Tricks, revisionism, and evil: A response to Jason Brennan

1.0 Introduction

Antirealism is on the receiving end of a cavalcade of misguided objections that rely more on rhetoric, negative framing, tendentious labeling, and misleading implication than substantive criticism. I believe Brennan’s remarks exemplify at least some of these qualities. This is unfortunate because Brennan is a professional philosopher and I would have hoped that critiques of antirealism would be more substantive, feature less rhetoric, and involve less unwarranted suggestions that antirealists themselves are bad people or defend their views in objectionable ways.

A better critique would serve to illuminate the issue in question. I don’t believe Brennan’s remarks exemplify this, but instead read as a denunciation of a view without much substance. Here is the original remark:

2.0 Misleading modifiers

First, note that misleading modifiers appear in all four of the numbered remarks:

1. Racism is not really bad. (Or “It’s not true that racism is bad.”)

2. The holocaust was not really bad.

3. Slavery is not really bad.

4. It’s not really good to help people.

As I argued in a previous post, use of terms like “really” in contexts like this is what I called deceptive modifiers. I have opted to change the name to misleading modifier so as to be less likely to suggest intent on the part of those employing them (and not because alliteration, that’s just a bonus). Misleading modifiers are a rhetorical tool that exploit the ambiguity between an (often unstated) technical sense use of the term and various ordinary usages that carry pragmatic implications. Whether intentional or not, prompting others to conflate the two serves to misleadingly imply false implications of a view or to inappropriately evoke a positive or negative emotional reaction from readers or listeners. This doesn’t have to be intentional, and I don’t generally assume that it is. Instead, I think people who make use of misleading modifiers probably do so unintentionally and either fail to appreciate why it’s a bad idea to use them, don’t agree with me that they are using them, or don’t find that doing so is objectionable.

In this case, the impression Brennan’s remarks are likely to give to many people is that antirealists aren’t opposed, or at least as opposed, to racism, slavery, or the Holocaust or don’t favor helping people (or at least that this wouldn’t be consistent with being a sincere and committed antirealist). Suppose you observed the following exchange on the street:

Alex: Sam, do you think racism is wrong?

Sam: I don’t think it’s really bad.

What would you infer about Sam? I think most of us would infer Sam is not as opposed to racism as other people. Or suppose you saw an exchange like this:

Alex: Sam, what is your attitude towards the Holocaust?

Sam: Well, it’s inconsistent with my values, but I don’t think it was really immorally wrong.

Once again, Sam sounds like a pretty awful person. Even with the inclusion of the remark that it’s inconsistent with his own values, Sam’s remark still conveys a lack of repugnance, outrage, and condemnation that we’d expect from a good person.

The problem is that antirealism does not entail, imply, or necessarily lead to any changes at all in a person’s attitude, disposition, or behavior. It is entirely consistent with being an antirealist to fully dedicate one’s life to eradicating racism or preventing a holocaust. A person could consistently maintain that there are no stance-independent moral facts while literally in the act of giving their life to stop a genocide. Our moral conduct, attitudes, character, and behavior are orthogonal to our stance about the ontological status of moral truth claims. So to say that an implication of moral antirealism is that “Racism is not really bad” is the sort of remark one would hope a philosopher would handle with care, recognizing that if by “really” you mean stance-independent, that technically on such a construel antirealism implies (even entails) that “Racism is not really bad,” that this could easily mislead others into thinking that the antirealist can be (and probably is) just as opposed to racism as the realist. One could pragmatically cancel out this implication by explicitly adding that while an antirealist does not think racism or the Holocaust or anything else is stance-independently wrong, it is nevertheless consistent with such a view to be opposed to these actions. Failure to do so leaves the implication floating, and is likely to mislead people into making negative inferences about the character of moral antirealists.

If such an implication is intentional, then I think it’s immoral for realists to speak in this way. Intentionally misleading audiences into thinking those who hold a rival philosophical position are bad people and that this is somehow implied or entailed by their philosophical positions is not a philosophically legitimate move to make. Perhaps realists actually think antirealists are bad people for being antirealists. In that case, the kinds of remarks Brennan has made still wouldn’t be a good way to do this, because they may still mislead audiences into thinking that antirealists don’t care about, or don’t oppose racism, the Holocaust, slavery, etc. even when this is demonstrably untrue: it is not an implication of antirealism that antirealists must be less opposed to moral transgressions, nor is it an implication of the view that they are less likely to intervene to stop them. It’s certainly possible to argue that antirealists are bad people because they don’t take the right stance towards these issues, but if that’s the objection, why not make that objection?

If this is unintentional, then I’d urge realists to consider whether some of their ways of framing the dispute between realism and antirealism may inadvertently prompt misunderstandings in their audiences via normative entanglement. Either way, if realists think there is something immoral or unsavory merely about endorsing antirealism, there are better ways to express this than the way Brennan has opted for here.

One aspect of these disputes that puzzles me is why realists are sometimes hostile towards antirealism, and seem to regard antirealists as bad people. If you are a moral realist, suppose you discovered that moral realism was false, and that there were no stance-independent moral facts. Would believing this make you a bad person? If so, why? And if you did become a moral antirealist, would you be less opposed to racism, the Holocaust, or slavery? I sure hope not. If it would, then why, exactly, would that make you a morally better person than the antirealist? And if it wouldn’t, and you did become less opposed to racism, the Holocaust, or slavery, or if you stopped opposing them at all, then perhaps you should reconsider your attitude towards antirealists: antirealists typically oppose these things in spite of them not “really” being wrong. Perhaps that’s commendable!

Personally, when it comes to judging whether people are good or bad, I care about their attitudes and behavior. It is entirely consistent with moral realism and antirealism to have the same values, to act in the same way, and to have the same emotional responses towards helping and hurting others. Just about the only observable difference one might find between a realist and an antirealist is that a realist, when asked, might say things like “I’m a moral realist,” while an antirealist wouldn’t. The dispute between the two positions is a purely intellectual one that borders on epiphenomenal; it simply doesn’t make that much (if any) of a practical difference if someone is a moral realist or an antirealist. If any realists reading this disagree, I’d be really interested in why. Leave a comment below or get in touch. You’d be welcome to join me on my YouTube channel to discuss why you disagree.

Either way, philosophers who employ misleading modifiers could opt to employ clear language that accurately conveys their views. Instead, they often employ normatively loaded language without sufficient clarity to disambiguate the actual intellectually defensible implications of a view from the positive or negative associations or misleading implications misleading modifiers can foster in audiences. This is ironic, since Brennan goes on to accuse antirealists of engaging in “semantic tricks.” Perhaps if we want to draw a distinction, I would suggest that Brennan and others often unwittingly exploit “pragmatic tricks,” though I hesitate to call them this since I don’t want to imply that they do so intentionally. Rather, I think inattention to pragmatics and ordinary language more generally causes philosophers to make avoidable errors.

Applying these concerns to Brennan’s remarks, the term “really” bakes in the critic’s normative assumptions, metaphysical assumptions, or some combination of both in such a way so as to imply something unappealing about a rival position without coming out and clearly stating just what that unappealing feature of the position is. Some other set of words may fail to conceal the vacuity of the remarks; their force is conveyed almost entirely by pragmatic insinuation. Consider one of Brennan’s examples: that an implication of moral antirealism is that “Racism is not really bad.”

I’m a moral antirealist. Do I think racism is “not really bad”? No. I do think racism is really bad, and I am happy to say so: Racism is really bad. I am not obliged to grant Brennan’s conception of what it would mean for something to “really” be (morally) good or bad. I reject realism and I also reject the notion that only realists have an acceptable account of what moral goodness and badness are, such that, if something is to “really” be good or bad, it must only be good in the way realists suppose. Brennan is not entitled to force his conception of the meaning of these terms on me or anyone else, nor is he entitled to declare any alternatives to his conceptions as “revisionary,” since I never agreed ordinary language was committed to realism in the first place.

Realists often take their position to one for which there is a presumption in its favor, and that antirealists must overcome this presumption to establish their position. I deny that there is any presumption in favor of moral realism. I’m not the only one that challenges this presumption. Neil Sinclair (2012) wrote an excellent paper challenging the presumption of realism here, and Nils Franzén (2024) wrote a more recent paper questioning the presumption of realism here. The latter is interesting. Franzén draws on literature on “subjective attitude verbs” (see p. 1194). According to Franzén, subjective attitude verbs “designate a kind of attitude that an individual can have only towards a class of contents which are intuitively subjective” (p. 1194). In English, “finds” serves as an example, though Franzén notes that English exhibits idiosyncrasies that make it less than ideal for conveying the general issue (see note 1). Consider this example (adapted from Franzén):

Alex finds pineapple on pizza tasty.

This sounds perfectly copasetic. Now consider this remark:

Alex finds that pineapple is a fruit.

This doesn’t sound right. English speakers wouldn’t typically say that one “finds” factual claims like this to be the case. Franzén gives another pair of examples to cement the point:

Karl finds Tuesday nights boring.

Karl finds it Tuesday.

The former is a normal locution; the latter isn’t. Franzén quotes Sæbø (2009), who says:

[…]there is an intuition that it does not make sense to entertain a subjective attitude to something which is either a fact or not a fact, regardless of the subject of

the attitude […]. (p. 329, as quoted in Franzén, 2024)

I’m not a fan of intuition talk, nor the vague notion that “there is” an intuition; intuitions are something that, if they exist at all, are things people have. There aren’t just intuitions floating in the ether. I wish philosophers would be more careful about this sort of thing. Nevertheless, if we grant Sæbø is correct about how we use the term “Find” in English, we can apply this to moral cases. Franzén applies subjective attitude verbs to moral claims, noting that they don’t appear to be a poor fit in such cases:

The observation speaks firmly against the hypothesis, since moral predicates, unlike objective predicates, interact felicitously with subjective attitudes. That is, one can felicitously assert sentences like:

(7) Karl finds it wrong to eat meat.

(8) Karl finds the government’s stance on climate change morally reprehensible.

(9) Karl finds the EU’s new immigration laws cruel. (p. 1196)

I’m not generally in favor of armchair approaches, but even I recognize that these are totally ordinary remarks. Does this decisively smash the case that ordinary language is committed in some way to moral realism? No. But perhaps it should give philosophers who lean so heavily on armchair methods some pause, and some reason to consider that at least some aspects of the English language, and other languages besides, appear to be a better fit for folk subjectivism than folk realism. As Franzén observes:

The fact that moral predicates, just like intuitively subjective predicates, interact felicitously with subjective attitude verbs stands in stark contrast to the widespread view among metaethicists in recent decades that subjectivist theories are in prima facie disaccord with how we speak and think about morality. (p. 1196)

My point in providing this example isn’t to declare the matter settled in favor of folk subjectivism (I don’t think that it is), but to note that realists often help themselves to presumptions that I don’t think they’re entitled to. I don’t appeal to these kinds of armchair theoretical considerations. Instead, I endorse a language as use view, and consider questions about how nonphilosophers speak and think to be empirical questions. And, at present, I believe the best available data does not support the notion that most people speak, think, or act like moral realists or in a way that best fits a realist analysis (I have several posts address the empirical case against folk realism, such as this one). Franzén’s claims may best reflect some aspects of ordinary language (in fact, I’d be surprised if this weren’t the case), but I don’t think the best or only way to address how nonphilosophers think or speak is by consideration of decontextualized sentences. That doesn’t mean I think the practice is useless, though.

This isn’t the only example Franzén provides, either. Franzén provides an example from The Big Lebowski.

In one scene of the movie, Jeffrey “The Dude” Lebowski’s bowling team is approached by Jesús, an opponent for an upcoming game. The following exchange takes place:

● Jesús: Liam and me, we are going to fuck you up [in Saturday’s game].

● The Dude: Well, that’s just, like, your opinion, man. (p. 1196)

According to Franzén:

The comic effect created by the scene is partly due to the fact that The Dude’s comeback is linguistically strange. Who is going to win the bowling game is not a matter of opinion since, presumably, there is or will be a fact of the matter to the question. (p. 1196)

Once again, when the same language is applied to moral claims, for many of us it many not prompt a similar degree of disaccord or strangeness (examples adapted from Franzén):

In my opinion, pineapple on pizza is tasty.

In my opinion, The Big Lewbowski is not that great of a movie.

In my opinion, abortion is morally wrong.

Moral claims more naturally fit with the kind of opinion talk we’d normally associate with presumptively nonrealist considerations, like taste in food or movies. Again, does this settle the matter? No. But what we don’t have is decisive, unilateral linguistic “data” in favor of folk realism.

One could reasonably and sincerely believe that the case is not settled in favor of the view that ordinary people speak and think like realists. One doesn’t have to be a revisionist (by their own lights) or engage in “tricks” to think this. Perhaps we are correct. In that case, is Brennan a revisionist? Is Brennan engaged in tricks? I wouldn’t say so. I’d just think Brennan was wrong, and, perhaps, a bit presumptuous. Perhaps we are incorrect. In that case, well, then we’re incorrect. That doesn’t mean we’re engaging in tricks.

Returning to Brennan’s use of the term “really”, what does “really” mean in this context? In ordinary language, “really” has a number of uses. It is sometimes used to contrast something that is in fact the case versus something that may only appear to be the case but isn’t:

He wasn’t joking. He really meant it.

At other times, it is used to emphasize the strength of something:

Wow! Did you look outside? It’s really raining out there.

What does Brennan mean by “really” in the original comment? I am not entirely sure. But here’s one possibility: while antirealists may say that something is immoral, or bad, or wrong, they don’t think it’s stance-independently immoral, or bad, or wrong. If by “really” one just means “stance-independently” wrong, we could substitute “stance-independent” for “really” in each of the statements:

5. Racism is not stance-independently bad. (Or “It’s not true that racism is [stance-independently] bad.”)

6. The holocaust was not stance-independently bad.

7. Slavery is not stance-independently bad.

8. It’s not stance-independently good to help people.

If this is all Brennan means, this is a profoundly trivial observation. Antirealism just is the view that there are no stance-independent moral facts, so of course an antirealist would agree with all of these remarks. But what does this show? Absolutely nothing. It is just a restatement of what antirealism commits one to. It’d be like saying that one of the awful, terrible, no good implications of atheism is that it implies that “God doesn’t exist.” If this is all Brennan means by “really,” then why not use the term “stance-independent”? That, at least, would be clear. This isn’t.

But Brennan may very well not mean stance-independent or something entailed or implied by there being stance independent moral facts. If so, this yields a dilemma. Either the sense in which Brennan thinks moral antirealism implies that nothing is “really” good or bad is constitutive of or entailed by moral goodness and badness being construed in terms of stance-independence, or not. If the former, Brennan’s objections are trivial. If the latter, then antirealism may not imply such implications at all. After all, if the notion of something “really” being good or bad is logically consistent with antirealism, antirealism wouldn’t imply that nothing could be good or bad in the relevant sense, or at least would entail it. It’s also not clear what Brennan has in mind by “imply,” though I reject both entailments and even any substantive implications. But maybe, as Bentham’s Bulldog has suggested, “really” has some primitive meaning here that cannot be fully articulated. If so, this would shroud one obscure notion in another, and I’d hardly feel inclined to endorse a pair of obscurities rather than just the one.

Whatever interpretation we’re supposed to take away from this characterization, realists might hold that only realist conceptions of morality capture a notion of morality that one regards as important, or desirable, or authoritative. Sure, the relativist or the noncognitivist may say that it’s good to help people, but all the relativist means is “I approve of helping people” and all the noncognitivist means is “Yay helping people!” Neither of these presents us with obligations that are binding on us independent of our own attitudes and beliefs. They’re toothless interpretations of such remarks.

If that’s the objection, why not make it? Again, an antirealist might just agree that their position implies things aren’t “really” bad in this sense. But an antirealist may not take these to be significant costs or problems. But note that if this is what people mean when they make moral claims, and the antirealist is correctly recognizing this, is it their fault? Is the moral antirealist somehow responsible for the content of ordinary language?

3.0 Appraiser relativism

Relativism, in its most basic form, is a descriptive account of the nature of morality. Yet Brennan and others often characterize proponents of relativism as though they are evil for endorsing relativism. Indeed, Brennan explicitly made such an argument in the comment section under this post:

1. It is evil to think that whether the holocaust was good or bad for you depends upon whether you like it or approve of it.

2. Some people subscribe to meta-ethical theories which imply that whether the holocaust was good or bad depends upon whether they like it or approve of it.

3. So some people are evil.

This is not a good argument. One reason why it’s not a good argument is that it fails to disambiguate agent and appraiser relativism, and direct whatever the objection is supposed to be towards the appropriate position. If it were specifically directed at agent relativism, I might even agree with the argument. But the language is too ambiguous for it to be clear which form of relativism this is supposed to be an objection to: agent, appraiser, both, or neither. This is a problem because, while such an objection may make sense if directed at agent relativism, it’s not clear that it’s a good argument against appraiser relativism.

Here’s why. According to appraiser relativism, moral claims include an implicit indexical element. As a result, claims like “the Holocaust was wrong,” mean something like “the Holocaust was wrong relative to my moral standards.” If this is what ordinary moral claims mean, then this allows the truth of moral claims to vary as a function of which standard the claim is indexed to. If Alex thinks the Holocaust was wrong, and Sam thinks it was not wrong, then you could have the following claims:

Alex: “The Holocaust was wrong.”

Sam: “The Holocaust was not wrong.”

Each of these statements would contain an implicit indexical, and could be translated as something like the following:

Alex: “The Holocaust was wrong according to my moral standards.”

Sam: “The Holocaust was not wrong according to my moral standards.”

If the Holocaust isn’t inconsistent with Sam’s moral standards, then all it would mean to say that, when Sam says that “The Holocaust was not wrong” is that the Holocaust isn’t inconsistent with Sam’s standards. And if it isn’t, such a statement would be trivially true.

What practical implications does that have for you and me? Absolutely none whatsoever. The appraiser relativist judges actions as right or wrong according to their own standards (or their own culture, in the case of appraiser cultural relativism), not the standards of the person performing the action (or that person’s culture). An appraiser relativist can and hopefully would regard Sam’s moral standards as repugnant and evil.

But the more important takeaway here is that appraiser relativism is, first and foremost, a view about the meaning of moral claims. If it turns out that this is how people used moral language, why would it be evil to think this?

Brennan’s characterization in the above syllogism is consistent with both agent and appraiser relativism. This is bad because it fails to draw an important distinction that, if made, might only work as an objection to some forms of relativism but not others. If this distinction seems arcane or unimportant to you, note that it is notable enough to be drawn in the Stanford Encyclopedia of Philosophy entry on moral relativism:

[...] that to which truth or justification is relative may be the persons making the moral judgments or the persons about whom the judgments are made. These are sometimes called appraiser and agent relativism respectively. Appraiser relativism suggests that we do or should make moral judgments on the basis of our own standards, while agent relativism implies that the relevant standards are those of the persons we are judging (of course, in some cases these may coincide). Appraiser relativism is the more common position, and it will usually be assumed in the discussion that follows. (Gowans, 21, emphasis mine)

If Brennan’s criticism were directed at appraiser relativism, I don’t think it would work. If it is directed at agent relativism, it may work, but note that Gowans describes appraiser relativism as the more common position and works with an appraiser conception of relativism in the SEP entry (I’m not sure that’s true or what Gowans is basing this claim on; I’ve seen no evidence appraiser relativism is more common). So what is Brennan objecting to? Agent relativism in particular? If so, why not draw this distinction? Analytic philosophy is all about making distinctions where relevant. And yet he simply doesn’t do so here. This is common for critics of antirealist positions. Nathan Nobis does this, too. I can understand having an issue with agent relativism. But is there an issue with appraiser relativism, too?

If the charge of being evil were leveled at the appraiser relativist, it’d be very strange. Again, appraiser relativism begins with a descriptive claim: a semantic thesis about the meaning of moral claims. If it turns out that the semantic thesis is correct, this is all the appraiser relativist needs for appraiser relativism to be true. And if we take the semantics of ordinary moral language to fix the nature of morality (I don’t, but many traditional 20th century metaethicists do. If you disagree with such an approach, take it up with them!), then what possible sense would it make to charge the appraiser relativist for being evil merely for having the correct stance about how ordinary people talk? Descriptive positions on how people use words are not plausible candidates for rendering one evil.

Second, appraiser relativism doesn’t carry any substantive normative implications. All an appraiser relativist would be committed to in saying that sometimes statements like “torturing babies for fun is morally good” is true is that when a person (that the appraiser relativist may very well regard as depraved and evil) who is pro-baby-torture makes such a claim, they’re reportings something true about their preferences.

For comparison, suppose that when people make claims about what food is good or bad they are making propositional claims about their food preferences. When Sam says, “vanilla ice cream is good,” this means “I like vanilla ice cream.” Now suppose Sam likes the taste of feces. Sam says “feces is delicious.” According to this analysis, this statement is true. Why? Because it means something like:

Sam: “I like the taste of feces.”

Sam does like the taste of feces, so this statement is true. Does this mean you have to like the taste of feces? And that if I present you with a bucket of pig shit you must want to scarf it down? No. Just so, the appraiser moral relativist is not obliged to act any differently towards racism, slavery, or the Holocaust as a moral realist. They can regard such acts as evil and act accordingly. So why would they be evil for endorsing “relativism”? I would have thought that what made one evil was one’s character and actions. A person is evil when they go out and do evil things, or at least desire to do so, and would if they could. Insofar as relativism is relegated to a metaethical position, it doesn’t have anything to do with how people feel or act; it’s just a view about the meaning of ordinary moral claims.

Now, if we’re talking about agent relativism: sure, that carries some nasty implications. But, as I’ve noted before, this is just one form of relativism; one does not refute “relativism” by refuting a specific form of it. And, as I’ve noted before, the irony is that what makes it so repugnant are its realist-like features: if someone else thinks baby torture is good, this makes it good in such a way that you are bound to acknowledge and are obliged to act accordingly. In practice, agent relativism results in each of us having a host of moral obligations and impositions placed on us all predicated on other people’s moral values rather than our own. So, whereas moral realism obliges you to comply with stance-independent moral truths, agent relativism obliges to comply with a host of stance-dependent moral truths that nevertheless impinge on your agency and make demands of you that may be misaligned your values…just like moral realism. Note, too, that this makes agent relativism, in part, a hybrid theory that incorporates both metaethical and normative components, and that it is its normative implications that make it so objectionable.

4.0 More rhetoric

Brennan next remarks:

Of course, many moral anti-realists don’t quite want to say these things or admit their view implies this.

This is rhetoric. Note the use of the term “admit.” As if the antirealist were guilty of something naughty that they don’t want to confess. Note, too, the focus not on how antirealism is mistaken, but about how there is something unsavory about its proponents. The language here suggests antirealists as dishonest and perhaps intellectually cowardly, because they are reluctant to concede implications of their view that presumably would be embarrassing. Antirealists can’t “admit” something if they don’t think it’s true.

Next, Brennan says:

So, the trick is that they always interpret 1-4 as sentences “inside” moral discourse, governed by semantic rules inside that discourse. They’ll say things like “Oh, 1-4 are first-order claims and moral anti-realism is a second order theory.”

What, exactly, is the trick here? What’s wrong with drawing a distinction between first-order and second-order claims? And why is it a “trick” to do so? Note that Brennan is claiming that the antirealist is engaged in some kind of duplicity in drawing this distinction. It’s a distinction antirealists hold because they sincerely endorse it. They’re not engaged in tricks (I address the first-order/second-order distinction below in note 2).

Brennan continues:

Or they’ll say “Oh, I’m not a revisionist about moral language.”

I don’t think antirealists are “revising” moral language because I don’t take it to presuppose realism to begin with. So I certainly would say that I’m not a revisionist.. If Brennan thinks that as an antirealist that I am committed to some type of revisionism, or thinks that moral language presupposes realism, he’s welcome to have a discussion with me about this. I discuss the topic a bit with Joe Schmid here.

I don’t endorse any sort of semantic thesis according to which ordinary moral claims involve an implicit commitment to realism. I don’t think most nonphilosophers are realists, nor do I think they implicitly speak, think, or act in ways that presuppose realism. I reject folk moral realism, and any notions that are associated with the idea that moral realism is “commonsense,” the “default” position, and so on. Is this a trick? I wrote my dissertation on this topic. Did I dedicate a decade of my life to this topic and counting just to try to play tricks? Maybe, but if a critic thinks so, I’d like to see them attempt to sustain that accusation. Brennan continues:

They won’t let you argue that their view leads to 1-4 because of these semantic tricks.

Should we really call appealing to a distinction between first-order and second-order moral claims or the denial that one is a revisionist about moral language “semantic tricks”? The first isn’t even a semantic distinction; it’s a conceptual one, and the latter is a substantive disagreement about the nature of moral language. Error theorists may not even endorse revisionary ways of speaking, while expressivists and constructivists don’t agree with realists about the meaning of ordinary moral claims, and so they would grant that they’re revising anything.

Brennan continues:

But we don’t have to play along. The following is a valid argument: If moral antirealism is true, then it is not true that the holocaust was bad. The holocaust was bad, therefore moral anti-realism is not true.

Play along? Again, the implication is that antirealists are up to no good, as if antirealists are dishonest and want to dupe others by playing some sort of insincere game. We’re not trying to trick Brennan. We disagree. And we sincerely disagree. Why does Brennan think we’re playing tricks or don’t want to “admit” to something Brennan thinks is the case? What are these claims based on?

Lastly, Brennan says that “The following is a valid argument” then presents us with the following:

P1: If moral antirealism is true, then it is not true that the holocaust was bad.

P2: The holocaust was bad [...]

C: [T]herefore moral anti-realism is not true.

This is a modus tollens. Of course it’s valid. This is also a valid argument:

P1: If moral realism were true, dinosaurs would still exist.

P2: Dinosaurs don’t still exist.

C: Therefore moral realism is not true.

Whether an argument is valid or not tells us very little about whether the argument is sound. Proclaiming one’s argument valid as though this had any dialectical relevance makes about as much sense as claiming you don’t have to “play along” with someone else’s position because you can speak in grammatical sentences.

It’s unclear how the validity of Brennan’s argument relates to the claim that realists don’t have to “play along” with antirealists, and it’s unclear why Brennan thinks merely providing a valid argument achieves much of anything.

For what it’s worth, this is one of those syllogisms that, when properly disambiguated, is either sound but trivial or unsound. I’ve critiqued similar arguments here. There, I consider this argument against illusionism:

P1: People sometimes feel pain.

P2: If illusionism is true, no one feels pain.

C: Illusionism is false.

This argument is not good. Why? Because either what is meant by “feel pain” just is in a sense that presupposes phenomenal states, in which case the argument amounts to saying that there are phenomenal states, so if a view were true according to which there were no phenomenal states, then there would be no phenomenal states, then such a view must be wrong. We can similarly disambiguate Brennan’s argument to show that it is either vacuous or unsound.

P1: If moral antirealism is true, then it is not true that the holocaust was bad.

P2: The holocaust was bad.

C: Therefore moral anti-realism is not true.

The only way that it would be the case that if moral antirealism were true, that nothing would be (morally) “bad” would be in the sense of being stance-independently bad. And moral antirealism just is the view that nothing is stance-independently (morally) good or bad. But if the first premise is cashed out this way, we get something like:

P1: If the view according to which nothing is stance-independently bad is true, then it is not true that the holocaust is stance-independently bad.

P2: The holocaust is stance-independently bad.

C: Therefore, the view according to which nothing is stance-independently bad is not true.

We have to avoid equivocation, so we have to plug in the same meaning to “bad” in P2 as we do for P1: stance-independently bad. Yet the result is a vacuous argument that relies on a premise according to which there are stance-independent moral facts. If, instead, we use some other sense of “bad” that doesn’t require stance-independence, then it’s not necessarily the case that if moral antirealism is true, then the Holocaust isn’t “bad.”

What we get here is a dilemma:

If by “bad” we mean something that would ensure the first premise is true, this also ensures the second premise begs the question against the antirealist, and an antirealist would be sure to reject P2: Of course the antirealist isn’t going to accept P2 if P2 just is the claim that something is stance-independently bad, as this is logically inconsistent with antirealism!

If, instead, by “bad” we mean something consistent with antirealism, then antirealism won’t necessarily entail that nothing is “bad” in the relevant sense, and the antirealist is free to reject P1.

In other words, any disambiguation of this argument leaves the antirealist in a position to either reject the first or the second premise with little issue. This argument is either outright false because P1 is false, or presupposes the very thing that antirealists deny. There is little intellectual value in presenting arguments like this. They have no dialectical force. At best, they might serve rhetorical purposes by exploiting ambiguity, but presumably that isn’t why philosophers present them. For those who do present syllogisms like these, I’d like to hear from them as to why they do so.

5.0 Conclusion

Brennan’s characterization of antirealism is fairly common among moral realists. I hope moral realists will see posts like mine and consider that there may be more clearer and more productive ways of characterizing and criticizing antirealism. Even if realists ultimately maintain that antirealism is indefensible and that antirealists are bad people, I’d like to see the case for such claims fleshed out more, so that we’d have something of greater substance to engage with.

Endnotes

1. Most contemporary analytic philosophy is carried out in English. To the extent that the English language differs from other languages and shapes, in extremis (e.g., when doing really abstract theoretical work in academic philosophy, not just talking to a person on the street), the way people think about a given issue, focusing entirely on English may be a profound mistake among analytic philosophers. That’s too much of a digression to get into here, but I want to flag that contemporary analytic philosophy is absolutely obsessed with language, is almost exclusively carried out in one language and in one dominant cultural landscape (the Anglophone world), and yet its proponents often treat their thinking and methods as though they transcend culture and language. Maybe they do, but this would be a task one would hopefully be capable of demonstrating empirically, not just assuming on the basis of armchair principles.

2. Moral antirealism is the view that there are no stance-independent moral facts. Such a view does not require you to believe that nothing is morally bad, or that there’s nothing immoral about genocide, or that it isn’t wrong to torture people for fun. It does not require you to take any particular normative moral stance towards anything. Moral antirealism is exclusively a view about the nature of moral truth: namely, it is a rejection of the notion that there are facts about what is morally right or wrong that are true independent of anyone’s stances. That’s it. It has no further necessary implications.

A first-order, or normative moral stance is a stance about what is morally right or wrong, or good or bad, morally permissible or impermissible, and so on. “It is wrong to torture people for fun” is a normative moral stance. Antirealist positions turn, centrally, one how to interpret utterances of this kind. The standard antirealist positions are error theory, stance-dependent cognitivist views, and expressivism/noncognitivism.

Error theorists agree with moral realists about what people are trying to do when they make these claims. Error theorists hold that people are attempting to describe stance-independent moral facts. But, since there are no such facts, all such claims are false.

Stance-dependent cognitivist views hold that moral claims can be true or false, but in a way that is made true by our stances. An individual relativist, for instance, might hold that moral claims like “stealing is wrong” are best understood to mean something like “I disapprove of stealing.” Such statements can be and often are true.

Expressivists/noncognitivists vary from crude and straightforward approaches to more nuanced ones that attempt to vindicate much of the allegedly realist-seeming nature of ordinary moral language, but nevertheless maintain that moral claims are not primarily used to express propositional claims that can be true or false, but are instead used to express nonpropositional attitudes: emotions, imperatives, and so on.

Note that in all three cases, the denial that an action is stance-independently right or wrong is completely orthogonal to one’s attitude, feelings, and motivations. An error theorist may agree that nothing is “really” right or wrong, but this tells us nothing about that person’s attitudes, dispositions to act, or general character. They can act exactly like a moral realist without any inconsistency in every respect that could plausibly matter in everyday life. Denying that torture is stance-independently wrong doesn’t imply that one is any less opposed to or disgusted by torture, or any less motivated to try to stop it.

The same holds for expressivism and for most stance-dependent theories (agent relativism is one notable exception). Aside from agent relativism, which does carry weird normative implications, most traditional forms of moral antirealism are consistent with having any set of attitudes, feelings, and motivations such that the antirealist could be just as motivated to stop torture, genocide, and so on as a moral realist. There is no inconsistency or contradiction in an antirealist having and acting on such motivations at all.

References

Franzén, N. (2024). The presumption of realism. Philosophical Studies, 1–22.

Gowans, C. Moral relativism. In E. N. Zalta (Ed.) The Stanford encyclopedia of philosophy. https://plato.stanford.edu/entries/moral-relativism/

Sæbø, K. J. (2009). Judgment ascriptions. Linguistics and Philosophy, 32, 327–352.

Sinclair, Neil. 2012. Moral realism, face-values and presumptions. Analytic Philosophy, 53(2): 158–179.