Monday, November 12, 2018

What does Believe Mean?

"Do you believe in God?"
"That depends on what do you mean by believe."
Since he has been talking about the Old Testamentreligion, and God, Jordan Peterson gets asked this question a lot. He seemingly dodges the question by asking what they mean by believe and what they mean by God. I get why this is an unsatisfying answer, but lets take it seriously for a second.

What do we ever mean by believe?

The Petersonian answer is, what you act like is true. If I act like my car works, then I believe my car works. If I act like my coffee is hot, then I believe my coffee is hot. If I act like God exists, then I believe God exists.

Okay, now let's throw some harder hypotheticals at this definition:

1) Jordan Peterson's definition does not capture high severity low probability scenarios.

I act like I'm going to get in a car accident when I put seat belts on my children, does that mean I believe I'm going to get in a car accident? I think not.

If the stakes are high enough, we will act as if something is true even when we don't believe it. It's unlikely that global warming is an existential threat, but a 1% chance may be high enough to act like it is. I install a smoke alarm even though I don't believe I'll need it. Batman's logic is questionable, but it seems wrong to say he believes Superman will wipe out the entire human race.

“He has the power to wipe out the entire human race, and if we believe there’s even a one percent chance that he is our enemy, we have to take it as an absolute certainty”
2) Jordan Peterson's definition does capture acts in indifference

There are two doors, each one of them has a 50% chance of having a pot of gold behind it. So you pick a door at random. You act like that door has the pot of gold behind it. But if I asked you whether you believe the pot of gold was behind the door I think you'd say, "I don't believe it because I have no idea."

It doesn't have to be 50%. This problem exists in any situation where your confidence is more or less a coin flip. Belief correlates with only high confidence scenarios, whereas actions may only need the probability to be on your side (or much less than that if the stakes are high enough).

3) Jordan Peterson's definition captures ideas we have absolutely no awareness of.

Evolutionary psychologists have long pointed out that men are attracted to women with high hip-waste ratios because it signals fertility. Most people don't know that, but they act in a way that that is very well consistent with that idea. If you pay attention to psychology or especially Robin Hanson on Signalling, you find out that we don't know why we do a lot of the things we do. As Jordan Peterson himself puts it,
We act out things we don’t understand, all the time. If that wasn’t the case, we wouldn’t need psychology, or sociology, or anthropology, or any of that, because we’d be completely transparent to ourselves, and we’re clearly not.
I'm not sure how Peterson puts these pieces together. To me, this is not the way people use the word belief. Like, ever. Belief is a cerebral act, or else people don't use that word.

When we use belief to describe background motivations for behaviors, people end up with a lot of strange beliefs.

"You married her because you believed she could spread your genes to the next generation"
"You voted for universal health care because you believe it will make you look generous"
"You fight against climate change is real because you believe it will signal political loyalty."
"You laughed at that joke because you believe it will improve your social bonds."

If you think, "those motives didn't even occur to me!", remember that according to Peterson it doesn't matter. It isn't relevant if you're aware of your motivations, if you act like they're true then you believe them. If we define it this way, we have ground to accuse people of believing lots of things that never occurred to them. This seems like fodder for unjustified moral berating.

Beliefs must be conscious, or at least resides in latent subconsciousness where it can be pulled up with some introspection. Beliefs sleep and can be woken up, but they're never ideas you're completely ignorant of. We act not only according to ideas we've never thought of, but according to ideas we could never think of. Those kinds of ideas don't seem like beliefs at all.

Nobody believed in evolution before Darwin thought of it, even if they acted like it was true.

Those are some of my objections to Jordan Peterson's definition. I have some others. Like I'm not sure how well hypocrisy is maintained as a meaningful concept if beliefs and actions are always consistent.

I'm also not sure how well Peterson is stuck on his own definition. He always says things like, "one way you could define believe is..." or "some people define believe as..." or whatever. He never states that it's his definition, or that it's a good one for how other people use the word.

So what's a better definition of believe?



I think belief is a commitment to the truth of a proposition. It's an epistemic tie between the self and the conclusion. It hurts when your beliefs are wrong. Belief is a conviction, which is more than mere confidence. A belief is to be done thinking because you're already sold.

Under this definition beliefs are necessarily stubborn and good bayesian never believe anything. Even if you have 99% confidence of something you don't quite believe it because 1/100 beliefs like that are wrong. That pretentious intellectual that tells you that they love when their beliefs are wrong because it means they learned something, yeah total nonsense.

If we can just cope with the negative nature of beliefs, which I know will put anyone off to this definition immediately, I think the definition holds.

I notice people don't go around voluntarily stating that they believe the sky is blue. But they will state they believe in Jesus or the Lakers or global warming. If you prompt them; however, they will state they believe the sky is blue. "Of course I believe the sky is blue". I think that statement forms the belief.

Think about it. If you hadn't asked me if the sky was blue and the sky turned out to be red, I wouldn't feel weird about it at all, even though I was wrong. But if you made me say it, now I'm wrong in a different way. I'm wrong in a way where I feel embarrassed, like I have excuse myself because nobody could have predicted the sky would turn out to be red.

Uttering the words, "I believe" turns boring old wrongness into an identity issue. That doesn't mean you can't have quiet beliefs, but stating a proposition alongside "I believe" is one way of committing yourself and forming the belief itself.

I think this explains why people can believe in concepts without any specific propositions attached to them. When you believe in God or climate change it means you believe they exist. Okay, so what does it mean to believe in Jesus, Israel, social justice, free speech, the death penalty, or yourself? Of course these things are real, so what specific proposition are you saying is true about them?

None. When you believe in these things it means you've committed yourself to their side. It means you're allies with them. Belief is about you and where you stand in an invisible war, and sure sometimes that war is over the truth of a proposition, but sometimes it's merely about a lone concept or person.

I think this explains why beliefs are so often submerged in moral waters. If you believe life begins at conception, to pro-choicers you're more than just mistaken. You are on the wrong moral side of the culture war.

I also think this explains why the language of beliefs surrounds tribal issues like politics or religion. Not 100%, but you are far more likely to "believe" in a political statement than that WalMart closes at 10pm.



Do I believe in this definition? Let's just say I'm not committed to it.