A Hundred Pounds in His Pocket

About ten years ago I got into a debate in a Facebook group that has stuck with me ever since. The people in it were sharp, highly educated, and better read in philosophy than most. And a lot of them, including people on my own side of the argument, thought the thing I was saying was absurd.

What I was saying was this. Some things are logically possible, and some things are physically possible, and those are different. You can say that something is logically possible without knowing whether it's physically possible.

The subject was gods. Depending on how you define one, a god might be logically possible. That doesn't tell you whether it's physically possible. The debate went on for weeks.

The pocket

One guy, a British atheist, set out to prove I was wrong. He asked me: is it possible that he had a hundred pounds in his pocket?

I said I didn't know. I didn't know the size of his pocket.

It's logically possible that he had it. Nothing about a hundred pounds in a pocket is contradictory. But he hadn't asked about pockets in general. He'd asked about his pocket, right then. If his pocket was too small to hold that much cash, it was not physically possible for it to be in there. I didn't know, so I couldn't say it was possible.

He brought this up as ironclad proof that I was full of shit.

Possible is a claim

Saying something is physically possible is a positive claim. It means the laws of physics, as far as you know, permit it. That's a claim you have to back up like any other.

Take the god they were all so eager to call possible: all-knowing and all-powerful. Are you really going to tell me the laws of physics have nothing to say about something being all-knowing? The speed of light is a thing. As far as we know, it's a hard limit on how fast information can get anywhere. Maybe there's something we don't know about that gets around it. That maybe is what keeps it logically possible. It doesn't make it physically possible.

A lot of the atheists in that group wanted to be seen as open-minded, so they'd concede that a god is possible and then argue for why they didn't believe in one. My point was that they didn't know it was possible. Conceding it wasn't open-mindedness. It was a claim they had nothing to back up.

What else should we see?

I'm not interested in the god question here. I'm interested in how lazy we are, in general, when we talk about what's possible.

People treat belief as though they're justified in believing anything at all, as long as it's logically possible for it to be true. That's bullshit.

What people don't do, or don't do nearly enough, is ask: if that were true, what else should we see? Could it be true without this other thing also being true?

And it runs the other way. When you deny something, or deny that you could be wrong about it, you're also committed to whatever has to be true for your denial to hold. People rarely check that either. Believing something without evidence is a failure of critical thinking. So is disbelieving something when you have the evidence. They're the same failure. Critical thinking is knowing what's worth believing, and you have to check both directions.

Keep it abstract first

I'm deliberately not opening with specific examples, and here's why. The moment you bring up one, somebody who believes or disbelieves that one thing wants to fight about it, and now they're taking issue with a basic principle they'd otherwise have no problem with. Sometimes I'm happy to have that argument. Sometimes I'd rather get the principle across.

There's a better reason too. If you can only state your principle inside one specific belief, do you actually understand it? Do you know what it entails? Are you even following it, or did you invent something for that one case so you could sound smart?

Examples come after. You state the principle, you work it out, and then you test it: you make a prediction and see whether it holds. Pick cases that are hard. The moon landing happened and vaccines don't cause autism, but those are easy, trivial applications. School doesn't even touch those; in school we take on weaker ones still and pretend we've done critical thinking.

You're not doing critical thinking until you're challenging a belief you want to believe, or looking for reasons a belief you want to reject might be true. If it never makes you a little uncomfortable, you're not doing it.

On myself

So here's one that makes me uncomfortable.

A while back I wrote that AI isn't going to take your job. A lot of what I said, I still absolutely stand by. But not long ago I was thinking about employees I've had: the good ones, the ones who do just enough, the bad ones, and how much managing each of them took. And I realized that AI, as far short as it falls in a lot of ways right now, is still better than a huge number of bad employees. A great employee using AI is going to knock it out of the park, and I'd happily pay for their AI. The bad ones? I would actually replace them with AI.

That forces me to confront my own claim. So I look at what else we should see. If AI were already replacing jobs at scale, employment should be falling. Instead it's near record levels, while America's war with Iran, the war in Ukraine and China's threats against Taiwan, the Philippines and Japan should all be making employers hold off on hiring. Something seems to be propping the job market up, and the most logical candidate is AI. At the big-picture level, the job apocalypse doesn't seem to be happening.

That's the evidence about today. About the future, I don't know. I keep coming back to all those employees who do just enough. They're replaceable. They're not putting in the effort.

Ask what else would have to be true, and aim the question at yourself.