On Fact and Opinion

Against an all-too-common distinction

Everyone is taught the distinction between fact claims and opinion claims in grade school. The ability to sort fact from opinion is widely considered to be a skill that every educated adult should have, and a crucial component of media literacy. This distinction between fact and opinion is essentially a philosophical one; categorizing different sorts of claims and coming up with theories of the natures of facts and opinions is precisely what philosophers like me spend our days doing. So you may be surprised to learn that every single professional philosopher that I know – and I know a lot – rejects the idea that there is a distinction between claims of fact and claims of opinion. It’s a dogma that can’t withstand serious scrutiny. I know, I know!¹ But hear me out.

Let’s begin with what the distinction is supposed to consist in. Here’s a representative account: “A fact is a statement that can be verified. It can be proven to be true or false through objective evidence. An opinion is a statement that expresses a feeling, an attitude, a value judgment, or a belief. It is a statement that is neither true nor false. Or it may feel true for some, but false for others.” To see the problem, begin with how an opinion is described: opinions express beliefs. Yet many of our beliefs concern matters of fact! Consider the claim “Paris is the capital of France.” That is a claim that pretty much anyone would categorize as a fact claim. Yet anyone who asserts it would be expressing one of their beliefs. (You do believe that Paris is the capital of France, don’t you?)

The deep problem has to do with what philosophers call propositions. Propositions play many different roles in philosophy. They are the meanings of (declarative) sentences. They represent the world as being a certain way; as such, they are “the bearers of truth and falsity,” capable of being true and false. And they are “the objects of the attitudes.” When philosophers say this, they are referring to the fact that we have a lot of different attitudes that get paired with ‘that’-clauses in English. We can believe that something is the case, hope that it is the case, fear that it is the case, know that it is the case, and so on. The thing that is expressed by the ‘that’-clause, the thing that we believe/fear/hope/know/desire/regret/doubt (etc, etc) is the case, is a proposition.

Fact claims are propositions. They represent the world as being a certain way, and are therefore capable of being true or false. But opinion claims are also propositions! If you believe that something is the case – if you are of the opinion that it is the case – then the thing that is your opinion is a proposition. Propositions are the objects of the attitudes like belief and also the bearers of truth and falsity. So “fact claims” and “opinion claims” are actually the same thing. They’re both propositions.

If you need more convincing, pick a claim at random and ask: “Does this claim represent the world as being a certain way? Is it capable of being true or false?” If the answer is yes, it would seem to be a factual claim. Yet it would also be an opinion: it’s the sort of thing that you could believe, and so someone could express their belief by stating it. If the answer is no, then it wouldn’t be a factual claim. Yet by the same token it could not express an opinion! If a claim doesn’t represent the world as being any particular way, then it doesn’t make sense to believe it. “Gronkniks are plinkly” is a nonsense claim that doesn’t represent the world as being any particular way, so it would be decidedly odd for anyone to claim to believe that gronkniks are plinkly. What are you believing anyway?

For all this, most people think they can sort sentences into fact claims and opinion claims, and they’re generally consistent in how they do so. So people are latching onto something when they draw the distinction. But what?

Primarily, two things. First, people call claims that are unsupported by evidence “opinion claims.” This is a holdover from the verificationist idea that claims that can’t be supported by evidence are meaningless. But it’s strange to say that claims which can’t be supported by empirical evidence can’t be factual claims. Consider the claim that God exists. That is clearly making a factual claim about the world: If God exists, His existence is a fact. If God doesn’t exist, that’s also a fact. “God exists” is either true or false. Yet that can’t claim can’t be supported by evidence.² So “claims that are supported by evidence” does a bad job of picking out what we mean when we talk about factual claims. Another problem with this definition is that opinion claims are then claims that are unsupported by evidence. To say that you have an opinion is thus to say that you believe something is true even though you have no reason to think that it’s true. If this is what an opinion claim is, then it’s always irrational to hold an opinion!

A second class of claims that are frequently regarded as “opinion claims” are moral or, more broadly, evaluative claims. To call something good or bad, right or wrong, better or worse, is supposed to be merely a matter of opinion. This is often what differentiates an “opinion” article in a newspaper from one that is “straight news.” Opinion writers freely make evaluative claims and recommend different courses of action, while straight news stories avoid doing this. And there is a real difference between claims that evaluate and those that merely describe. Being able to distinguish these two kinds of claims is useful.

But there’s still good reason to hesitate to apply the “opinion” label to all evaluative claims. Is it really the case that there is no fact of the matter about morality? Does the claim that genocide, rape, and torture are all morally wrong not state a fact? Most philosophers would agree that it’s an objective moral fact that the Holocaust was a grave moral crime.

Now there are philosophical views according to which evaluative claims wouldn’t count as knowable matters of fact. According to a view known as non-cognitivism, moral claims don’t express propositions. They don’t represent the world as being any particular way; rather, they are expressions of emotion or something like that. According to non-cognitivists, “The Holocaust was evil” means something like “Boo to the Holocaust!”³ This used to be a very popular view, but the view is much less popular now (although it still has adherents). And moral skeptics hold that, while moral claims do express propositions, it’s impossible to know whether those propositions are true. For a moral skeptic, “The Holocaust was evil” does describe the Holocaust as being evil; but who knows whether it actually was evil? The universe is a great mystery!

None of this is to say that you shouldn’t be a non-cognitivist or a moral skeptic. Those are respectable philosophical positions (although their adherents are in the minority among philosophers). The point is just that to call an evaluative claim just an “opinion” is to saddle yourself with some highly controversial philosophical claims about the nature of morality or normativity more broadly.⁴ These are hardly common sense claims that should be obvious to any grade schooler.

This all matters because the fact/opinion distinction can be used as an excuse for sloppy thinking. Any genuine opinion represents the world as being a certain way. If you’re going to think of the world as being a certain way, you have a responsibility to form that belief carefully, on the basis of evidence. But opinions aren’t the sorts of things that need to be supported by evidence (according to the common dogma), so calling something “my opinion” is a way of avoiding responsibility for your beliefs. In a recent interview with Megyn Kelly, when Kelly asked Trump about the classified documents that he took to Mar-a-Lago, Trump replied, “I’m allowed to take these documents, classified or not classified. And frankly, when I have them, they become unclassified. People think you have to go through a ritual. You don’t — at least in my opinion, you don’t.” That might well have been Trump’s opinion; but if so, his opinion was wrong.

It’s not just Trump, of course. “That’s just my opinion” is deployed all the time as a shield from criticism. But it should be treated as an invitation to criticism. All claims of opinion are claims of fact. The objects of the attitudes are the bearers of truth and falsity. If you have an opinion, we can ask whether that opinion is true, and if it’s supported by evidence. And if it’s not, then it shouldn’t be your opinion anymore.

1. People actually get mad at me sometimes when I try to tell them the distinction between fact and opinion is bogus.

2. That’s controversial, of course. But figuring out whether or not the claim that God exists can be supported by evidence will involve a complex tangle of questions about what evidence is and what it takes for a claim to be supported by evidence. The difficulty of those questions constitutes another huge problem for verificationism and the fact/opinion distinction that is based on it. But these issues are too thorny to get into here.

3. Contemporary versions of non-cognitivism are substantially more sophisticated than this, but this is the basic idea that the more sophisticated versions developed from.

4. If you’re interested in this stuff, though, I just wrote a book about this subject! You can find it here.