relation between metaethics and normative ethics, while the latter may be more a question of human psychology that cannot readily be settled without empirical evidence.
Another feature of moral discourse that only seems grounded if morality is in fact objective is moral theorizing. People come up with theories of what makes something moral or immoral, or at least general rules of thumb, and decide what particular actions are right or wrong based on those theories. This is a regular feature of objective discourses - we make predictions about the results of an experiment based on our physical theories or about the outcome of an event based on rules of thumb we've learned from previous similar events. And it's obviously grounded in the objective nature of the discourse - if morality is objective, then there should be some description of what it is, and that description will likely have features like simplicity that theories of objectively existing things tend to have. But if morality is stance-dependent, then we don't have to worry about theoretical virtues - it can be whatever our stance says it is. Furthermore, we would never have any reason to base our judgement of a particular case on a general theory, just as you wouldn't change your stance on whether gummy bears taste good even if you had learned a general rule of thumb that implied that they should or shouldn't taste good.
Another feature is completeness - this goes nicely with consistency as mentioned in the article. When discussing moral matters, we assume that there is some fact of the matter about any particular case, even one we haven't thought about before. We assume, for example, that either it is wrong to lie in order to make someone happy (in a specific case that specifies all morally relevant details), or it is not wrong. But if morality is stance-dependent, it could be that our stance is simply silent on the matter. Our stance neither judges it to be wrong, nor for it to be non-wrong because it is an ambiguous stance. In fact, for almost all accounts of stance-dependent moral facts, this will be the case for most moral dilemmas. But then why debate moral dilemmas in the first place?
Yeah, one thing I regret is that I didn't do more to spell out exactly what I meant by "consistency," because it's basically an analogue to what you're saying here. It's hard for me to understand how antirealism can ground universal claims like "Torture is wrong" solely by reference to individual experiences of negative feelings towards torture - we expect that kind of consistency in situations where we're discussing objective properties, but there's nothing weird about the idea that you might disapprove of several instances of torture but semi-randomly approve of another.
As an error theory adherent of sorts (local error theory i.e. categoricity-only rejection) I'm not sure if our moral discourse needs justification in a reflective sense. That is, the whole fact that we have a moral discourse. But I also don't say that serious revisions to whatever implicit second-order beliefs people happen to hold would have no bearing on actual moral behavior. Seems like it could. An adherent to error theory could also advise a noble lie, or to let sleeping dogs lie, or have a whole other attempt at making moral behavior intelligible:
https://partiallyexaminedlife.com/2025/03/05/the-life-of-the-volunteer/
In any case, I don't perceive myself to need to believe anything like categoricity to operate in everyday life irt descriptive language. So, this part
"The essential features of moral discourse are best justified in other non-moral domains by reference to the objectivity of the facts involved, and this gives us good reason to believe that objectivity is what justifies those essential features in moral discourse as well."
doesn't strike me as decisive even if we grant it, because objectivity is not enough for any strong moral realism. You need categoricity too.
This is a good example of a well-made philosophical argument in that once you start agreeing with it you find it very difficult to stop (and I don't just mean that it's deductively valid). I would, of course, disagree with or make modifications to basically every premise and assertion in this article (lol) but I can feel why it would be compelling to someone who is more on the fence than I am. It kind of speaks to a belief in "mere" moral discourse that must either be wholly disregarded or inevitably accepted as indicative of realism. Also, thanks for mentioning error theorist fictionalist 🥂.
Error theory fictionalism*
Very interesting. I'm not terribly familiar with the debate between realism and non-realism, but I think I have a counter example of a domain that's not objective, where we engage in similar ways.
Take for example a workplace, where the boss sets the expectation that staff will follow a certain dress code, and complete certain tasks by certain times. The boss's expectations here are not objective facts, but something she wilfully sets. But she will still assign out praise and blame based on those expectations, still be expected by others for her expectations to be consistent, and in all likelihood she will be intransigent and largely unwilling to compromise on her expectations.
We might look at other instances of non-objective expectations/standards as well, such as standards of spelling, grammar, or good manners. We absolutely try to impose these expectations/standards on others, assigning praise and blame, expecting consistency, and being intransigent about them.
Yes, this is definitely a complicating factor that you're right to bring up. My take on it would be, briefly, that the assignation of praise and blame (and the presence of related features) is justified only to the degree that the subjective standards are themselves justified on the basis of objective second-order standards. So, for example, it might be justified on an objective basis to have *some* sort of dress code, because it prevents conflicts and improves the image of the business. And once that objective need has been established, then it would be derivatively justified to adopt whatever subjective standards are needed to fulfill that goal. Etiquette sorta works like this too - the behaviors it polices are subjective, in the sense that there's nothing objectively bad about picking your nose at dinner or something, but *that there be some set of social standards that are enforced* could still be objectively justified in a way that makes the specific subjective standards we pick justified too.
I think that would work, but is it implausible that they might just be unjustified and not really need justification? The boss might make up rules just because, and while they're unjustified, she has the power and can wield it how she likes.
Perhaps morality likewise is nothing more than us setting expectations and standards on each other, and the only grounding for it is our will/desires. The justification would then be just a question of whether we can convince (or coerce) others to uphold the standards.
The oddity with the liquorice case is perhaps just, why would anyone care what sweets others like? But then, people do, strangely, care about such things, like opposing pineapple on pizza, or worrying about others' sex lives. It seems your argument requires not condemning pineapple on pizza!
I just wrote an absurdly long response to this, but I was immensely and especially puzzled by this remark:
>>Antirealists might also be tempted to adopt a more radical approach that denies the need to justify these features of discourse entirely. They can just say these sorts of discursive practices are appropriate in one domain but not another, and that’s just the way it is. But a move like this imposes a major theoretical cost, in that it reduces all questions of justification to mere social convention or else renders the entire project incomprehensible.
Why should I think there's anything "radical" about this? What does that mean? It never even occurred to me consider whether the features of moral discourse were "justified." What does that that mean, too? Justified according to what standard? I'm not even sure it makes sense to say they're "appropriate in one domain but not another."
People engage in moral discourse. Why would I need to think doing so was "justified," and that I have to explain why it's justified? And why would I think I'd be paying any sort of theoretical cost if I didn't think it was justified? What cost would that be?
I don't even know what notion of "justification" you're working with. And perhaps the projects you are engaged in simply are incomprehensible. I certainly think a lot of analytic philosophy is pointless or nonsensical. I don't think I'd be paying any sort of cost if I thought the same of a bit more of what philosophers do.
What I suppose might help is if you say a bit more about what you think justification is, why you think moral discourse is justified, and why someone would think they'd be paying some kind of cost if they denied this.
Honestly, I'm not exactly sure how to answer this question. If you're asking me for a complete theory of normative justification, then I wouldn't say I have one. But if you're just asking what I mean when I make claims about certain beliefs, attitudes, actions, etc being justified in the context of first-order discourse, do we really disagree on that? Using the example in the piece, would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears? As I said in the piece, I'm not attaching any particular normative framework or analysis at the start. But would you just, in daily life, say that was unjustified or no?
I don't need a complete theory. I simply don't know what you have in mind by "justification." Like at all.
>>But if you're just asking what I mean when I make claims about certain beliefs, attitudes, actions, etc being justified in the context of first-order discourse, do we really disagree on that?
I think we probably do, yes.
>>Using the example in the piece, would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears?
I don't know, because I don't know what you're asking me. I think it'd be mean for someone to do that, and I wouldn't like if they did it. I might even think they were insane. But what do you have in mind when you suggest they would be "unjustified" in doing this?
>>But would you just, in daily life, say that was unjustified or no?
Probably not. If I did, I'd mean something extremely reductive like that it was rude or I didn't approve of them doing it. Is that what you mean by justification? I imagine not.
Are you saying, then, that you never think of actions or beliefs you encounter in your own life as justified or unjustified? If you were in a meeting at work and made a suggestion, and your boss said "I don't think that would be justified right now," you wouldn't know what they were trying to say or what you would need to do in order to dispute their judgment? Or if I asked you what would justify leaving a restaurant without paying, you wouldn't be able to conceive of a coherent answer?
I certainly don't think explicitly in terms of them being "justified" or "unjustified." I really mean it when I say I don't know what you mean by that. I use the *term* "justified," but this is just another way of saying something about my expectations or attitudes in most contexts. It might mean something different in some other context. I don't think there's some shared, underlying notion of "justification" that drives how I speak or think (nor do I think this is the case for pretty much anyone else, except maybe philosophers).
>>If you were in a meeting at work and made a suggestion, and your boss said "I don't think that would be justified right now," you wouldn't know what they were trying to say or what you would need to do in order to dispute their judgment?
I'd infer their meaning from the context. And what people mean by the term "justified" would vary from one context to another. I don't have enough information about the context here to have a sharper sense of what my boss meant, but I can imagine a context where this would make perfect sense. For instance, suppose the company is having serious financial problems and I suggest we file for bankruptcy. If they said they didn't think that'd be justified, I'd interpret them to mean something like "that's a drastic action, and our finances aren't so bad that I'd consider that the best course of option just yet." The use of "just now" would imply something like that they're open to it being an option in the future, though.
My boss could have just as easily have phrased the remark without any reference to "justification." I don't think this kind of remark calls for any distinctive conception of "justification" to do the relevant work it was doing in some real world context.
So when you start talking about justification in a philosophical context, this is some other context where the relevant cues I think I and nonphilosophers typically use to infer someone's meaning are no longer present. In other words, whatever it is you are doing with the term "justification" when you write about philosophy, I do not take you to mean the same thing as what I take my boss to mean. I've written about this quite a bit: I think what people mean by the words and phrases they use only emerges in the relevant contexts of usage. While I am fully capable of inferring the meaning of a term in a non-philosophical context, I don't grant that we can transpose meaning in those contexts over to philosophical contexts. I think philosophers thus have a responsibility to explain what they mean in the relevant (philosophical) contexts of usage.
So if your question is intended to suggest I struggle to understand everyday use of the term "justification" No. I don't. But I am explicitly denying that such usages necessary match your usage outside of those contexts.
The same goes for this remark:
>> Or if I asked you what would justify leaving a restaurant without paying, you wouldn't be able to conceive of a coherent answer?
It would be easy to give a coherent answer. Here are some possible "justifications": I told them I have a food allergy and they served food loaded in whatever it is I am allergic to, I couldn't eat it, I got sick, and they insisted I pay. Or the waiter physically assaulted me. Or the restaurant is on fire. Or if I own the restaurant and don't have to pay. And so on.
Note that if you think these are perfectly fine justifications, I can revisit the gummy bear situation:
>> would you *not* affirm the claim that it's unjustified to (sincerely) scream at someone and call them a depraved piece of garbage because they prefer licorice over gummy bears?
If I thought that people who prefer licorice over gummy bears were evil and should not be accepted in polite society, I'd not only think it was justified to scream at them, I might think it was mandatory to do so. Screaming at them might not even be going far enough. It might be mandatory to consign the person to life in prison. And I don't think I'd be making any errors in doing so, and I don't think anyone could provide a single decent reason for me to think someone who felt this way was making any kind of error (and I would very strongly object to any suggestion that such a stance involves "admitting defeat".)
FWIW: I think you and I might have very different views about language and meaning, and that his may influence our downstream positions on these matters.
Okay, this helps! It seems to me that you do understand what people mean when they talk about justification in a general first-order sense (or at least what I mean when I do), because you were able to give coherent and meaningful responses to everyday first-order justificatory claims. So I guess I'm not sure what the issue is? I don't mean that in a snarky way at all - I'm legitimately unsure of what else it is you're asking. Like I said, I don't have a fully worked out theory for the exact conceptual boundaries of justification, how it works, etc - but ironically, now it seems like I'm the one saying that shouldn't matter for the related general first-order discourse! All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about. In an entirely first-order sense, do you disagree with that? Your response above doesn't really answer the question, because you've amended a bizarre belief you don't hold that alters the situation as described. I'm looking for answers like the ones you gave for the "leaving the restaurant" questions - the actual answers you'd give as an agent in the situation.
>>So I guess I'm not sure what the issue is?
The issue is that I don't know what you mean by the term in your blog post. As I said, I think terms acquire meaning in their contexts of usage. While I can infer meaning in ordinary contexts of usage, you're not using the term "justification" in an ordinary context of usage. You're using it in some kind of philosophical context of usage. So I don't know what you mean.
I am very puzzled by all of this. Can you just tell me what you mean by “justification”? I’m not asking for a theory. I’m just trying to understand what you mean by the term.
>>All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about.
I don't think that's all your argument needs. People mean different things in different contexts when they say things are "unjustified" and it's an empirical question what anyone means in any given context. I think it's probably going to vary, and many even vary in ways that are mutually inconsistent.
>>All my argument needs to get going is just the basic fact that, inside our general first-order discourse, it's considered unjustified to react to entirely subjective disputes with the sorts of behaviors I'm talking about. In an entirely first-order sense, do you disagree with that?
Yes. I disagree with you. Some people would consider it justified and some wouldn’t. So if you’re asking a descriptive question about how other people think, lots of people are, I believe, going to think that some people’s subjective preferences are morally wrong. People moralize getting tattoos, sexual orientation, what music people listen to, and so on. They have historically, and they continue to do so today.
Also, I think if people’s subjective preferences had practical implications for others they’d be much more likely to not consider such actions unjustified. I do not think that people’s judgments about what is justified divide along objective/subjective grounds. I think they divide along what the person values vs. what the person doesn’t value. I think people’s subjective preferences and values are doing the work here, not their beliefs about what’s objectively right or wrong.
>>I'm looking for answers like the ones you gave for the "leaving the restaurant" questions - the actual answers you'd give as an agent in the situation.
I might say it’s unjustified. And most people might. But all this tells me is something about the kinds of words people are likely to use in English in these situations, and this would be conditioned on contingent facts about those people’s psychology. I don’t think any answer I gave is going to do the kind of philosophical work you put “justification” to in your post.
Paraphrasing the gist of the core of the argument as understood by me:
1. How we discuss and engage with claims in various domains diverge in ways that correlate with whether or not there must be exactly one correct position for claims made in that domain.
2. When there is exactly one correct position, the correlation occurs as a consequence of the nature of that characteristic.
3. The correlation holds true for the moral domain.
4. It is sensible to think that it holds true for the same reason.
I'd say there is a more exact correlation with how people perceive the severity of disagreement. That mostly comes down to how much of a difference there is between alternative positions, the stakes involved, and any factors that aggravate the emotions involved.
There are certainly scenarios where the three essential features mentioned in the article won't correlate with objective/subjective but do correlate with severe/not-severe disagreement.
But there is a lot of apparent overlap, for a couple of main reasons.
In a dispute between a proven position and a disproven position, it's hard to disagree by only a little bit, unless the stakes are low. So disputes about established objective facts are prone to being severe disagreements.
The second reason is that pretty much any subjective question that has high stakes will have some moral or morality-adjacent issue attached to it. If you exclude all of those, then you're mostly just left with the low-stakes stuff.
What's an example of a completely subjective question that has high stakes in the non-moral realm? I have to admit I can't really think of one.
I stress severity in my response to BSB, but it isn't just about severity, I argue, but about scope and practical relevance more generally.
When it comes to taste preferences, my standards typically concern my own conduct. They have a personal scope: I want to eat food I like, and avoid food I dislike. This simply doesn't apply to other people's conduct.
When it comes to moral preferences, my standards concern everyone's conduct. They are universal in scope: I don't want anyone to do anything immoral. This clearly does apply to other people's conduct.
In the examples BSB provides, the cases of snack preferences and hobbies, the moment those actions interfered with our values or threatened to harm us, we would moralize them and care about them. As such, the practical relevance of other people's actions can account for when and why we'd be concerned with condemning those actions.
I don't think an implicit conception of moral concerns as objective and nonmoral concerns as subjective does much work in explaining any of this. On the contrary, I think accounting for our attitudes in terms of our subjective goals and values does. If I really cared about everyone loving gummy bears, I'd be outraged at someone hating them. That most readers would think this is insane and maybe "unjustified" can be accounted for by their shared subjective values; no need to posit a commitment to objectivity.