Skip to content

How opinions should work

Can an opinion be wrong? What would it mean to say "yes", or "no", for that matter? And what is an opinion, that makes it different from a fact?

I'm gonna try to answer these questions from a very opinionated point-of-view (see what I did there?) by crafting some definitions that I think are, at least, somewhat helpful to drive this discussion. But keep in mind, this is only my opinion, in a very strict sense of opinion, that I'm gonna explain to you.

To define what is an opinion, I think we have to contrast it with the complementary concept of fact. I'm gonna say there are two types of statements you can make about any subject or object: opinions and facts, and these are disjoint sets.


What is a fact?

I'm gonna define a fact as an is-statement, for example, the snow is white, or coffee is bad for health. Alternatively, I'm gonna say facts are objective statements.

Facts can be right or wrong, in a very strict sense: whether they correspond to objective reality or not. However, just trying to define what objective reality is, is tremendously complex. So I will assume that most of us share a common idea of what objective reality is, even if we don't agree that some statements are part of, or can even refer to, that reality.

More generally, facts can have a degree of accuracy, so they are neither right nor wrong, but somewhere in between. For example, when you say Earth is round, that is not exactly right, but is a far better description of Earth than, say, Earth is flat. Hence, among two contradictory facts (two facts that cannot be true at the same time), I'll generally say that the most accurate is "right", and the other, "wrong", but this is nuanced and contextual.

The reason we care about facts, is that they are useful. Facts allow us to communicate with each other about reality. Most of us, I think, we'll agree on a large common set of facts, such that something seems to be pulling you to the ground, and that you cannot breath under water.

However, many of us will not agree on some facts, even if they are a very accurate description of objective reality. This disagreement can be due to several reasons. The most obvious one, is that we have different tools to observe objective reality, such that we can be seeing different portions of the same phenomenon and thus disagreeing by a simple mismatch of data. A more complex reason is that, even when starting with the same data, we can have different reasoning processes, some better than others, and thus we can reach different conclusions. Finally, a less obvious reason, is that we don't agree on what a fact is, and one of us is taking a fact as an opinion, or an opinion as a fact.

The good thing about facts is that there is, in principle, an objective way to solve any disagreement. We just have to show each other our data and describe our thought processes. This doesn't mean we can always know if a fact is right or wrong, sometimes we won't even have the technology to be able to collect or analyze that data. But at least, we should be able to agree if a fact seems right from the collective data we have gathered, or if we cannot decide yet. This is, in part, how Science works.

With all this in mind, we can reach an alternative definition for facts. I would say:

📝 Facts are the type of statements about objective reality that can be, at least in theory, agreed upon by anyone who has access to the same set of observations and uses a principled reasoning procedure that doesn't produce contradictions.


What is an opinion?

If facts are is-statements, then I'll say opinions are ought-statements. For example, education ought to be free, or people ought to be able to say whatever they like. Alternatively, I'll say opinions are subjective statements.

Do we need opinions in the first place? Can't we just get along with facts? Well, sadly, there are fundamental questions we can only answer if we allow subjective statements. The most obvious case is possibly organizing a community. For example, if you want to build a society that allows people to live together in harmony, you'll probably want a rule like people ought to respect each other's freedom.

Now there's a funny thing that happens with statements. It's called the "Hume's guillotine", and it basically says you cannot produce an ought-statement from a chain of is-statements by pure logical reasoning. For example, you might want to say weather is cold today, and cold is bad for health, so you ought to wear a jacket. Seems reasonable, right? But the thing is, you're making an assumption, which is that I want to protect my health. And you cannot derive that as a fact from objective reality. At most, you could say you ought to protect your health, and so, you need to introduce an opinion that we need to agree on.

So, bottom line is, opinions, as I have defined them here, cannot be proven logically from facts. There is no science that can validate if an opinion is "right", starting only from pure objective truths. You have to start with an opinion somewhere, that is taken at face value and accepted as a universal truth. A subjective truth.

Keep in mind, though, that some things which may look as opinions are actually facts, and vice-versa. For example, if you say, I like coffee, that is not an opinion, that is a fact about yourself. It is a fact because we can precisely define what like means, and we can agree within reasonable accuracy, that the observations we all perceive of your coffee consumption behavior are consistent with what we refer to as liking.

And if you say, democracy is a successful sociopolitical system, that is also not an opinion, it is a fact, provided you can describe what "successful" means in this context. It can either be a true or a false fact, depending on which metrics you pick, or somewhere in between. However, whatever the case, that fact doesn't directly imply that people ought to prefer living in a democracy, by XXX razor. That is an opinion, and thus cannot be scientifically proven from objective reality.

Hence, to conclude this section, I think we can come up with an alternative definition for opinion. I would say:

📝 Opinions are the type of statements about subjective reality, for which there are equally-valid alternative statements, that cannot be differentiated by using just true facts and any principled reasoning procedure.


⚠️ Before we move on, I want to stress that the distinction between facts and opinions is not a linguistic issue.

Is not the case that something is a fact because it is expressed as an is-statement. Or vice-versa, that I can turn a fact into an opinion by changing the way it is expressed. These syntactical "rules" are just a tool for us to communicate the semantic meaning of the word "fact" and "opinion", but they are not the definition. The definitions, as hard as they are to get right, rest on the key distinction that facts are statements that can be proven right or wrong, independent of your beliefs; and opinions are exactly the opposite, statements that cannot be evaluated outside of a belief system.


Agree to disagree?

Can an opinion be wrong? We could say an opinion is "trivially wrong" when it contradicts objective reality, or simply put, if it contradicts some true fact(s). For example, if you say electrons ought to be positively charged; but I'll argue that those are, at the very least, useless opinions.

More importantly, then, can an opinion be wrong when it doesn't contradict objective reality? For example, if you say people ought to be able to decide if they want to vaccinate their children? Yes, vaccines are proven to help fight diseases, that is a fact. But we have to agree, as before, that people ought to do what's best for public health, and that, again, is an opinion (not necessarily held by a majority).

If we say useful opinions cannot be measured against objective reality, then we can only compare opinions with other opinions. Can we agree on a set of basic opinions that determine which other opinions are valid or not? I'm gonna propose a few.

I'll start by proposing that opinions ought to be only about subjective reality. This means that, in my view, you cannot have an opinion on something that can be proven scientifically. The Earth is either round or flat, and you're not entitled to believe it is any different. You can disagree with the science used to determine it. You can be a skeptic and attempt to do the experiments yourself. But you cannot believe that people are entitled to decide what charge of the electron, or what shape of planet, they want to believe in.

It is not easy, at all, to determine what can be scientifically proven, so there will be statements which, at some points in time, some people will rightfully argue there is still room for opinions. But I think that in most of the statements that matter for practical purposes, even if we don't know their truth value, we can pretty much agree if they are about objective or subjective reality.

Next, I'll propose that opinions ought to acknowledge their status. Hence, you cannot have an opinion, and then believe that opinion to be anything else than an opinion, i.e., you are not allowed to believe your opinion is a fact. But it's OK as long as you understand all you have is an opinion, and thus, it cannot be compared to a fact.

Finally, I'm gonna argue that people ought to be allowed to have opinions that do not conform to the dominant moral dogma. I'm totally OK with, and I will fight for, people's right to hold an unpopular opinion, even if I don't agree with that specific opinion.

All in all, if an opinion contradicts one of the previous three statements, I'm gonna call it a "wrong opinion". For example, from my point of view, you cannot say it's my opinion that vaccines don't work, but you can say people should be able to decide if they want to vaccinate. Maybe I won't agree with you, but I'll defend your right to have an opinion I disagree with, and I'll do my best to try and convince you otherwise.


And this is my opinion on how opinions ought to work.