Are moral values like taste preferences?

Twitter Tuesday #47

Photo by Carissa Gan on Unsplash

Sometimes I describe my moral standards as preferences. Critics of this view often insist that this isn’t adequate because other sorts of preferences, like food and music preferences, differ from moral stances.

I agree. Moral values do differ from other preferences. Usually. I think the boundaries between moral and nonmoral preferences are often blurred, as I argue here. But how do they differ? Critics of people who construe moral values as preferences often imply or explicitly insist that moral standards are treated in a way that would only make sense if we were realists about our moral values, and that, as such, moral values can’t be preferences. Checkmate, antirealists! Consider this exchange over on Twitter:

Hannah asks a great question. I would’ve preferred a direct response from Nathan, but this is what we have instead. I suppose it’s intended to be a sort of evidence, but if so it’s a fairly obscure and presumptuous sort of evidence. Nathan doesn’t outline the precise ways moral standards differ from food and drink preferences, and this exercise relies on the respondent arriving at the same conclusions as Nathan, which they may not.

With respect to the question itself: do my taste preferences differ from my attitude towards school shootings? Sure. But none of the ways they differ have anything remotely to do with realism, e.g., perhaps I’m expected to think school shootings would be wrong regardless of my preferences or attitudes, while this would make no sense for my preference for strawberries over blueberries. But that difference isn’t present in my appraisal of food preferences and moral considerations. Both seem to me to be a kind of preference, albeit ones for which I have very different attitudes and reactions.

I’ve seen realists draw attention to alleged differences between preferences and moral standards many times, but I suspect moral realists are being a bit hasty and a bit unimaginative. Sure, we typically treat our moral values differently from our taste preferences. But treat them differently how? Here are some possibilities:

• We treat our moral values as claims about what is stance-independently right or wrong, while our taste preferences merely express our own stances or attitudes.

• Our moral values concern other people’s conduct (often everyone), while our taste preferences are typically limited in scope to our own conduct.

• Our moral values matter a whole lot to us, while our taste preferences matter comparatively less.

• Our moral values prompt a particular set of emotional reactions, such as disgust or outrage. Our taste preferences rarely evoke these emotions, or when they do, are highly attenuated relative to our reaction to moral considerations.

There are other possibilities (and if you think of any, please leave a comment; I’d be happy to add it to the list and credit you too). Speaking for myself, the second and third of these are the most relevant: my moral preferences tend to differ from my taste preferences in two ways:

• Scope: My moral preferences are preferences about how other people (often all or most people) act, not just myself.

• Intensity: I care more about my moral preferences than my taste preferences. I might not eat nasty food if you serve it to me, but I would be far more opposed to someone setting me on fire or stealing my car.

These differences do all or nearly all of the work of distinguishing my moral preferences from my nonmoral preferences. In fact, that these are the relevant differences does a better job of explaining my moral and nonmoral values than proposing a realism/antirealism distinction. Why? Because scope and intensity exist on a spectrum and vary in degree, whereas realism/antirealism doesn’t. This better accounts for the way in which moral and nonmoral attitudes and values blur at the edges and overlap with one another.

Some of my moral values aren’t that intense, while some of my taste preferences are quite strong. I have very strong preferences for certain kinds of music over others and it genuinely bothers me that some people like music I consider absolutely terrible, like Drake. Yet I am relatively indifferent to certain actions that I might think are technically morally wrong but unlikely to plausibly cause harm: white lies, keeping an extra sugar packet from a restaurant, that sort of thing. I'm not frothing in righteous fury if someone is a little rude to someone else without good reason. I’m also not actively attempting to stop anyone from doing these things. Not all moral values are imposed on others. Some aesthetic sensibilities very well might be.

Just like taste preferences, some of my moral standards are only applicable to me. Many readers will have heard of people who “live by a code” or abide by certain principles or standards. They may moralize them to an extent, or even a great deal, even if they have no expectation that others abide by those standards. Some vegans and vegetarians may be this way. Some people take this stance on abortion: that they think it’s wrong to get one, but that everyone should have the choice to do so. As a matter of actual human behavior, many people don’t impose their moral standards on others or judge everyone by the same moral standards. Some of our moral standards are personal codes of conduct.

Viewing the distinction between moral and nonmoral concerns as one in which moral concerns tend to be broader in scope and more intense while taste preferences tend to be narrower in scope and less intense fits more naturally with my own way of thinking about matters of morality and taste. I wouldn’t be surprised if this did a better job of accounting for how many other people think, as well. Hannah suggests another difference:

I’m not sure what you mean by “literal preference.” The former kind of preferences don’t invoke my empathy or conscience whereas school children being shot does. How does that translate to moral realism?

I’m not sure what Nathan means by a “literal” preference, either. This does strike me as another difference between these preferences: my preference for strawberries doesn’t prompt empathy while my attitude towards school shootings does prompt empathy. So there are psychological differences in my (and many other people’s) emotional response to these scenarios. It’s a very important difference, too.

Richard Yetter Chappell chimed in at this point, stating:

Taste preferences seem arbitrary (you could just as well flip them around without error); moral preferences seem subject to rational assessment: they may be supported by good reasons or bad ones.

Richard perpetuates the objectionable use of unqualified terms like “seem” that I frequently criticize on this blog. Seems to who? My moral preferences seem to me to be no different from my taste preferences in this regard: they do not appear to be subject to rational assessment. At least, not in a way inconsistent with antirealism. Here’s why. Suppose on entirely stance-dependent grounds I endorse utilitarianism: I prefer that we maximize utility. I don’t believe this is true in some stance-independent respect. Now, there is a separate question of whether any given course of action would, in fact, maximize utility.

Would donating to shrimp welfare maximize utility? That’s an open empirical question that I don’t know the answer to. This consideration is subject to rational assessment even as a moral antirealist because it’s a question about whether some course of action would be conducive to your goals, and such facts are contingent on nonmoral facts, i.e., facts about how much shrimp and humans are capable of suffering, how much shrimp suffer in the way they are currently killed by humans, how effective the charity is at reducing that suffering, and so on. However, it does not seem to me that my non-instrumental values (e.g., my preference to maximize utility) are subject to rational assessment in such a way that I could be in error (though I do think it’s possible to be mistaken about one’s values or preferences, in which case reflecting on whether you do value something is subject to rational assessment; but again this is entirely consistent with antirealism).

I’m not a utilitarian, so let’s instead take the set of all my non-instrumental moral values. That set of values does in fact seem to me exactly like my taste preferences with respect to whether it is subject to rational assessment: it isn’t. Things just don’t seem to me the way they do to whoever Richard has in mind when he says it “seems” a given way. I value my family. I don’t think it makes any sense to rationally assess my reasons for doing so. I don’t need, want, or care about having reasons to care about my family. I just do care about them, and that’s good enough for me. I find the notion that one could or would have any interest in rationally assessing their moral values completely alien.

Hannah doesn’t take this route, but instead says:

Lots of preferences, moral and amoral, are not arbitrary—for example, passing on one’s genes through procreation—and are supported by reasons. I don’t see how that makes them anything more than subjective preferences/desires/values.

I can’t say I disagree. One might very well have reasons for their preferences. But it doesn’t seem to me that preferences require reasons, whether moral or nonmoral. Even so, this shows a route by which one might approach the matter: one could argue that neither moral nor nonmoral values “seem” to be subject to rational assessment (my approach), or one might argue that both appear to be subject to rational assessment and/or that one can provide reasons for those preferences. I don’t endorse the latter approach but it does strike me as more plausible than thinking preferences aren’t subject to rational assessment but moral considerations are. But who cares how plausible things seem to me? How plausible things seem to different people will vary, so this isn’t a good way to try to settle these matters in the first place. And if plausibility is all you have, maybe you should reconsider your methods or your intellectual endeavors.

At this point Nathan steps in to say:

I do not understand how this example-how some people do or would prefer to have children and others do or would not-is supposed to support anything interesting. Again, preferring chocolate to vanilla is very much unlike preferring that children not be shot over their being shot.

Note that Nathan doesn’t provide any arguments for this. He simply asserts it. Second, Nathan still hasn’t specified the respects in which they differ. How are they unlike? Third, if this is a claim about phenomenology, unlike to who? If it’s a philosophical claim about their actual nature (rather than how they seem), why should we think this is true (especially if it doesn’t seem this way to us)?

This exchange continues with input from all three:

Nathan continues to simply assert that there are differences:

Yes, you keep insisting that (although what's the difference between a preference & a "subjective" preference?). Again though, "I would prefer to have a child some day" is quite different from "I prefer children not being shot over them being shot" re. the reasons given for each.

Sure, one can give different reasons for each of these preferences. But what does this have to do with realism?

Meanwhile, Richard says:

Maybe a key thing to stress here is that there can be some cases where you can have good reasons for going either way (depending on personal factors), but that's compatible with there being other cases (like shootings) where the reasons don't depend on how you feel about it.

Sure, that’s compatible. But recall what Hannah had initially asked:

Moral realists: what evidence do you have that your intuitions are anything more objective than preferences?

It doesn’t look to me like Nathan or Richard have given a satisfactory response to this question.

The last post in the exchange as of this writing is from Nathan. It is a reference to an alleged problem with subjectivism:

A standard objection to thinking that moral judgements are "preferences" is that that entails that what often seem to be moral disagreements are not that since 'A prefers X' and 'B prefers Y' are not contradictory; you can adapt this to ethics:

This is weak tea. Nathan links to his recent article about what laypeople might mean when they say morality is subjective, which I thought was quite bad and criticized here. This is a standard objection. It’s also a terrible objection. It only works by arbitrarily restricting the notion of a “disagreement” to a dispute about the truth of moral propositions. But not all “disagreements” are disputes about what’s true. They are often disputes about what to do.

A group of friends may disagree on where to go for dinner. There need not be any stance-independent fact of the matter about where they should go, but they may nevertheless present reasons why they ought to go to one restaurant over another.

When a person sells an heirloom at the pawn shop, there is no stance-independent fact of the matter about the price of the heirloom; there are simply facts about what the buyer and seller are willing to agree upon. Those facts are fully contingent on their values and preferences. Whatever they agree to goes. And yet they can disagree on the price of the heirloom even if they agree on exactly what it is, what it’s made of, and so on. The pawn shop might even acknowledge and entirely appreciate its sentimental value, but still insist that they’re just not willing to pay more for it because they simply don’t want to.

Similar examples can be repeated indefinitely. Simply put, people often come into conflict with one another because they have conflicting preferences. Such conflicts can easily and fully account for moral disputes. The realist would have us believe the only reasonable way to make sense of two people arguing about whether abortion is right or wrong is if one thinks it is wrong independent of anyone’s stance while the other thinks it isn’t wrong independent of anyone’s stance. But what if one person really doesn’t want anyone to get an abortion, and the other really wants people to be able to get an abortion, they both realize this is the case, they both explicitly endorse subjectivism, and they both recognize that they don’t disagree about any moral facts. Would they just pack up and go home and stop fighting with one another? Of course not! The reason they’re yelling and arguing with one another is more plausibly attributed to their conflicting preferences than it is to conflicting beliefs. After all, even if one person believes abortion is stance-independently wrong, and another thinks it is stance-independently not wrong, this won’t guarantee they’ll care very much, or have any motivation at all to go argue about the matter, much less get into a fight on the street. Conversely, if someone has an intense preference that nobody ever gets an abortion, and another person has an intense preference that everyone has the right to one, this can perfectly account for why two people would yell at each other on the street.

But if we define “disagreement” as “conflicting beliefs about the truth of a proposition,” then the two people screaming at each other about abortion don’t disagree. The friends arguing about where to get dinner don’t disagree. Nobody who has conflicting preferences disagrees with each other, even if they are arguing, yelling, or presenting reasoned defenses of their respective perspectives. Diplomats negotiating peace don’t disagree even if they reject a proposal or argue about it provided they don’t dispute the facts. This restrictive notion of disagreement is ridiculous; I predict it doesn’t match ordinary usage (I’d be willing to take bets on experimental philosophy studies vindicating me on the matter). It’s just a hollow victory through stipulative gerrymandering: a common practice for analytic philosophers.

Realists make a big fuss about how realism better explains everyday moral discourse and practice. Not only does it not do so, it’s not at all clear to me that it doesn’t do a worse job of explaining how people interact with each other. At best, I think it’s a tie. Even if every single person were a committed moral antirealist, this wouldn’t entail people wouldn’t have conflicting goals or values and desire to impose those values onto one another, nor would it suggest we’d never expect to see anyone yelling or arguing or disagreeing about what to do. I elaborate on this objection to disagreement arguments in section 8.0 of this article.