People liked last week's video essay, and I had fun doing it, so I decided to do a few more. I'll record a short video, the AI will make a transcript, I'll use AI to help me edit that into essay form, and then I'll post both. I'll run the experiment for the next few weeks and see how it goes.

For this first official video essay, I want to talk about the idea of good faith in arguments.

Traditionally, good faith is defined in opposition to bad faith—where good faith means you're not engaged in deceptive argumentative behavior. Bad faith behaviors are generally those that are deceptive in some way: directly lying about facts you know to be false, intentionally using faulty logic and talking over your interlocutor, or any rhetorical tricks that avoid arguing straightforwardly. Good faith, then, means you're arguing in a straightforward way without intending to deceive.

As a definition of good faith, this is fine. Certainly, no one wants to argue with someone who's arguing in bad faith. It makes sense in many contexts, especially debates. If you're arguing in a courtroom or political debate, it's hard for the audience to figure out what's going on if the debaters or lawyers are lying. So it's important that debaters and lawyers argue in good faith.

That said, I think this definition doesn't go far enough. In a debate, it's fine because the audience decides what's true. Similarly, in a courtroom, lawyers don't decide what's true—the judge or jury does. There's an audience making the judgment, and the arguers' job is to present the maximally strong version of their position and let someone else sort it out.

This works in some contexts, especially legal ones, though it'd be nice if we could develop other ways of discovering legal truth. The adversarial system has evolved features that make it, I'd posit, our best option for finding truth in a courtroom—not the best possible system, just the best we've found so far.

However, this idea of good faith doesn't go far enough for the types of disagreements I often have, which thankfully aren't in courtrooms or on debate stages, but in conversations with friends or people online where we're disagreeing about whether a theory is true or about some interpretation of facts.

In those conversations, what I want isn't for us to debate while others draw their own conclusions (though I hope folks reading do that). I want myself and the folks I'm disagreeing with to come to some better understanding of the truth than what we had going in.

Yes, it's always possible that one of us is just right and the other wrong. But more commonly, we're all in some state of confusion about what the truth really is, and together we're trying to become less wrong, to have a more correct sense of how the world works.

To do that, it's valuable for people to be curious, to try understanding one another's positions, and importantly, for all sides to update their positions in light of evidence and arguments presented by others.

This is where the traditional idea of good faith falls flat. If you enter such a discussion with merely good faith debate norms, the experience becomes like conversing with a brick wall. With debate norms, you dig into your position, argue for it maximally, and give the least charity possible to everyone else. But this feels like bad faith in the discussions I'm trying to have because there's no curiosity, no attempt to update—just an attempt to push a position and let others decide if it's true.

What's better is if we can engage with one another to figure out together what's true. This means trying to make sense of the other person's position, offering them some charity—not maximal charity, but some amount given that conversations are hard. Sometimes it's difficult to explain what we mean, or we think we've explained clearly, but the other person understands our words differently than expected.

Beyond charity, we need real willingness from all parties to update their positions in light of what's argued. Otherwise we're not seeking truth together—it's more like one person's seeking truth while the other's just holding their position and refusing to budge.

Of course, there are degenerate cases. I don't mean that if someone believes the moon's made of cheese while the other knows we've been there and it's clearly not, there's any real update to have about cheese-moon probability. That's not what I'm talking about.

But if you're debating with a Flat Earther or someone who believes something you're sure is false, there's still the question of why they believe it. What caused them to have that belief? What's the crux of your disagreement? What do they believe that you don't, where if one or both of you changed your minds, the downstream beliefs would also update?

That's the ideal: in conversations, it'd be great if people could find agreement, everyone updates, and we all have a better understanding of the truth. But that won't always happen. Sometimes the gap's too wide to bridge in a single conversation. When that's the case, it's important to at least try getting down to the crux of the disagreement and understand what belief, if changed, would change your minds.

So that's my idea of what good faith really means. I don't think it's radical, but it's stronger than the traditional notion. My experience is that these norms—where you have curiosity, you're trying to update, and you're offering charity—lead to better conversations for finding truth.