Kyeli and I were talking about philosophy and the underlying nature of the universe the other day, and she told me her beautiful beliefs about what she thinks is the underlying nature of the universe. I said, “That’s really beautiful, it’s a shame that the fact that it would make us happy to believe it makes it a priori less likely to be true.” “What?!” she replied.

“People are more likely to believe something that makes them happy. So if someone has a belief about the nature of the universe that makes them unhappy, then a likely explanation is that it’s true. But it someone has a belief about the nature of the universe that makes them happy, then there are two likely explanations: one, that it’s true, and two, that they just made it up to make themself feel happy, safe and secure.

“Imagine an experiment, where each person goes into a room to listen to a radio signal, and they’re supposed to write down the instructions given on the radio and then hand them to the experimenter to carry out. The radio voice is going to tell you how much you will be paid for your participation in this experiment, between $1 and $100. So first you do a control experiment just to figure out how many people lie, and then you do the real experiment, where half of the radio signals are unintelligibly garbled. The subjects are told that if they can’t hear the signal clearly, to just make something up, it’s fine. So I bet what we would find is that people who heard the garbled signals would write down higher numbers than people who heard the clear signals. Therefore the conditional probability of ClearSignal given LowNumber is relatively high, whereas the conditional probability of ClearSignal given HighNumber is relatively low. It’s more likely that the HighNumber was due to GarbledSignal. This is a metaphor for people trying to understand the true nature of the universe. Most of us probably get garbled signals, some of us may get clear signals, but all things being equal, a high number (a belief that makes us happy) is less likely to be true (clear signal) than a low number.

“Here’s another analogy. You have two big bags full of stones. Call them the Bag Of Truth and the Bag Of Bullshit. The Bag of Bullshit contains almost all white stones, and the Bag Of Truth, nobody knows what colour stones it contains. So if someone comes out with a white stone in their hand, chances are they got it from the Bag Of Bullshit. Sure, they may have gotten it from the Bag Of Truth, but it’s just as likely, maybe even more likely, that they got it from the Bag Of Bullshit. Whereas if someone comes out with a black stone in their hand, it’s much more likely that they got it from the Bag Of Truth, because there aren’t many non-white stones in the Bag Of Bullshit. So the white stones are beliefs that make you happy, and the bags, well the bags are pretty obvious.”

“So what you’re saying is that the universe probably sucks, because if you believe anything good, it’s probably bullshit!”

“Not probably. It’s just a priori less likely to be true. It’s Ockham’s Razor. If there are two possible explanations for a bunch of people having a belief, and one of them requires a bunch of things you consider to be improbable and the other one is simply ‘they made it up so they could feel happy, safe, and secure’, then the simplest explanation is probably best. Now of course ‘simplest’ is a matter of opinion, for instance the ancient Greeks would have probably said ‘The Gods made everything’ was the simplest explanation. So any of these beliefs has to compete with ‘they made it up to be happy’, which is a pretty darn simple explanation.”

We went round and round like this for two days, me talking about science and conditional probability, and Kyeli wondering at how much it must suck to live in a world where nothing beautiful could possibly be true. (“Not possibly, just a priori less likely.”)

But suddenly this morning when I woke up, I realized that Kyeli was right and that I was wrong. The problem with my argument is similar to the problem with Pascal’s Wager — it’s making bogus quantifications of the unknown. Who knows what the underlying nature of the universe is? It could be such that it causes us to believe things that are true, even if we just make them up! It could be such that true beliefs about such things tend to make us happy! It could be something else entirely that we can’t really comprehend, and if so, it makes absolutely no sense to impose any sort of probability distribution on it.

I feel like a weight has been lifted from me. I’ve been unconsciously feeling bogged down by this belief, which has subtly influenced me toward believing that the true nature of the universe probably sucks. But now, I know that I do not know, and I am free.