Enjoying reading this, only time to skim. On point II Lance and I have talked about this before and we both believe that the problem with Frege Geach embeddings is the theory of language and truth that it depends on. Your logical space fixes in place views of language and truth in a way we would dent. The first claim is that the use of "true " is supposed to apply to propositions which are truth apt sentences, the second claim is that of compositional semantics where the constituent parts of propositions have little atomic meanings that can be recombined in certain sentence structures, the third is that the meaning of a word is an internal, phenomenally private representation which "lines up" when a word is correctly used and doesn't when it doesn't -- this is in opposition to both Lance and my own views of meaning as use.
So, not that either of us do exactly believe this, but this makes it possible that moral utterances are (say) expressions of Boo/Hurrah, and that we go on to say our moral assertions are true ( which we would explain as being caused by our Boo/Hurrah emotion ). This means that when placed in grammarical constructs that have inferential form we can apply a logical calculus and there's no problem. The problem is introduced by the view of language/semantics.
If this hasnt helped Im happy to type a longer reply off my phone when I have time. Fwiw am also not an emotivist.
Thank you for the comments! If I understand you correctly, you're not objecting the way of dividing up metaethical positions, but just objecting to the Frege-Geach problem?--please correct me if I've misunderstood.
Anyways, if that's right then I think I pretty much agree with what you say, in the sense that if you hold the views of language you describe, then there shouldn't be any problem there. In fact, I suspect that the Frege-Geach problem is pretty surmountable in any case (though I've not thought too much about it).
That is pretty much correct. It wasnt the point I was that bothered about -- like you, I think taxonomising isnt really the interesting part. I do believe your taxonomy holds for views that depend on these accounts of language (which is fine for most contemporary analytic philosophy). If we were to introduce those claims into our modelling we would get different accounts of truth which would introduce more possible positions. Though I believe all the coherent positions that reject this kind of propositional and compositional semantics would be "ant-realist" positions broadly.
I basically agree with Nathan here. And I already did before considering of your post. The issue is that the categories you present rely on background assumptions I don't grant.
So while it'd be fine to say something like: "If you accept presuppositions {X, Y, Z,...} then there are exactly and only three antirealist positions." The problem is that I don't accept all of the presuppositions.
So your whole account never goes through for me. It relies on assumptions I don't grant. The logical space of positions to take in metaethics isn't exhausted by an exhaustive list of possibilities that are conditioned on some foundational set of claims because philosophers are free to reject those foundational claims!
> I also suspect he might object that there isn’t any well-defined class of “moral judgements,” as it’s just a subset of ordinary discourse, with no clear boundary.
This seems like more of an argument against moral anti-realism than a counter to arguments against it. If there's no clear boundary between moral discourse and non-moral discourse, how can one say that all moral discourse in not stance-independently true, unless one also says the same about non-moral discourse, or at least a large segment of it?
An easy way to see whether this works is tell me where my position falls within the description you provided. I don't think it does, which shows both that the claim that there's only three positions is wrong and that the process of elimination approach won't work.
Well, I take that you think it's often indeterminate what people mean with moral statements (correct me if I'm wrong). Now, there are several ways of attempting to classify this, depending on exactly how this is spelled out, I think.
If indeterminate uses of moral statements are not truth-apt, and if all uses of moral statements are indeterminate (or if some are, and the rest determinately express some non-truth apt attitude) then your view would count as non-cognitivism as defined here.
If indeterminate uses of moral statements are not truth-apt, but people sometimes use moral statements to express truth-apt moral judgements, all of which are false, then your view would count as error-theory as defined here.
Neither of these options require that moral judgements always determinately express some attitude, and that this attitude is sometimes a non-truth-apt one. Rather it just requires that when it is indeterminate what people express, that is not truth-apt.
There is perhaps also the option that what people express is neither truth-apt nor not truth-apt. In that case, then your positions really doesn't fit in the framework I sketch. But as I mention, that needn't matter in the context of the elimination argument (though I admit it *does* matter if we're simply talking about what positions an anti-realist might take outside of this context--something I maybe didn't make quite clear enough) for two reasons:
1) Steps in the process of elimination argument also rule out this view, given they work, even if it was not explicitly mentioned at the start (namely the error-theory and relativism steps). I don't claim this step works--in fact I think it'll be very minimally convincing to someone who's already an anti-realist. Especially if, like you, one doesn't feel any realist pull from the examples, even if one is very strongly opposed to them.
2) If there still are some cases where what people mean with moral statements is determinate, then we can just stipulate the moral domain to only include those statements. From there the argument can proceed as follows (though the realism that's the goal might not exactly be what the proponent was looking for, if the moral domain ends up being much smaller than initially expected).
Though I don't want to assume things on your behalf, so please correct me if I'm making some mistakes here! :)
You can always redefine categories to include anything you want. If someone says there are only three kinds of fruit: apples, bananas, and oranges, and I say no there are other kinds of fruit and I give the list, someone could just redefine all of those other fruits to count as either an apple, a banana, or an orange. Nothing of substance is accomplished by doing this. It's just verbal gerrymandering.
I assume we both agree that there is no "true" way of dividing up the categories. In that case how we should divide them seems more to be a question of what's useful or fruitful (no pun intended). The reason we shouldn't divide fruits as you suggest is that it's a supremely confusing and useless way of classifying fruits--at least in many contexts.
What about metaethical positions? In some cases it might be more useful to use alternatives to the classification I have given, as it smooths over some distinctions that we might want to draw, depending on the context. But for the purposes of this argument, the divisions I propose are perfectly serviceable. Not only do they match the common usage of the labels reasonably closely (I think), but they are also very clearly defined so that there should be no confusions even on the points where they don't. And as I have argued, if the arguments made in the process of elimination work, they work against all positions within these categories, successfully eliminating all alternatives to realism. That is surely a strong reason to use these classifications when discussing this argument.
I agree that nothing of substance is gained or lost by how you carve out the categories--but the same is the case for carving up the categories in an alternative way to what I have done. If the argument can be run with the categories I have given, then it can be run with the categories I have given, and there is no point in drawing the categories a different way. This is not to say that the argument *does* work, but it *is* to say that the framing of the argument is not what is at issue if it doesn't.
My point is that my initial claim was that the traditional three positions did not include my position. This is true. This is not refuted by altering the initial three categories to try to include my position and then say that there were three positions after all. This misleadingly implies that my initial claim was false.
Yeah, I think we pretty much agree then. I was trying to argue simply within the context of the elimination argument, where I think this division does make a lot of sense. But if we're just talking how the labels are generally used, then they are not typically mutually exclusive.
I will also add that I think the people making this argument are often not explicit enough in how they define the positions. For example people often say that error theory claims that "moral judgements are truth apt and false," but that leaves the scope underspecified: Are *all* moral judgements truth apt, in which case there is plenty of room between non-cognitivism and this, or is it *some*?
Robert Hall in his post was quite explicit on this point, and in the conversation between Huemer and Loeb, the division I propose was also explicitly made (though by Joe Schmidt, so Huemer doesn't really get the credit there). But in many cases proponents aren't as careful, and then it merits being pointed out (though it doesn't actually affect the strength of the argument, but just means that the proponent has been sloppy in spelling it out).
If someone -- in the context of a process of elimination argument -- says something like _There are exactly three positions other than mine_, he is presumably not making an empirical claim about what the standard positions are.
Instead, he might be saying something like _There are three dialectically relevant logical possibilities other than the logical possibility that I endorse_ -- in other words, he is implying that he has divided up the positions such that he has objections to those three positions, and that the objections will cover the denial of his own view, either because those positions are logically exhaustive, or the three positions plus their combinations are logically exhaustive. In the case of Michael Huemer, I think that he does not define the positions to be exclusive of each other, and acknowledges that one could have a combination view -- but his objections apply to combination views as well, so this doesn't matter for the argument. In my case, and the case of Silas Abrahamsen's essay, we divided them up in a way to be exclusive of each other and logically exhaustive.
(Actually, what Bentham's Bulldog said was 'Given that, as previously discussed, moral realism is the view that there are true moral statements, that are true independently of people’s beliefs about them, there are three ways to deny it'. I don't think that he was making an empirical claim about the standard positions.)
Of course, if all you have ever done is made the empirical claim that the positions, as most commonly defined, don't cover yours, then you may well be right. But if you are objecting to the division into three anti-realist positions in the context of someone's presentation of an argument for moral realism, to show that you are correct, you would need to show that the stipulated positions (or the positions and their combinations) are not logically exhaustive, or that an objection fails to address the relevant position.
//Instead, he might be saying something like _There are three dialectically relevant logical possibilities other than the logical possibility that I endorse_ -- in other words, he is implying that he has divided up the positions such that he has objections to those three positions, and that the objections will cover the denial of his own view, either because those positions are logically exhaustive, or the three positions plus their combinations are logically exhaustive.//
Huemer's objections to the standard positions do not, as far as I can tell, include any objections to my views.
//In the case of Michael Huemer, I think that he does not define the positions to be exclusive of each other, and acknowledges that one could have a combination view -- but his objections apply to combination views as well, so this doesn't matter for the argument.//
My position is not a combination of the three traditional views *and* it's not clear Huemer's view does apply to combination views, since combination views may rely on different stances towards language/semantics. That concern is moot though, since I don't endorse pluralism and have little interest in defending it.
//In my case, and the case of Silas Abrahamsen's essay, we divided them up in a way to be exclusive of each other and logically exhaustive.//
If Silas's three categories include my view then they aren't the three categories I was saying aren't the only three. Anyone can arbitrarily make anything fit into three categories, or even just two. For instance, all objects are either potatoes or non-potatoes. Number of categories is often arbitrary or a matter of utility or convention for some specific purpose. I am not concerned with the number of categories, but their contents.
According to Huemer, there are three antirealist positions: noncognitivism, error theory, subjectivism. I'm an antirealist and I don't endorse any of those positions. My position is not logically impossible and is not reasonably included within any of the those three positions. Neither is incoherentism and possibly pluralism isn't either. So there are two, maybe three positions excluded by Huemer's claim, and possibly more. So Huemer is mistaken.
...This is not that complicated. What I said was true, is true, and will remain true, and neither you nor anyone else has said anything that could be reasonably interpreted as a refutation of my initial claim.
Creating a new set of three categories, gerrymandered by design to be mutually exhaustive, does not refute my original claim.
If, for instance, someone said that there are only three fruits: apples, bananas, and oranges, and I say "no there aren't, behold the watermelon." This could not be refuted by saying:
"Aha, but suppose I do this: Apples, bananas, and the orange/watermelon category! Now there are still three categories!"
The issue was never about how many categories there were. It was about the specific presuppositions and strictures underlying the three categories Huemer insisted were the only three options. Since all three options rely on e.g., specific presuppositions about semantics that I reject, my position definitely does not fall within those three categories.
Of course if you create some new set of three categories you can force-fit my position into it. This is a profoundly uninteresting and pointless exercise.
I don't think this is that great of a test because you can always just be vague enough about your position that it's not clear which category it falls into or hold a position that is agnostic as to which category is true. This wouldn't prove that the categories are non-exhaustive, but it would still be impossible to sort your position into one of them. Similarly, you could always hold a position that clearly seems to fall into one of the categories, but deny that it does.
What do you object to about the revised classification scheme? Given that it's logically exhaustive, you can't just say that it ignores certain options. I guess you could say that the words used to define the categories (truth-apt, true, stance-independent) are themselves ill-defined, but in that case moral anti-realism itself becomes an ill-defined position.
One thing that frustrates me is that philosophers don't take the time to explain to people who aren't familiar with a philosophical question why it matters. What are the stakes of the moral realism question? How do moral realists and anti-realists see the world and act differently? I've seen people debating moral realism on here for months and have just as little understanding of the stakes as when I started. It all seems more like a game than an issue that I should take seriously.
That's a great article! So why do people go on talking about moral realism and anti-realism then? Is it bullshit in the Frankfurt sense that no one in the debate cares whether what they're saying is true or not? Maybe they're just seeking some kind of approval or something.
Thank you! I can't say what everyone's motivation is. I certainly don't think they're spewing Frankfurtian bullshit--I assume people care about saying what they believe is true in the domain. Personally I just think it's an intrinsically interesting subject, which also touches on other subjects that I find interesting (I suspect this is the case for many people talking about it). As I mentioned at the end of the aforementioned article, it also has at least some relevance for theism/atheism (something that itself might have some practical relevance).
Thanks! I'm currently studying for a logic exam, though, so I thought I'd use it as an excuse to tell myself I'm being productive. But I appreciate the consideration:)
I appreciate posts like these because it creates opportunities for those that don't really fit into the traditional analytic discourse to point out that there are many other ways to approach language and intersubjectivity. Your exhaustive list of positions all assume something like correspondence between thoughts/beliefs/desires and their representations in language. For various forms of quietism, pragmatism, or incoherentism, there is no correspondence. I fall somewhere in this camp. Given my understanding of language and subjectivity, when we talk about ethics with one another, we are engaged in a purely verbal language game, united by a family of resemblances. Furthermore, it's merely practical that we describe our aspirations and ethical desires in terms of empirical or synthetic a priori descriptions of reality, because we also play those language games all the time it usually gets the job done! It's like going to play chess, realizing you can't find the chess pieces, so you borrow checkers pieces. But then you play chess-with-checkers-pieces long enough with the same people, until it doesn't feel incoherent, given a particular context, to incorporate some of the functions of checkers into the chess game (like denoting pawn promotion with a stack of checkers pieces, and keeping in the function of double-jumping). Imagine if Stockfish (the best chess computer) were to analyze a game between two players that used checkers pieces and incorporated double-jumping for stacked pieces, but only when the pawn had been promoted. I imagine Stockfish would simply say this action is not valid or meaningless. Well, a lot of analytic philosophy is like that to me. Analytic traditions try to delineate moral utterances from non-moral utterances in terms of the proper rules (syntax, rules of validity) and pieces (atomized concepts); and it borrows rules and pieces from other language games, such Kantian analyses of synthetic a priori knowledge. But this is a superficial distraction from the activity of morality, which connects desires/aspirations with actions within a particular situation. The purpose of philosophy, to me, is to help us pay attention how we are using and misusing philosophical language when attempting to represent or justify moral activity. In other words, there's no metaethics outside of the attempt within analytic philosophy to make moral utterances do stuff it was never meant to do.
I endorse this response. As a pragmatist and quietist I simply reject the foundational assumptions within which contemporary analytic metaethics is framed. I'm not stuck with analytic distinctions.
Every moral utterance is driven by cares & concerns, and most of them appeal in some way to facts that are ostensibly independent of any cares / concerns, yielding one of various forms of strategically-suggestive blurtings, e.g., "It's bad to do X" "Do Y!" "Z is permitted" "X is obligatory" "Booo Y!" "Z is forbidden." This in turn becomes a "shitmess" due to this wide variety, the utter slop, and the chronic omission and/or mixing-up of the cares/concerns and the strategic facts. Attempts to shoehorn this into tight fourfold taxonomies are just drawing leylines through the wild scatter.
"Also we should generally be weary of disjunctive arguments like these, as you can give plausible-sounding objections to almost all philosophical positions."
Such dychotomies as "objective/subjective" or "real/unreal" or "stance-independent/stance-dependent" are very unhelpful. Artifacts of ancient philosophy which didn't understand map-territory relations, treated mind and matter as separate magisteriums and therefore was tremendously confused about basically everything, unable to separate baby from the bathwater.
From which "stance-independent/stance-dependent" is probably the worst offender. With subjective/objective it's possible to conceptualize that things that are subjective are also objective. With real/unreal it's harder, but you can come up with some notion of "being real but in a different way that was initially implied and therefore, unreal in the initial sense". But the notion of dependency/independency... it's either incredibly strict or completely meaningless.
Suppose I define morality as a certain generalization of human values as developped by evolution through natural selection. Is it stance dependent? Well yes, I'm literally using moral stances of people throughout history as an input before I use it to train statistical classificator. On the other hand I'm pointing to something in the real world, not just my interpretation of it. And after my classifier is already trained it doesn't matter how much we modify the training data. So if alien ray modifies the stances of all people currently living on Earth to think that torturing babies is good, it wouldn't affect the answer of my classifier, so the standard argument against relativism doesn't work. But also there are multiple different ways to generalize human values from the data.
In the end the framework of stance-dependancy misses the actually interesting meta-ethical questions
I would put the four limbs slightly differently and more generically:
1. All claims about X are incapable of being truth-apt.
2. Claims about X are conceptually capable of being truth-apt but in all our contexts none ever are.
3. Claims about X can be truth-apt for us, but only ever where the evaluation of those claims depends upon some stance(s) (being stance(s) not intrinsic to X or the surrounding context potentially available to us).
4. Claims about X can be truth-apt for us without needing to employ any non-intrinsic stances.
If in some context 'X' has a consistent connotation, then those four possibilities about the truth-apt-ness of X must be mutually exclusive within that context. (One must be correct, though people's positions about which is correct need not fit neatly into any one box.)
That's one concern... whether/which one of the 4 limbs holds true with respect to some notional set of all possible claims about X.
The other concern is the characterisation of what people are doing when they make claims about X. As in, if someone says "atrocity always bad", the question of whether the person is or isn't intending to express it as a morally real truth and, if you say "not", whether you can then only deny it on the basis of the first 3 limbs.
The answer to the latter being no, you are not bound by those 3 limbs.
E.g. if someone says "it's ok to steal from the extremely rich to feed the starving" it might be that they (1) are expressing their approval of that, (2) consider morality to be inapplicable (neither moral nor immoral), (3) consider that it is a true statement according to them but not necessarily for all value-systems, (4) consider that it is a true statement according to some universal true moral system, (5) making a claim without considering its truth, (6) making a claim without considering whether it is universally true, (7) making a claim while holding a contradictory position, (8) misspeaking, (9) some combination of those things... plus probably some other things that I haven't included. And only (4) involves the adoption of a realist position.
It is worth noting that for a claim to be 'neither truth-apt nor not truth-apt' would entail a logical contradiction, so that isn't a logically possible position.
Enjoying reading this, only time to skim. On point II Lance and I have talked about this before and we both believe that the problem with Frege Geach embeddings is the theory of language and truth that it depends on. Your logical space fixes in place views of language and truth in a way we would dent. The first claim is that the use of "true " is supposed to apply to propositions which are truth apt sentences, the second claim is that of compositional semantics where the constituent parts of propositions have little atomic meanings that can be recombined in certain sentence structures, the third is that the meaning of a word is an internal, phenomenally private representation which "lines up" when a word is correctly used and doesn't when it doesn't -- this is in opposition to both Lance and my own views of meaning as use.
So, not that either of us do exactly believe this, but this makes it possible that moral utterances are (say) expressions of Boo/Hurrah, and that we go on to say our moral assertions are true ( which we would explain as being caused by our Boo/Hurrah emotion ). This means that when placed in grammarical constructs that have inferential form we can apply a logical calculus and there's no problem. The problem is introduced by the view of language/semantics.
If this hasnt helped Im happy to type a longer reply off my phone when I have time. Fwiw am also not an emotivist.
Thank you for the comments! If I understand you correctly, you're not objecting the way of dividing up metaethical positions, but just objecting to the Frege-Geach problem?--please correct me if I've misunderstood.
Anyways, if that's right then I think I pretty much agree with what you say, in the sense that if you hold the views of language you describe, then there shouldn't be any problem there. In fact, I suspect that the Frege-Geach problem is pretty surmountable in any case (though I've not thought too much about it).
That is pretty much correct. It wasnt the point I was that bothered about -- like you, I think taxonomising isnt really the interesting part. I do believe your taxonomy holds for views that depend on these accounts of language (which is fine for most contemporary analytic philosophy). If we were to introduce those claims into our modelling we would get different accounts of truth which would introduce more possible positions. Though I believe all the coherent positions that reject this kind of propositional and compositional semantics would be "ant-realist" positions broadly.
I basically agree with Nathan here. And I already did before considering of your post. The issue is that the categories you present rely on background assumptions I don't grant.
So while it'd be fine to say something like: "If you accept presuppositions {X, Y, Z,...} then there are exactly and only three antirealist positions." The problem is that I don't accept all of the presuppositions.
So your whole account never goes through for me. It relies on assumptions I don't grant. The logical space of positions to take in metaethics isn't exhausted by an exhaustive list of possibilities that are conditioned on some foundational set of claims because philosophers are free to reject those foundational claims!
> I also suspect he might object that there isn’t any well-defined class of “moral judgements,” as it’s just a subset of ordinary discourse, with no clear boundary.
This seems like more of an argument against moral anti-realism than a counter to arguments against it. If there's no clear boundary between moral discourse and non-moral discourse, how can one say that all moral discourse in not stance-independently true, unless one also says the same about non-moral discourse, or at least a large segment of it?
An easy way to see whether this works is tell me where my position falls within the description you provided. I don't think it does, which shows both that the claim that there's only three positions is wrong and that the process of elimination approach won't work.
Well, I take that you think it's often indeterminate what people mean with moral statements (correct me if I'm wrong). Now, there are several ways of attempting to classify this, depending on exactly how this is spelled out, I think.
If indeterminate uses of moral statements are not truth-apt, and if all uses of moral statements are indeterminate (or if some are, and the rest determinately express some non-truth apt attitude) then your view would count as non-cognitivism as defined here.
If indeterminate uses of moral statements are not truth-apt, but people sometimes use moral statements to express truth-apt moral judgements, all of which are false, then your view would count as error-theory as defined here.
Neither of these options require that moral judgements always determinately express some attitude, and that this attitude is sometimes a non-truth-apt one. Rather it just requires that when it is indeterminate what people express, that is not truth-apt.
There is perhaps also the option that what people express is neither truth-apt nor not truth-apt. In that case, then your positions really doesn't fit in the framework I sketch. But as I mention, that needn't matter in the context of the elimination argument (though I admit it *does* matter if we're simply talking about what positions an anti-realist might take outside of this context--something I maybe didn't make quite clear enough) for two reasons:
1) Steps in the process of elimination argument also rule out this view, given they work, even if it was not explicitly mentioned at the start (namely the error-theory and relativism steps). I don't claim this step works--in fact I think it'll be very minimally convincing to someone who's already an anti-realist. Especially if, like you, one doesn't feel any realist pull from the examples, even if one is very strongly opposed to them.
2) If there still are some cases where what people mean with moral statements is determinate, then we can just stipulate the moral domain to only include those statements. From there the argument can proceed as follows (though the realism that's the goal might not exactly be what the proponent was looking for, if the moral domain ends up being much smaller than initially expected).
Though I don't want to assume things on your behalf, so please correct me if I'm making some mistakes here! :)
You can always redefine categories to include anything you want. If someone says there are only three kinds of fruit: apples, bananas, and oranges, and I say no there are other kinds of fruit and I give the list, someone could just redefine all of those other fruits to count as either an apple, a banana, or an orange. Nothing of substance is accomplished by doing this. It's just verbal gerrymandering.
I assume we both agree that there is no "true" way of dividing up the categories. In that case how we should divide them seems more to be a question of what's useful or fruitful (no pun intended). The reason we shouldn't divide fruits as you suggest is that it's a supremely confusing and useless way of classifying fruits--at least in many contexts.
What about metaethical positions? In some cases it might be more useful to use alternatives to the classification I have given, as it smooths over some distinctions that we might want to draw, depending on the context. But for the purposes of this argument, the divisions I propose are perfectly serviceable. Not only do they match the common usage of the labels reasonably closely (I think), but they are also very clearly defined so that there should be no confusions even on the points where they don't. And as I have argued, if the arguments made in the process of elimination work, they work against all positions within these categories, successfully eliminating all alternatives to realism. That is surely a strong reason to use these classifications when discussing this argument.
I agree that nothing of substance is gained or lost by how you carve out the categories--but the same is the case for carving up the categories in an alternative way to what I have done. If the argument can be run with the categories I have given, then it can be run with the categories I have given, and there is no point in drawing the categories a different way. This is not to say that the argument *does* work, but it *is* to say that the framing of the argument is not what is at issue if it doesn't.
My point is that my initial claim was that the traditional three positions did not include my position. This is true. This is not refuted by altering the initial three categories to try to include my position and then say that there were three positions after all. This misleadingly implies that my initial claim was false.
Yeah, I think we pretty much agree then. I was trying to argue simply within the context of the elimination argument, where I think this division does make a lot of sense. But if we're just talking how the labels are generally used, then they are not typically mutually exclusive.
I will also add that I think the people making this argument are often not explicit enough in how they define the positions. For example people often say that error theory claims that "moral judgements are truth apt and false," but that leaves the scope underspecified: Are *all* moral judgements truth apt, in which case there is plenty of room between non-cognitivism and this, or is it *some*?
Robert Hall in his post was quite explicit on this point, and in the conversation between Huemer and Loeb, the division I propose was also explicitly made (though by Joe Schmidt, so Huemer doesn't really get the credit there). But in many cases proponents aren't as careful, and then it merits being pointed out (though it doesn't actually affect the strength of the argument, but just means that the proponent has been sloppy in spelling it out).
How does the elimination argument work against my position?
If someone -- in the context of a process of elimination argument -- says something like _There are exactly three positions other than mine_, he is presumably not making an empirical claim about what the standard positions are.
Instead, he might be saying something like _There are three dialectically relevant logical possibilities other than the logical possibility that I endorse_ -- in other words, he is implying that he has divided up the positions such that he has objections to those three positions, and that the objections will cover the denial of his own view, either because those positions are logically exhaustive, or the three positions plus their combinations are logically exhaustive. In the case of Michael Huemer, I think that he does not define the positions to be exclusive of each other, and acknowledges that one could have a combination view -- but his objections apply to combination views as well, so this doesn't matter for the argument. In my case, and the case of Silas Abrahamsen's essay, we divided them up in a way to be exclusive of each other and logically exhaustive.
(Actually, what Bentham's Bulldog said was 'Given that, as previously discussed, moral realism is the view that there are true moral statements, that are true independently of people’s beliefs about them, there are three ways to deny it'. I don't think that he was making an empirical claim about the standard positions.)
Of course, if all you have ever done is made the empirical claim that the positions, as most commonly defined, don't cover yours, then you may well be right. But if you are objecting to the division into three anti-realist positions in the context of someone's presentation of an argument for moral realism, to show that you are correct, you would need to show that the stipulated positions (or the positions and their combinations) are not logically exhaustive, or that an objection fails to address the relevant position.
//Instead, he might be saying something like _There are three dialectically relevant logical possibilities other than the logical possibility that I endorse_ -- in other words, he is implying that he has divided up the positions such that he has objections to those three positions, and that the objections will cover the denial of his own view, either because those positions are logically exhaustive, or the three positions plus their combinations are logically exhaustive.//
Huemer's objections to the standard positions do not, as far as I can tell, include any objections to my views.
//In the case of Michael Huemer, I think that he does not define the positions to be exclusive of each other, and acknowledges that one could have a combination view -- but his objections apply to combination views as well, so this doesn't matter for the argument.//
My position is not a combination of the three traditional views *and* it's not clear Huemer's view does apply to combination views, since combination views may rely on different stances towards language/semantics. That concern is moot though, since I don't endorse pluralism and have little interest in defending it.
//In my case, and the case of Silas Abrahamsen's essay, we divided them up in a way to be exclusive of each other and logically exhaustive.//
If Silas's three categories include my view then they aren't the three categories I was saying aren't the only three. Anyone can arbitrarily make anything fit into three categories, or even just two. For instance, all objects are either potatoes or non-potatoes. Number of categories is often arbitrary or a matter of utility or convention for some specific purpose. I am not concerned with the number of categories, but their contents.
According to Huemer, there are three antirealist positions: noncognitivism, error theory, subjectivism. I'm an antirealist and I don't endorse any of those positions. My position is not logically impossible and is not reasonably included within any of the those three positions. Neither is incoherentism and possibly pluralism isn't either. So there are two, maybe three positions excluded by Huemer's claim, and possibly more. So Huemer is mistaken.
...This is not that complicated. What I said was true, is true, and will remain true, and neither you nor anyone else has said anything that could be reasonably interpreted as a refutation of my initial claim.
Creating a new set of three categories, gerrymandered by design to be mutually exhaustive, does not refute my original claim.
If, for instance, someone said that there are only three fruits: apples, bananas, and oranges, and I say "no there aren't, behold the watermelon." This could not be refuted by saying:
"Aha, but suppose I do this: Apples, bananas, and the orange/watermelon category! Now there are still three categories!"
The issue was never about how many categories there were. It was about the specific presuppositions and strictures underlying the three categories Huemer insisted were the only three options. Since all three options rely on e.g., specific presuppositions about semantics that I reject, my position definitely does not fall within those three categories.
Of course if you create some new set of three categories you can force-fit my position into it. This is a profoundly uninteresting and pointless exercise.
I don't think this is that great of a test because you can always just be vague enough about your position that it's not clear which category it falls into or hold a position that is agnostic as to which category is true. This wouldn't prove that the categories are non-exhaustive, but it would still be impossible to sort your position into one of them. Similarly, you could always hold a position that clearly seems to fall into one of the categories, but deny that it does.
What do you object to about the revised classification scheme? Given that it's logically exhaustive, you can't just say that it ignores certain options. I guess you could say that the words used to define the categories (truth-apt, true, stance-independent) are themselves ill-defined, but in that case moral anti-realism itself becomes an ill-defined position.
One thing that frustrates me is that philosophers don't take the time to explain to people who aren't familiar with a philosophical question why it matters. What are the stakes of the moral realism question? How do moral realists and anti-realists see the world and act differently? I've seen people debating moral realism on here for months and have just as little understanding of the stakes as when I started. It all seems more like a game than an issue that I should take seriously.
Well in my previous post I argued that it doesn't really matter much.
That's a great article! So why do people go on talking about moral realism and anti-realism then? Is it bullshit in the Frankfurt sense that no one in the debate cares whether what they're saying is true or not? Maybe they're just seeking some kind of approval or something.
Thank you! I can't say what everyone's motivation is. I certainly don't think they're spewing Frankfurtian bullshit--I assume people care about saying what they believe is true in the domain. Personally I just think it's an intrinsically interesting subject, which also touches on other subjects that I find interesting (I suspect this is the case for many people talking about it). As I mentioned at the end of the aforementioned article, it also has at least some relevance for theism/atheism (something that itself might have some practical relevance).
Oh and to add for your proods, there are tableaux tree proof generators you can try: https://www.umsu.de/trees/
Thanks! I'm currently studying for a logic exam, though, so I thought I'd use it as an excuse to tell myself I'm being productive. But I appreciate the consideration:)
Good luck!
Graham Priests book on non classical logics has a good intro on tableax proofs with trees if it might be of interest.
Thank you, and thank you for the recommendation!
I appreciate posts like these because it creates opportunities for those that don't really fit into the traditional analytic discourse to point out that there are many other ways to approach language and intersubjectivity. Your exhaustive list of positions all assume something like correspondence between thoughts/beliefs/desires and their representations in language. For various forms of quietism, pragmatism, or incoherentism, there is no correspondence. I fall somewhere in this camp. Given my understanding of language and subjectivity, when we talk about ethics with one another, we are engaged in a purely verbal language game, united by a family of resemblances. Furthermore, it's merely practical that we describe our aspirations and ethical desires in terms of empirical or synthetic a priori descriptions of reality, because we also play those language games all the time it usually gets the job done! It's like going to play chess, realizing you can't find the chess pieces, so you borrow checkers pieces. But then you play chess-with-checkers-pieces long enough with the same people, until it doesn't feel incoherent, given a particular context, to incorporate some of the functions of checkers into the chess game (like denoting pawn promotion with a stack of checkers pieces, and keeping in the function of double-jumping). Imagine if Stockfish (the best chess computer) were to analyze a game between two players that used checkers pieces and incorporated double-jumping for stacked pieces, but only when the pawn had been promoted. I imagine Stockfish would simply say this action is not valid or meaningless. Well, a lot of analytic philosophy is like that to me. Analytic traditions try to delineate moral utterances from non-moral utterances in terms of the proper rules (syntax, rules of validity) and pieces (atomized concepts); and it borrows rules and pieces from other language games, such Kantian analyses of synthetic a priori knowledge. But this is a superficial distraction from the activity of morality, which connects desires/aspirations with actions within a particular situation. The purpose of philosophy, to me, is to help us pay attention how we are using and misusing philosophical language when attempting to represent or justify moral activity. In other words, there's no metaethics outside of the attempt within analytic philosophy to make moral utterances do stuff it was never meant to do.
I endorse this response. As a pragmatist and quietist I simply reject the foundational assumptions within which contemporary analytic metaethics is framed. I'm not stuck with analytic distinctions.
Every moral utterance is driven by cares & concerns, and most of them appeal in some way to facts that are ostensibly independent of any cares / concerns, yielding one of various forms of strategically-suggestive blurtings, e.g., "It's bad to do X" "Do Y!" "Z is permitted" "X is obligatory" "Booo Y!" "Z is forbidden." This in turn becomes a "shitmess" due to this wide variety, the utter slop, and the chronic omission and/or mixing-up of the cares/concerns and the strategic facts. Attempts to shoehorn this into tight fourfold taxonomies are just drawing leylines through the wild scatter.
"Also we should generally be weary of disjunctive arguments like these, as you can give plausible-sounding objections to almost all philosophical positions."
Do you mean wary?
Such dychotomies as "objective/subjective" or "real/unreal" or "stance-independent/stance-dependent" are very unhelpful. Artifacts of ancient philosophy which didn't understand map-territory relations, treated mind and matter as separate magisteriums and therefore was tremendously confused about basically everything, unable to separate baby from the bathwater.
From which "stance-independent/stance-dependent" is probably the worst offender. With subjective/objective it's possible to conceptualize that things that are subjective are also objective. With real/unreal it's harder, but you can come up with some notion of "being real but in a different way that was initially implied and therefore, unreal in the initial sense". But the notion of dependency/independency... it's either incredibly strict or completely meaningless.
Suppose I define morality as a certain generalization of human values as developped by evolution through natural selection. Is it stance dependent? Well yes, I'm literally using moral stances of people throughout history as an input before I use it to train statistical classificator. On the other hand I'm pointing to something in the real world, not just my interpretation of it. And after my classifier is already trained it doesn't matter how much we modify the training data. So if alien ray modifies the stances of all people currently living on Earth to think that torturing babies is good, it wouldn't affect the answer of my classifier, so the standard argument against relativism doesn't work. But also there are multiple different ways to generalize human values from the data.
In the end the framework of stance-dependancy misses the actually interesting meta-ethical questions
I would put the four limbs slightly differently and more generically:
1. All claims about X are incapable of being truth-apt.
2. Claims about X are conceptually capable of being truth-apt but in all our contexts none ever are.
3. Claims about X can be truth-apt for us, but only ever where the evaluation of those claims depends upon some stance(s) (being stance(s) not intrinsic to X or the surrounding context potentially available to us).
4. Claims about X can be truth-apt for us without needing to employ any non-intrinsic stances.
If in some context 'X' has a consistent connotation, then those four possibilities about the truth-apt-ness of X must be mutually exclusive within that context. (One must be correct, though people's positions about which is correct need not fit neatly into any one box.)
That's one concern... whether/which one of the 4 limbs holds true with respect to some notional set of all possible claims about X.
The other concern is the characterisation of what people are doing when they make claims about X. As in, if someone says "atrocity always bad", the question of whether the person is or isn't intending to express it as a morally real truth and, if you say "not", whether you can then only deny it on the basis of the first 3 limbs.
The answer to the latter being no, you are not bound by those 3 limbs.
E.g. if someone says "it's ok to steal from the extremely rich to feed the starving" it might be that they (1) are expressing their approval of that, (2) consider morality to be inapplicable (neither moral nor immoral), (3) consider that it is a true statement according to them but not necessarily for all value-systems, (4) consider that it is a true statement according to some universal true moral system, (5) making a claim without considering its truth, (6) making a claim without considering whether it is universally true, (7) making a claim while holding a contradictory position, (8) misspeaking, (9) some combination of those things... plus probably some other things that I haven't included. And only (4) involves the adoption of a realist position.
It is worth noting that for a claim to be 'neither truth-apt nor not truth-apt' would entail a logical contradiction, so that isn't a logically possible position.
This is not true. A claim could be indeterminate. One is not obliged to regard all claims about language as bound to the diktats of classical logic.