Error Theory and Subjectivism Still Do Not Entail Absurd and Unacceptable Results

Talis' arguments against error theory and subjectivism trade in equivocations, ambiguities, and falsehoods

Talis wrote an article arguing against error theory and subjectivism. I wrote a response. They’ve now released an article responding to my response. This is my response to their response to my response.

Talis misses the mark yet again.

Equivocation

Talis provides the following argument against error theory:

The Genocide Argument Against Error Theory

1. If moral error theory is true, then there is nothing wrong (or right) with genocide.

2. But, there is something wrong with genocide.

3. So, error theory is false.

Talis claims that their argument against error theory doesn’t assume non-naturalism and can in fact be utilized by people with other meta-ethical views (such as subjectivists) to argue against error theory. However this poses a problem for Talis: As I said in my initial article, when an error theorist affirms that there is nothing wrong with genocide, what they’re affirming is that there is no stance-independent non-natural moral fact that genocide is wrong.

If this is not the proposition being talked about in Talis’ argument against error theory, then they are simply equivocating on “nothing is wrong with genocide”. Talis is making an error that many people make: conflating semantic disputes with propositional disputes. The fact that two people use words in different ways does not mean there is a proposition in contention between them.

What Talis is doing is tantamount to following scenario:

Suppose Charlie says he has a bat in his house. Tasha responds by claiming that Charlie’s statement is false because there is no winged nocturnal mammal in his house. However, suppose that when Charlie claimed that there was no bat in his house, he meant a baseball bat.

Tasha’s response here did not actually address the proposition Charlie was affirming. Likewise, if Talis is talking about something other than stance-independent non-natural moral facts when they say there is something wrong with genocide, they simply are not addressing what the error theorist is saying when they claim nothing is wrong with genocide.

What Does “Wrong” Mean?

This also leaves open the question - what exactly is Talis referring to when they say “there is something wrong with genocide”? Without clarification on what they mean by these terms, how are we even supposed to evaluate the truth value of the premises?

An Analogous Argument Against Realism?

Talis proposes the following hypothetical argument against moral realism:

The Genocide Argument Against Realism

1. If moral realism is true, then there is nothing wrong with genocide.

2. But, there is something wrong with genocide.

3. So, moral realism is false.

Talis claims that if the first premise of this argument were true, then it would be a good argument against moral realism that could be utilized by subjectivists. However, they say if we were to interpret the argument in the way I’m apt to, subjectivists could no longer use the argument:

The Genocide Argument Against Realism

1. If moral realism is true, then there is nothing stance-independently wrong with genocide.

2. But, there is something stance-independently wrong with genocide.

3. So, moral realism is false.

I think this latter interpretation of the argument is the one that makes the most sense. If this is not the intended interpretation, I struggle to understand what the argument is even trying to say.

The PhilPapers Survey

In my original response, I noted that if Talis were to say error theorists are affirming something absurd and unacceptable when they say “there is nothing wrong with genocide”, Talis would have to say that most philosophers and meta-ethicists are committed to something absurd and unacceptable as well since they affirm the same proposition¹.

Talis claims I don’t take sides on whether we should pay attention to the survey results. But I did take sides: I explicitly said in my article that I don’t think a view being endorsed by most philosophers immunizes it from accusations of absurdity.

There are two main reasons I brought this up:

1. I don’t think Talis realized that what they were calling absurd and unacceptable is the majority view among philosophers and meta-ethicists

2. Non-naturalists often love to cite PhilPapers survey to argue against anti-realism. But the proportion of philosophers who endorse non-naturalism is about the same as the proposition who endorse anti-realism. Why then would we have any more reason to disbelieve anti-realism than non-naturalism on the basis of this evidence?

Too Narrow An Interpretation?

Talis critiques my interpretation of their argument as “too narrow” because it leaves open the possibility that other meta-ethical views are true. Talis claims that if we interpret the first premise as “If error theory is true, then there is nothing stance independently wrong (or right) in a non-naturalist sense with genocide”, then this leaves open the possibility for other meta-ethical views to be correct. Naturalist realism, subjectivism, and expressivism all reject the existence of stance-independent non-natural moral facts as well, but these theories are incompatible with error theory.

I am unsure what the problem here is supposed to be. Suppose there is a view called “Travisism”. According to Travisism, there are no unicorns. It also says that cats exist.

Suppose that someone provides the following argument against Travisism:

P1. If Travisism is true, there are no unicorns.

P2. There are unicorns.

C. Therefore Travisism is false.

One could object to P1 by saying it “leaves open” the possibility that cats don’t exist and thus it is “too narrow”. But this doesn’t seem to be a problem: P1 is still true, and the argument is sufficient to disprove Travisism (assuming the premises are correct).

What Error Theory Claims

Talis claims:

If error theory is true, then there is nothing wrong (or right) with genocide in any sense, not just in the non-naturalist realist sense.

But this is not true. An error theorist is only committed to saying that the way the statement “there is something wrong with genocide” is used in ordinary moral discourse, it expresses a false proposition - because it refers to stance-independent non-natural moral facts, and these facts do not exist.

But this does not mean an error theorist will think that anyone who utters the statement “There is something wrong with genocide” is always saying something false regardless of what they mean by it.

If someone said “There is something wrong with genocide” but all they meant by this statement was “I disapprove of genocide”, an error theorist could agree that the proposition expressed by “There is something wrong with genocide” in this case is correct.

Let’s go back to Travisism. Suppose someone claims that unicorns exist, but what they really mean by “unicorns” is cats. Is a Travisist dedicated to saying that such a person is saying something false? No - a Travisist could agree that such a proposition is true. They may assert that it’s a non-standard usage of the term “cat”, but this doesn’t mean they think the proposition expressed by it is false.

Another example would be atheism. An atheist might say since God doesn’t exist, all statements referencing God are false.² There are actual people named God. If I were to say “God was forced to prove his existence in court” - referencing the human named God - would I be saying something false by the atheist’s lights? Of course not. When the atheist says that all statements assuming God are false, they obviously mean statements that assume the existence of an all-good, all-knowing, all-powerful creator of the universe. They don’t mean any statements which just so happen to use the word “God” regardless of what the term refers to.

Likewise, when an error theorist says that all moral statements are false, what they mean by this is that all statements which assume the existence of stance-independent non-natural moral facts are false. What they don’t mean is that any sentence that uses terms like “right” or “wrong” - regardless of context of use - are false. I mean, would Talis say that an error theorist is dedicated to saying the statement “People who say that vaccines cause autism are wrong” is false - since it asserts that something is wrong? Of course not. This is simply not the sense of “wrong” that error theorists are concerned with.

Everything Is Permissible?

In my initial response, I commented that I felt it was important to clarify that error theorists do not think genocide is permissible. A similar sentiment is echoed in this article on the SEP:

The last example (“Stealing is not morally wrong”) calls for an extra comment. In ordinary conversation—where, presumably, the possibility of moral error theory is not considered a live option—someone who claims that X is not wrong would be taken to be implying that X is morally good or at least morally permissible. And if “X” denotes something awful, like torturing innocent people, then this can be used to make the error theorist look awful. But when we are doing metaethics, and the possibility of moral error theory is on the table, then this ordinary implication breaks down. The error theorist doesn’t think that torturing innocent people is morally wrong, but doesn’t think that it is morally good or morally permissible either. It is important that criticisms of the moral error theorist do not trade on equivocating between the implications that hold in ordinary contexts and the implications that hold in metaethical contexts.

Talis responds to this point by claiming that error theorists do in fact think genocide is permissible because for X to be permissible is just for X to not be impermissible. While one certainly can use the terms in this way, I question why they would want to given that the obvious connation of such a claim is that one is fine with genocide.

To be fair, Talis does acknowledge in their article that one can be an error theorist and oppose genocide. But the insistence on claiming that error theorists think genocide is permissible strikes me as bizarre and rhetorical in nature.

On Subjectivism

This is Talis’ argument against subjectivism.

The Rescue Argument Against Subjectivism

1. If subjectivism is true, then anyone who doesn’t care to rescue drowning children doesn’t have to rescue drowning children.

2. But, they do have to rescue drowning children (all else being equal).

3. So, subjectivism is false.

Talis says the following:

Travis also accuses me of thinking subjectivism is a normative ethical theory since I’m supposing that subjectivism determines that something is right (or wrong) so long as someone approves (or disapproves) of it. (Seriously?)

Talis can make snide remarks all they want, but their initial article just did give off the impression that they think subjectivism carries certain first-order normative implications: namely that it would be permissible for someone to refrain from saving a child if they didn’t want to. If they didn’t want their article to be interpreted in that way, they should have taken the time to be clear instead of making vague and ambiguous remarks.

Travis asserts that subjectivism is just the metaethical theory that when someone says “X is right”, they mean “they approve of X”. But the theory doesn’t only concern itself with moral claims. Subjectivism also asserts that the preferences of individuals determine what the content of morality is (for those individuals). So, yes, subjectivism posits that X is right so long as someone approves of it. If this were false, then I’m not sure how subjectivists would understand moral claims as posited by Travis.

It is unclear to me what exactly Talis means when they say “subjectivism also asserts that the preferences of individuals determine what the content of morality is (for those individuals”. I don’t know how they’re differentiating this from the assertion that moral statements are made true by the speaker’s attitudes.

My description of subjectivism is totally in line with standard definitions of the view. For example, here’s how the IEP describes subjectivism:

Ethical subjectivism can be understood as the view that the meanings of ethical claims are propositions, but propositions about speakers' attitudes.

Regarding the claim that “subjectivism posits that X is right so long as someone approves of it”, this is true in one sense and false in another. On subjectivism, there is no “right” or “wrong” simpliciter. Things are only right or wrong relative to some standard.

Compare this to other, non-ethical forms of subjectivism. Suppose one is a gastronomic subjectivist. Does the fact that there’s some pervert out there who loves the taste of feces mean that you as a gastronomic subjectivist are committed to thinking that feces is tasty? Well, you’re committed to thinking that feces is tasty relative to the taste preferences of those who enjoy the taste of feces. But you can still assert that feces is not tasty relative to your taste preferences.

Likewise, does the fact that there’s some psychopath out there who approves of torturing babies mean that you as an ethical subjectivist are dedicated to saying that torturing babies is right? Just that it’s right relative to the moral standards of people who approve of torturing babies. You can still say it’s not right relative to your own standards!

Subjectivism and Moral Obligations

Talis says:

[The subjectivist is] dedicated to saying that, for the person who doesn’t care, there is no moral obligation to save children from drowning in any sense whatsoever–they’re not affirming a tautology.

A naive subjectivist might argue that since they personally have a preference for others to rescue children, that this also means that others have a moral requirement to rescue children. But this is clearly not true. Our preferences do not necessarily imply that others have moral requirements–at least under subjectivism.

I have no idea what construal of subjectivism Talis is working off of that leads them to think a subjectivist can’t say that others have a moral obligation to rescue children on the basis that they approve of rescuing children.

Again, subjectivism just is the view that moral statements are statements about your own attitudes. “People have a moral obligation to rescue children”, on subjectivist semantics, just is the same proposition as “I prefer that people rescue children”. So of course a subjectivist can say that others have an obligation to save children on the basis that they prefer for others to save children - because these statements mean the same thing!

Illegitimate Criticisms?

In their original article, Talis claimed that subjectivists are unable to “legitimately criticize” the moral views of others. In this newest article, they elaborate more on that claim:

By this comment, I simply mean that it’s impossible for subjectivists to legitimately criticise those in question, since subjectivism cannot provide the grounds for such a thing…This comment stems from the idea that in order to criticise, one must point to some mistake that a person is making. When talking about morality, we’re talking about moral mistakes. But, under subjectivism, if someone holds a coherent set of beliefs in relation to what they stand for, there is no mistake to be made.

Presumably the notion of “mistake” Talis has in mind here is broader than just a factual error, since it seems that one could be aware of all the stance-independent moral facts and simply not act in accord with them - yet I imagine Talis would still say such a person is making a mistake. So I’ll treat acting out of accordance with the stance-independent normative facts as under the umbrella of “mistake”.

It seems bizarre to me to insist that you need to point to some mistake someone is making in order to criticize. By Talis’ lights, it seems that that one can’t engage in food or movie criticism unless they’re a gastronomic and aesthetic realist.

At best, the subjectivist can say that they have different preferences, and ‘criticise’ the misogynist insofar as their preferences differ. But I’m not sure how much of a ‘legitimate criticism’ this would amount to. After all, the misogynist could happily return the same ‘criticism’ and it would be just as ‘legitimate’.

What I would say about this as someone who uses moral language in a subjectivist manner is that “legitimate criticism” is a normative notion.³ So whether a criticism is legitimate or not depends on your stances - it might be legitimate relative to a feminist’s stances, but illegitimate relative to a misogynist’s stances. I do not see the issue with this.

Normative Entanglement

Talis claims that my accusations of normative entanglement are unfounded. They say:

I’m only including this segment to cover my bases. I’m pretty sure Travis is NOT arguing for this, but denying that someone has a moral requirement to X, does not mean you’re fine with them not doing X. It’s on this point that someone might accuse my argument of relying on normative entanglement. For instance, “you’re arguing that subjectivism implies that subjectivists should be fine with the actions of anyone so long as it’s in line with that other person’s preferences/ moral requirements”.

But this is not at all what I’m arguing. This assumption would be built upon the false view iterated above. A subjectivist can very coherently judge someone who refrains from rescuing a child, even if the subjectivist thinks the individual had no moral requirement to rescue them. I never argued otherwise either.

I agree that one can deny that one has a moral obligation to do X while still opposing them refraining from doing X. This is what error theorists do.

But it’s pretty obvious what the connotation of stating that someone thinks there’s no moral obligation to rescue children is - to the ordinary person, this sounds like such a person thinks it’s permissible to refrain from rescuing children.

I think many people - if they have little familiarity with the positions at stake - would come away from Talis’ article thinking that subjectivists think it’s okay to refuse to save a child if you don’t want to. At the very least, I think people would come away from the article thinking Talis believes that subjectivists are committed to saying it’s okay to refuse to save a child if you don’t want to.

You’ll notice that nowhere in Talis’ original article do they take the time to clarify that subjectivists can still think it’s wrong for someone to refrain from rescuing children even if they have no desire to.

Talis acts like this is an unreasonable interpretation of their post but it is simply not. First of all, Talis already demonstrated in their article that they don’t understand subjectivism - by erroneously claiming that subjectivists can’t claim people have moral obligations to do things they don’t want to - so it is not unreasonable to assume they misunderstood the view in other ways.

Second of all, it is routine for moral realists to argue against subjectivism by erroneously claiming as long as someone approves of something, subjectivists can’t say it’s wrong for them to do that thing.⁴

Given that realists routinely misrepresent⁵ subjectivism, Talis should have made it crystal clear what subjectivists are and aren’t committed to. Talis thinks they can just say that they didn’t intend to imply that subjectivists to think it’s okay to ignore drowning children and their hands are clean - they’re cleared of any charges of normative entanglement. No.

It is simply not enough that one didn’t intend to give off these misleading impressions. One has an active duty to take care to clearly present things so that audiences don’t come away with the wrong impressions. Talis did not make a serious effort to do this. And so my original charge - that normative entanglement artificially makes their objections to anti-realist views appear far stronger than they actually are - still stands.

1. Obviously they wouldn’t phrase it the same way, but the proposition at stake here is just “There is no stance-independent non-natural moral fact that genocide is wrong” - which basically everyone but non-naturalists would affirm.

2. Excluding statements which deny the existence of God.

3. At least in the context of the moral domain. I think there are other contexts where we operate under a non-normative notion of “legitimate criticism”.

4. For an example, see the following excerpt from this paper: “For instance, individual moral relativism, in its simplest form, says that which actions are right or wrong is relative to individuals: the morality of my actions is determined by my own moral standards, and the morality of yours by yours. But this is clearly false, too. It implies that the actions of a racist are right, as long as the racist is living up to their own standards.”

5. By “mispresent” I don’t mean to imply that realists are intentionally doing this. I think most of them are just confused.