Myths about relativism: A response to Nobis
Myths about relativism: A response to Nobis
TikTok Tuesday #1
1.0 Nobis’s question about relativism
In this TikTok video, Nathan Nobis poses a question about moral subjectivism/relativism. Nathan says:
If you can help me figure out what people mean when they say something…suppose somebody says ethics is subjective or relative. If you ask that person ‘well alright what do you mean and what comes from that, or what results from that…like what are they going to say? Like that anybody should be able to do whatever they want, or anything any culture does is fine, or everybody’s views are sort of equally good or plausible or reasonable or whatever or what. I mean, what does it really come down to if somebody says ethics is subjective or relative?
My goal in this post will be to clarify the various dimensions and formulations of “relativism,” distinguishing four key orthogonal dimensions:
Ordinary vs, technical use of “relativism” and “subjectivism”
Relativization target
Metaethical relativism vs. “pseudo-relativism” (normative and descriptive relativism)
Indexical target
2.0 Ordinary vs, technical use of “relativism” and “subjectivism”
It is an empirical question what any particular person is intending to communicate when they use the terms “subjective” or “relative.” And people probably do sometimes invoke these terms in contexts where they convey a variety of different claims, sometimes bundled together in a confused and messy way. So nonphilosophers may mean a variety of different things when they invoke terms like ‘relative” and “subjective.” Philosophers use both terms in technical contexts, as well. Their meaning likely differs from instances that take place outside a technical context.
It’s not entirely clear whether Nathan is asking about these technical uses or ordinary usage. If the latter, it’s not clear whatever it is people use these terms to mean will be shared across speakers, or that whatever they mean will be all that philosophically interesting. It’s also worth noting to that to the extent that nonphilosophers invoke these terms, such invocations may reflect a kind of backflow from philosophy itself; that is, they are not your standard terms that figure into ordinary language, like “tasty” or “cool,” but a kind of quasi-technical recycling of technical terminology, much as nonscientists may use terms like “chemical” or “physics” in ways that relate to technical usages among the sciences but plausibly differ in everyday use.
Nonphilosophers may mean a variety of confused, muddled, and questionable things when they invoke these terms. If Nathan’s goal is to draw attention to this, then it is a worthwhile endeavor. However, it would have been helpful if Nathan were more clear about the dialectical objectives of the question. Since Nathan defends moral realism and critiques antirealist views, Nathan presumably rejects metaethical relativism in its more technical form, and isn’t merely interested in criticizing ordinary, non-technical uses of these terms. Given this, it would be in the interests of Nathan’s viewers if Nathan were clear about whether the target was ordinary or technical uses of these terms.
More generally, I want to note a kind of asymmetry in the way philosophers handle certain kinds of arguments. Philosophers will often make a point of criticizing “student relativism” or a kind of colloquial, folk relativism. Such forms of relativism are often easy targets: they are muddled, unsophisticated, and often internally contradictory. Once this is pointed out, the critic will then present realism as an alternative (why not some more defensible form of antirealism?).
This is an unfortunate way of framing metaethical disputes. It’s not as though when I criticize moral realists I find some muddled lay conception of realism, replete with bad metaphysics and confusions about language among nonphilosophers, critique that, then supply antirealism as the solution. I critique what philosophers say, which (I argue) often recapitulates, amplifies, or constructs altogether novel confusions about language and metaphysics. Why, then, do philosophers focus so much on folk realism rather than versions of realism that philosophers have defended?
3.0 Relativization target
With respect to terms like “relativism” and “subjectivism,” philosophers sometimes use the terms interchangeably, and sometimes draw a distinction between them. I prefer to use the term “subjectivism” to refer to a subset of metaethical relativist views which identify moral facts with the standards of individuals, and “relativism” as a broader category that includes both subjectivism (or “individual subjectivism”), and other forms of relativism, most notably cultural relativism, which holds that moral claims are true or false relative to the standards of different cultures. For our purposes, I will use the term “relativism” unless otherwise specified, for simplicity’s sake, to refer to the superordinate category, and subjectivism to refer to a form of relativism.
This distinction alludes to one of the key features of relativism: relativization target. All forms of (metaethical) relativism allow moral claims to vary in accordance to the moral standards to which they are relativized. That is, for any given moral claim, e.g., “hurting people for fun is morally wrong,” such a claim may only be judged, for relativists, as true or false relative to one or another moral standard. So this raises the question: which moral standards?
The two most popular standards are the standards of individuals and the standards of cultures. This yields the two standard relativization targets for moral relativism:
(Individual) subjectivism: Moral claims are true or false relative to the standards of individuals
Cultural relativism: Moral claims are true or false relative to the standards of cultures
For subjectivists, this means that if Alex and Sam are members of the same culture, and Alex thinks stealing is wrong, but Sam does not, then they are both correct relative to their own moral standards. Moral truth is not determined by their culture’s standards, but by each individual’s moral standards.
For cultural relativists, if Alex and Sam are members of the same culture, then claims about whether stealing is wrong, when made by either of them, will either be true or false relative to the standards of their shared culture. This thus technically allows moral disagreement to exist between members of the same culture.
It is possible to relativize moral standards to other standards as well, such as the standards of religions or species. It is even logically possible to relativize moral claims to bizarre standards, like one’s geographic location, age, what time or day of the week it is, one’s astrological sign, and so on. I’ve never heard of anyone endorsing views like these, but it’s possible to do so.
The most common standards to which a moral relativist may relativize moral claims will be the stances of individuals or groups. This means that standard forms of moral relativism are also forms of moral antirealism because they hold that there are only stance-dependent moral facts, but no stance-independent moral facts.
However, some of the standards moral claims could be relativized to are not stances, e.g., geographic location. If one believes that a moral claim is true when uttered on earth but false when uttered on Mars, the truth of such claims is relative, but it is not relative to anyone’s stances. This would technically be a form of moral realism, since moral claims would have stance-independent truth values. Again, this is super weird, but it reveals that relativism is, technically, consistent with realism. This works in reverse, as well: not all stance-dependent theories are forms of relativism. Ideal observer theory holds that the truth of moral claims is determined by what an ideally rational and fully informed agent would endorse under optimal circumstances. These accounts make the truth of moral claims dependent on the stance of a hypothetical agent. As such, it is a form of antirealism. However, moral claims on this account are not relative: there is a single correct standard of moral truth. This illustrates that stance-dependence and relativism are, technically, orthogonal to one another. This distinction is beautifully explained by Joyce in this excellent supplement in the Stanford Encyclopedia of Philosophy.
A lot of people seem to be unaware of this, so let me stress this explicitly:
Relativism does not entail antirealism. Not all forms of moral relativism relativize moral claims to stances, so not all forms of moral relativism deny that there are stance-independent moral facts. Likewise, not all accounts which hold that there are stance-dependent moral truths are forms of relativism, since some of these theories only allow for a single correct moral standard. As a result, they do not allow the truth status of moral claims to vary relative to different moral frameworks.
4.0 Metaethical relativism vs. pseudo-relativisms
So far, what I have been describing is a metaethical position, i.e., a position exclusively about the nature of morality itself. Metaethical relativist positions consist primarily of a semantic thesis about the meaning of moral claims. In particular, they hold that moral claims are propositions with an (usually implicit) indexical element, such that the truth or falsity of the claim can vary as a function of the standards to which it is relativized. We already use indexical terms all the time, e.g., “I,” “you,” “me.” When I say, “I am Lance,” the term “I” refers to me. Since I am Lance, this statement is true. Yet if you say “I am Lance,” and you are not Lance, this statement is false. Thus, the exact same statement, “I am Lance,” is true when someone who is Lance says it, and false when someone who is not Lance says it. This is because the term “I” refers to whoever is using it. Since whoever is using the sentence can vary, “I” varies in who it refers to. Since something that is true of one person may not be true of another, the inclusion of “I” allows the truth status of sentences featuring indexicals to vary.
4.1 Metaethical relativism
Metaethical relativism allows the truth status of moral claims to vary in exactly this way. It’s just that the the indexical element is implicit, rather than being explicitly indicated by the use of a term that functions in an an explicitly indexical role, like “I” or “you.”
Thus, metaethical relativism treats a moral claim like:
it is morally wrong to hurt people for fun
as having an unstated indexicalization. For subjectivists, this will be the speaker’s own moral standards. Such a statement thus means something like:
it is morally wrong to hurt people for fun according to my standards
I consider it morally wrong to hurt people for fun
Note that in these cases, the truth of the claims in question is often fairly trivial and uncontroversial. If someone considers it wrong to hurt people for fun, and they say “I consider it wrong to hurt people for fun,” even a realist would have to agree that this is true. A relativist simply maintains that this is all there is to moral claims of this kind (but see below for agent relativism, a special kind of quasi-relativism). As a metaethical position, metaethical relativism cashes out moral claims in purely descriptive terms: they consist of reports about the speaker’s moral standards, or their culture’s standards, or whatever, and nothing more. It has no normative implications at all.
Here’s the problem. Not everyone uses “relativism” (or “subjectivism”) to refer to a purely metaethical position. Instead, they often use such terms to instead refer to normative or descriptive claims, or, worse, bundle metaethical and non-metaethical notions into their use of the terms in ways that have resulted in immense confusion and misunderstanding.
There are two other major forms of “relativism.” They are types of “relativism” in name only: they are completely different from metaethical relativism, and have nothing to do with the truth status of moral claims varying as a result of having an indexical element. I think it’d be best to consider these, not as forms of moral relativism, but as their own thing (like David S. Pumpkins). I think it is best to think of these as forms of pseud-relativism. Unfortunately, people frequently mix up metaethical relativism with normative and descriptive “relativism,” conflating metaethical, normative, and descriptive theses into a single muddled “position.”
This distinction between a descriptive (or empirical) position, a normative position, and a metaethical position is explicitly described by Gowans (2021) in the the Stanford Encyclopedia Entry on moral relativism. This isn’t my distinction, and thus some potentially idiosyncratic way of drawing conceptual boundaries. As Gowans puts it, “[...] in moral philosophy ‘relativism’ is usually taken to suggest an empirical, a metaethical, or a normative position.”
4.2 Normative relativism
Normative relativism is (roughly) the normative moral position that one ought to respect the moral standards of other individuals or cultures. This may result in a moral prohibition against judging other people or cultures by your own moral standards, imposing one’s moral standards on individuals or cultures with different moral standards, on refusing to intervene in conflicts with people with different moral standards, or in some other way entail some sort of normative facts about how you should or shouldn’t treat people with different moral standards.
It is important to stress that this is not a metaethical position. It is a normative moral position. The distinction is, between a second-order and first-order moral claim, respectively. A metaethical, or second-order moral claim is a claim about morality, moral claims, and so on, whereas a normative, or first-order moral claim is a “direct” claim about what is or isn’t morally good or bad, right or wrong, and so on. Here are some examples of each:
Metaethical / second-order moral claims
There are objective moral facts
It is possible to have knowledge of the moral truths
When people make moral claims, they are expressing nonpropositional content
Normative / first-order moral claims
It is morally wrong to hurt people for fun
It is good to be courageous
You shouldn’t lie without a good reason
Normative relativism results in claims like
You should not impose your moral standards on people or cultures with different moral standards
You shouldn’t judge people with different moral standards by your own moral standards
You should respect people with different moral values
As you can see, statements (7)-(9) are more like statements (4)-(6), and less like statements (1)-(3). Normative relativism is a view about what you should or shouldn’t do, what kind of person you should be (i.e., tolerant, respectful), and what’s good or bad. It is not a view about the nature of morality itself. It is not a metaethical position. It is not really a form of “relativism” at all.
4.3 Descriptive relativism
Descriptive relativism is simply the empirical observation that different individuals or cultures have different moral standards. This has nothing to do with metaethics or normative ethics. It’s simply a descriptive claim about human institutions and human psychology. Nevertheless, people will sometimes say that moral relativism is true, and by this they are merely observing that moral standards vary between individuals and groups. This is likewise not really a form of relativism.
5.0 Indexical target
The final distinction is one of the most important. One of the most common objections to moral relativism is that it carries certain undesirable implications. In particular, people think that if you endorse moral relativism, you must tolerate or even approve of other people doing anything they want, no matter how monstrous you yourself might find it. I will call this the “approving evil” objection.
5.1 The approving evil objection to relativism
According to the approving evil objection, we should reject moral relativism because it entails that we would be required to approve of actions that are manifestly evil, or morally wrong.
For instance, one might interpret relativism as entailing something like the following:
Suppose you are an individual relativist. If so, you face a serious problem. You think that moral standards are only good or bad relative to each individual’s standards. As a result, this means you think that, no matter what other people’s moral standards are, if they think it’s good for them to do something, then it is good for them to do it. If Alex thinks it’s okay to torture people, then it’s good for Alex to torture people. This poses a serious problem for the relativist: if Alex wants to torture them, they have no right to object. After all, it’s good for Alex to torture them. So it would be morally wrong for them to protest or attempt to resist Alex’s efforts to capture them and drag them off to be tortured. More generally, the relativist must think that whenever Alex tortures anyone, that it is morally good for them to do so. The relativist is thus committed to thinking that so long as someone thinks it’s okay to torture people, it is okay for them to torture people. This is repugnant and absurd. Moral relativism is a horrific position that requires us to be okay with torture.
This isn’t a good objection to “relativism.” This is because, at best, it fails to disambiguate different forms of relativism. In particular, it fails to disambiguate the distinction between agent and appraiser relativism.
According to agent relativism, moral claims are true or false relative to the standards of the agent performing the action (or that agent’s culture, etc.).
According to appraiser relativism, moral claims are true or false relative to the standards of whoever is evaluating the action in question (or the standards of the appraiser’s culture), even if they are not, themselves, performing the action.
First, the approving evil objection would apply to normative relativism. But recall that this is a normative moral view, not a metaethical position, and it is a form of pseudo-relativism. Of course it trivially follows of any normative view according to which one must tolerate, respect, or approve of specific actions, such as torture, that one must tolerate, respect, or approve of those actions. And of course, if we are antecedently committed to the moral stance that such positions are wrong, and that we should not tolerate, respect, or approve of these actions, we would find ourselves in a contradiction. Yet this poses no threat to metaethical relativism, since metaethical relativism isn’t a position on what’s morally right or wrong in the first place.
Second, the approval evil objection might apply to agent relativism in particular. Here’s why. The agent relativist may hold that it is the standards of an individual (or their culture) that fix what is morally right or wrong, good or bad, for that person. If the agent relativist maintains that because it is the right thing, or the good thing for that person to do full stop, in virtue of the fact that it is consistent with their moral standards, such that the rest of us are obligated to behave accordingly, and regard it as what they should do or what is good for them to do ourselves, that person’s moral standards fix the moral facts in such a way that they have normative implications for everyone else. For instance, if Alex thinks that it is okay to steal, then this makes it true that it is morally good for Alex to steal, not simply from Alex’s point of view, but it is true in such a way that it has normative and evaluative entails that everyone else is obligated to comport themselves with, independent of their own moral standards, attitudes, desires, or preferences, including standards, attitudes, desires, or preferences about how Alex should act or what it would be good for Alex to do.
In other words, Alex’s moral standards “fix” the moral facts in such a way that they result in a set of moral standards that may be stance-dependently true, but don’t depend on your moral standards, or the standards of your culture. You are effectively obligated to treat that person such that their behavior is right or wrong, good or bad, and so on, independent of your own stance or the stance of your culture. This effectively results in a kind of de facto relativism: there are facts about what you should or shouldn’t do, and what’s good or bad, that you must comply with, independent of your goals, standards, or values. Those facts aren’t made true by mindless forces, but are instead made true by someone else’s moral stance, but they nevertheless bind and have authority over you in exactly the way moral realists think moral facts have authority over us and bind us.
I’m a moral antirealist. Central to my denial of moral realism is my objection to the notion that there are some special set of normative facts that I “must’ comply with, or that I “should” act on, that are not reducible to descriptive facts about my goals or desires. If agent relativism results in normative moral facts that exhibit an analogous form of authority over me, it would result in a “metaethical” position that has features that are functionally identical (or very similar) to those features that motivate one of my objections to moral realism. Agent relativism functions as something like a kind of moral realism where each person or culture is subject to a distinct set of rules everyone else is obligated to respect. This may technically be a form of antirealism: stances fix the moral facts, but insofar as stances create obligations everyone else is obligated to comply with independent of their stances, it functions as a kind of de facto realism. It is a kind of quasi-realism with a strange constructive procedure whereby individuals or cultures “create” moral facts that have normative implications for others that wield similar authority to realist accounts. This is why I’d prefer to call it a la carte realism.
Furthermore, note that if agent relativism carries normative implications of this kind, it is not really an exclusively metaethical position at all, and is thus not, technically, a form of metaethical relativism. Instead, it is a hybrid metaethical-normative moral account. Ironically, too, the most objectionable features of this account are its normative implications.
Appraiser relativism, in contrast, is not subject to these concerns. Appraiser relativism cashes out as the following:
Moral claims express relational facts about the standards of a speaker or their culture, e.g. “murder is wrong” = “I consider murder wrong.”
This indexical element allows the truth status of moral claims to vary relative to the person or group the claim is indexed to
These relational facts are descriptive and, conditional on the truth of the semantic thesis, trivial, e.g., if “murder is wrong” = “I disapprove of murder,” and the person does disapprove of murder, this statement is trivially true, and a realist has no legitimate basis for rejecting the claim.
The appraiser relativist has no obligation to respect anyone else’s moral standards because the view has no direct normative moral implications at all. For comparison, consider food preferences. I suspect most readers aren’t “gastronomic realists” that think there are stance-independent facts about what you should or shouldn’t eat, or what tastes good or bad, regardless of your own preferences, attitudes, or experiences with food. You may, as a result, find something like gastronomic appraiser subjectivism quite plausible: claims like “chocolate tastes good,” can be true or false relative to the attitudes of whoever is expressing such a judgment, and
I’m not an appraiser relativist. There are objections to the view, as well. However, it isn’t vulnerable to the juicy objections that critics heap on agent relativism. It is suspicious, then, that critics of “relativism” opt to focus so much on agent relativism and less so on appraiser relativism.
This propensity to hate on relativism is part of a broader pattern in the way some moral realists criticize moral antirealism: by denigration, innuendo, and disparagement. Here are a few objections you’ll see, even from professional philosophers, though commonly among nonspecialists engaged in debates “in the wild”:
The approval of evil problem
Relativists can’t object to others
Relativists can’t judge others
Relativism leads to contradiction
If relativism is true, nothing really matters
Such efforts have, unfortunately, proven quite successful. Realists have successfully framed the dialectic as one in which moral antirealists are often convincingly depicted as being more likely to do immoral things, tolerate moral things, to lack purpose, value, or meaning, and so on. Relativists can easily respond to all of these objections, which are all astonishingly weak. Yet they still persist. The responses to these objections are so straightforward and so definitive I consider the matter effectively settled with respect to these objections. I’ll quickly cover how silly (2)-(4) are just to illustrate the point.
5.2 Relativists can’t object to other people’s actions
Some critics will insist that if you’re a moral relativist, you can’t object if someone else harms you, threatens you, or mistreats you. This might only be true if the “relativist” were an agent relativist, i.e., an a la carte realist. This would not apply to appraiser relativism. If critics want to level this objection to a la carte realism, they should feel free to do so, but this objection has no force against appraiser relativism. For an appraiser relativist, judgments like “it’s wrong for you to harm me,” and even “it’s wrong for anyone to harm anyone,” are true relative to the speaker’s moral standards. Such standards can and do apply to other people’s conduct.
The appraiser relativist is not obligated to respect other people’s moral standards, unless, and only unless, doing so is a feature of their normative moral standards; they do not have to do so in virtue of the truth of appraiser relativism itself.
5.3 Relativists can’t object to other people’s actions
Some critics will insist relativists can’t judge other people’s actions, enforce their moral standards on others, and so on. One again, this would at best only be true for agent relativism, not appraiser relativism. Since it is simply and straightforwardly not an implication of appraiser relativism, it could not be used as an objection to appraise relativism.
5.4 Relativism is self-contradictory
Some critics think you can reject relativism because it allows contradictions, e.g., the statement “lying is wrong” could be both true and not true. Since this would result in a contradiction, relativism is absurd.
This is the worst objection to relativism. Relativism no more results in “contradictions” than statements like “I am in Brazil” are contradictory. If someone in Brazil says “I am in Brazil,” the statement is true, while if someone who says “I am in Brazil” is not in Brazil, it’s false. Should we conclude that everyday indexical language is absurd? No.
Critics who use this objection may have in mind some kind of a la carte realism, and mistakenly think that if Alex thinks lying is wrong, then it “really is wrong,” and if Sam thinks lying is not wrong” then it “really isn’t wrong.” Since this would result in a contradiction, relativism is a stupid position. This objection is so bad it doesn’t apply to agent or appraiser relativism. The agent relativism would at least hold that if Alex thinks stealing is bad and Sam doesn’t, that it is bad for Alex to steal but not for Sam to steal.
5.5 If relativism is true, nothing really matters
This is the most subtle of anti-relativism myths. The idea here is that if you’re a relativist, then things don’t matter in a way, that, well, matters. Unfortunately, this objection begs the question on what it would mean for something to “really” matter. If “really” mattering just requires realism, then sure, nothing “really matters” for a relativist. But this is just a rhetorically loaded way to say nothing stance-independently matters. This is hardly an objection: this is simply a restatement of what the relativist denies! This is like objecting to atheism on the grounds that the atheist “denies the existence of God.” I discuss why this objection is terrible elsewhere, including here and here.
5.6 Please stop with the terrible objections to relativism
Simply put: the philosophical community should recognize these objections as terrible, and discard them. Yet even professional philosophers employ these arguments. Given how easy it is for a relativist to respond to these objections, and how definitive those responses are, I will no longer treat them as objections, but as simply myths, misrepresentations and, when their proponents ought to know better, the persistence of willful ignorance on the part of their exponents. It is time antirealists stand up to the lazy and asinine rhetoric people online routinely employ against antirealist positions like relativism.
For whatever reason, critics of relativism and other antirealist positions often employ weak objections that seem like they were memorized at a sermon or handed out in tracts by a street preacher. Tom Gilovich draws an important distinction between the level or quality of evidence we demand for views we want to believe, versus those we don’t want to:
Can I believe it?
Must I believe it?
Many critiques of antirealism seem more like the former. They pass the simple, low bar of looking superficially like a convincing objection to antirealism. The bar suddenly becomes much higher for critiques of realism. Motivated reasoning is everywhere, and I am no exception. Yet when a particular mistake is repeated often enough, it crystallizes and becomes a kind of self-perpetuating myth maintained by the sheer volume of believers and proponents mutually reinforcing it. Much as it takes a critic to point out, openly, that the emperor has no clothes, I want to make a similar point here: I am not simply criticizing moral realism and defending moral antirealism. I’m making a meta-point about the state of the dialectic: it’s absolutely terrible. Arguments on both sides (antirealists included) are feeble, shriveled little things, mutually incapable of convincing the other side. This is only to be expected, as both arguments for and against realism consist of what few drops of moisture one can scavenge in the dialectical wasteland that is contemporary analytic metaethics. If we want to resolve these issues, we need to leave that desert, and seek verdant land elsewhere.
6.0 Nobis’s responses
It looks like my foray into metaethics on TikTok is in full swing:
(1) I posted this response to Nathan
(2) Nathan responded
(3) I responded to that response
I know I tend to emphasize criticism and people sometimes don’t react well when someone like myself just presents endless critiques, so it’s encouraging to see we appear to have (to me, at least) some common ground.
First, Nathan notes that I didn’t provide much of a definition of subjectivism. I hope the preceding sections rectify that remark.
Part of the reason I’m hesitant to offer “a” definition of subjectivism is that I don’t think there is singular, well-defined notion picked out by the term “subjectivism”. “Subjectivism” can be used to refer to various distinct, technical notions in contemporary analytic metaethics, it could be used to refer to technical concepts outside of analytic metaethics, and it could be used in everyday discourse to refer to a variety of often jumbled, semi-overlapping and sometimes internally inconsistent ideas.
Whatever students or nonphilosophers in general mean when they use the term would be a challenging empirical question, and not one I think could be easily resolved in classroom settings. To develop on this point, I want to emphasize two main points (I make similar remarks in the last of the three videos linked above):
First, the test cases Nobis presents, especially if they are presented publicly in a classroom setting, are not a good way to determine whether people are subjectivists.
Second, questions about whether people are subjectivists are empirical, and there is at least some empirical evidence that finds high rates of subjectivism/relativism (and antirealism more generally) among populations sampled in the US.
6.1 Classroom tests aren’t that informative about folk subjectivism
The proportion of nonphilosophers who endorse subjectivism is a challenging empirical question that cannot be resolved by informal classroom surveys. For a variety of reasons, classrooms are not an appropriate place to gather such data:
Students in college courses are not psychologically representative of people in general. So whatever findings we observe among student populations, even if we employed valid measures, would tell us little about how people in general think. See here for an extended critique of the tendency for philosophers to generalize from their own judgments (similar concerns apply to anecdotes). Students aren’t representative of the nations they come from, and the nations, cultures, and general populations most analytic philosophers are exposed to the most are psychologically unrepresentative of most of the rest of the world's populations. Students taking philosophy classes are also likely to differ from people who don’t take philosophy classes in ways that are relevant to how they respond to questions about philosophy.
Furthermore, philosophers themselves are inducted into ways of thinking that can introduce biases and prompt them to interpret their interactions in ways that funnel them towards particular conclusions about how nonphilosophers think. There may also be selection effects that cause certain people to be more likely to study philosophy than others, amplifying any potential training biases by selection effects.
Philosophers may also bias students towards certain kinds of responses by framing questions in certain ways (such as providing students with explicit response options, instead of asking more open-ended questions). Class dynamics can change the psychology of students in ways that differ from the general population. If students are reading philosophical texts prior to class, or are in the middle of a lecture, or are given novel thought experiments they never engaged in immediately before a question, they can be primed to theorize or reflect about a philosophical issue in ways that cause them to form views they didn’t have prior to reflection, a process I call spontaneous theorizing. Students in these circumstances may think or at least report thinking things that systematically differ from how they and others would respond or think outside philosophy classes.
However, the social setting is likely the most important factor threatening the reliability of the anecdotes philosophers provide about classroom discussion. If a student states that they are a subjectivist or relativist, and you ask them something like “well then, if someone thought it was okay to [insert atrocity], would it be okay for them to do it?” you may find that a student says “no,” thus seeming to show that, on reflection, they don’t really endorse subjectivism after all. This might be the case. However, there are a few other potential factors in play.
First, in a pedagogical context, questioning the student in this way can work much like leading a witness. When a student says something, and an instructor says “well, would you still think that if [...]?” this can give the impression that the student “ought to” change their mind. Students may appear to relent and say “no” in these circumstances to give an answer they expect an instructor to approve of, or to not appear foolish or uncooperative. Note that Nobis suggests part of the reason why students appear to endorse “subjectivism” in the first place is for social reasons. Such social objectives are not isolated to endorsing subjectivism, they emerge in other contexts as well, including instances in which instructors challenge students. I suspect you could get a fair number of students to appear to change their views even if you gave them irrelevant thought experiments, bad examples, or even asked them completely incoherent questions.
More generally, what students say in a classroom context, both before and after questioning by an instructor, may or may not reflect what they’d say, think, or do outside that context. This may even further undermine the case for folk subjectivism; my point here is simply that what students say in classes may tell us very little about how they think or what they do outside those contexts
Also, the kinds of examples that are often used to jostle students out of endorsing subjectivism could, when presented in public, give the impression that the student is permissive of, or even approves of horrific actions that they don’t approve of, in public, in front of others. Students may not have a well-developed sense of the distinction between agent and appraiser relativism, and may be drawn, in some inchoate form, to something more like appraiser relativism, not agent relativism. Yet when pressed, they may lack the conceptual resources, especially under severe social pressure, to draw the distinction and respond accordingly.
This can lead students to feel that they must recant their endorsement of subjectivism, or they will appear to fail to condemn or disapprove of horrible actions, even though doing so is completely consistent with certain forms of subjectivism (e.g., appraiser subjectivism). In short, students may suffer a kind of “dialectical short-circuiting” where they endorse a position that actually could be defended, and they simply don’t know how to defend it. If an instructor is unsympathetic to viable defenses and rebuttals available to subjectivists/relativists, they may unfairly take student concession to be an indication that the student is an intuitive realist after all, rather than simply a confused subjectivist who doesn’t want to look like an asshole in front of the whole class.
6.2 It’s an empirical question
Armchair analysis of how nonphilosophers speak or think about moral issues is simply inadequate to provide robust, reliable information. The best way to study how nonphilosophers think about philosophical topics (if they think about them at all) is to employ a variety of mutually corroborating forms of empirical evidence from fields such as psychology, anthropology, and linguistics.
At present, available empirical evidence simply does not provide much compelling indication that most people speak or think like moral realists. This, at least, is my conclusion after comprehensively viewing the literature and conducting studies of my own on the topic, discussed here.
References
Gowans, C. (2021). Moral relativism. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2021 ed.). https://plato.stanford.edu/archives/spr2021/entries/moral-relativism/