Peter Singer tweeted something the other day about an “interesting” NYT article about moral relativism in schools:

Since Singer is one of the most famous contemporary ethicists, and I have a professional interest in both philosophy and education, I went and had a look. And the article is indeed instructive—in that it gives won­derfully clear examples of terribly muddled thinking, i.e. bad philosophy.

Assuming that you have read the article (which would be rather helpful, to be honest), let’s get right into it. The author, philosophy professor Justin P. McBrayer, starts us off with a rhetorical question that is designed to activate the outrage module in our brains:

What would you say if you found out that our public schools were teaching children that it is not true that it’s wrong to kill people for fun or cheat on tests?

Well, that’s a misrepresentation of what McBrayer later shows schools are actually doing: they don’t say, ‘It’s not wrong to kill people for fun’, but ‘The category right/wrong doesn’t apply to moral questions’. One should think a philosopher would be able to see the difference and at least be fair to what others are saying, how­ever disagreeable their statements might be.

Now, the point is that McBrayer says here that the problem—i.e. people not agreeing with the proposition that ethical statements can be assigned truth values—stems from relativism. And he sees this relativism in a definition used by his son’s school:

Fact: Something that is true about a subject and can be tested or proven. Opinion: What someone thinks, feels, or believes.

So what’s wrong with this? According to McBrayer, this distinction “undermines the view that there are objective moral facts”. How so?

First, the definition of a fact waffles between truth and proof — two obviously different features. Things can be true even if no one can prove them. For example, it could be true that there is life elsewhere in the universe even though no one can prove it.

A couple of things need to be said here. Firstly, the definition doesn’t “waffle between truth and proof”, if you interpret it to mean that it is only possible to categorise a proposition as true if it is open to proof. That bit is not what is problematic. What is problematic, though, is that, in addition, they make the same mistake as McBrayer and say that being a fact is identical with being true. How is that a mistake? Let me explain.

McBrayer’s position on ‘truth’ is that it’s absolute: If there is life elsewhere in the universe, then that is true—whether anybody knows it to be true or not. Now, when he talks about “life elsewhere in the universe”, that is a feature of the world. That, however, is something that is usually called a “fact”. “Truth”, on the other hand, is usually seen as a property of propositions, i.e. statements about certain things. And that makes a lot of sense: facts either are facts or they’re not; calling them “true” doesn’t really add anything. To give an example: The statement, “There is life elsewhere in the universe”, is true if and only if it describes an actual fact. But facts on the one hand and statements about facts on the other are different things and should be clearly differentiated.

And here’s why. If you try to use an absolute definition of ‘truth’, you run into a somewhat embarrassing problem: Since every observation, argument, idea, etc. is fallible, there can be no guarantee that what we discern to be a fact is really a fact. And since the universe is practically (if not in fact actually) infinite, there is no way that any statement can be true to (i.e., correspond with) all the facts—and even if it could, we could never know that for certain. The inevitable conclusion is: once you insist on using an absolute definition of ‘truth’, you cannot say of any statement that is true. You have, in effect, defined the concept so as to make it unusable.

This leads us to the actual problem of the matter—one that is actually already visible in the school’s distinction between fact and opinion McBrayer took exception to. What that school wanted to say was that some things are purely subjective, while others are not. Basically, the distinction is this: Some statements can be critically, inter-subjectively discussed, for example those about facts. Other statements seem to be of a substantively different variety. Should I eat chocolate ice-cream or vanilla? That is a question of purely personal preference (mere opinion).

The relevant distinction then becomes not that between ‘relative’ and ‘absolute’, but that between ‘subjective’ and ‘objective’. What is important to see is that the two distinctions are not identical: truth can be viewed as absolute and subjective (a mere belief that happens to correspond with a fact), but, more importantly, it can be viewed as relative and objective. And since the truth we can reasonably want is not an absolute one, let’s examine why objective truth has all the properties we can reasonably expect it to have and why it doesn’t matter that it is a relative truth.

It is fairly obvious that a kind of truth that is relative to something subjective, i.e. a belief or any other kind of state of mind, is the kind of thing we fear when we (pejoratively) talk about “relativism”. But it’s not its relativism we fear, it is its subjectivism: the idea that truth should be something that by definition only the person whose mind it forms in can judge. Why shouldn’t we use such a definition? Because it would mean that we could no longer talk to others about this “truth”, except to relate the fact that we believe it—because then saying “X is true” would mean nothing more than “I believe X”. But clearly, when we talk about something being true, we want to say something more than just “I believe it”. That is the sense which McBrayer invokes when he describes what he fears some people are espousing: “In that case, E=MC² is a fact for a physicist but not for me.”

Here’s another point he mistakenly attributes to relativism, when it is in fact about subjectivism:

If we’ve taught our students for 12 years that there is no fact of the matter as to whether cheating is wrong, we can’t very well blame them for doing so later on.

That conclusion in no way, shape, or form follows from the premises. There is also no fact of the matter whether driving on the left or on the right is correct; the only thing that matters is that there be a consensus on the question. And it might be just the same with moral questions, where one might possibly argue, “Society would be impossible if we didn’t agree on certain rules (however arbitrary they are)”. That would be equivalent to repudiating subjectivism: “Ethics can’t work if the validity of ethical truths by definition extends no further than the edges of each of our individual (subjective) minds.” And again, the opposite of subjectivism is not absolutism but objectivism.

The only thing we need to do in order to be able to make objective truth (and objective ethics) possible is to refer to a frame (which in no way has to be accepted dogmatically but may remain criticisable) within which we can talk about logical relations. And it is crucial to notice that this is a step we have to take in all fields, even in logic and mathematics: without first accepting the principle of non-contradiction, no logic (as we commonly understand it) is possible; and without accepting certain mathematical axioms and definitions, no proofs are to be had. And we accept e.g. the axioms and definitions of mathematics not because they are inevitable (or in any other sense absolute) but because they are conducive to working towards a (mostly unstated) goal: the growth of knowledge.

By the way: Did you catch McBrayer’s actual argument for why he thinks there are “moral facts”? No? Well, maybe that’s because he doesn’t give one—he just dogmatically assumes the truth of his statement that moral facts (absolutely) exist. That is like saying that the sum of the internal angles of a triangle just (absolutely) is 180 degrees—which presupposes Euclidian space and thus, in fact, is relative. But it is still objective: once you imagine a triangle in Euclidian space, that object which you just imagined into being has autonomous properties that are independent of any one person’s subjective prejudices. And those properties, furthermore, can be inter-subjectively examined (i.e. tested).

The necessary premises—whether Euclidian space, certain definitions of ‘infinity’, or a goal for a certain intellectual activity—are never inevitable. No one can force me to accept that logic has to be non-contradictory; nor that science must make progress or produce knowledge; nor that maths has to be useful as a language to describe the real world. The only thing that is inevitable is that if I don’t accept those premises, then I play a different game, and it would be disingenuous to pretend otherwise. Nobody can force me to refrain from calling a game in which players use their hands and aluminium bats to play a ball “football”; but if my stated goal was to be understood, I would be objectively in the wrong.

The same goes for ethics. While nobody can force me to accept that the goal of health is well-being, there still is a commonly accepted usage. “Eating iron filings is not healthy” is in no sense absolutely true; but it is objectively true if your stated goal is the common notion of ‘health’. And this notion doesn’t have to be dogmatically accepted, it can be argued and critically examined. Similarly, the statement “Throwing acid in the face of a girl because she wants to go to school is wrong” can properly be called objectively true—we just have to be able to both spell out the underlying premises and critically discuss them. That is all that is required for objective truth—whether in empirical science, maths, philosophy, or ethics. ‘Absolute truth’, however, is a notion we have no need for and should discard.

Peter Monnerjahn is a writer and teacher based in Berlin. He is currently completing a PhD dissertation in philosophy on Popper’s ‘Open Society’ and the enlightenment.