A mortal sin of argumentation

Waterfall Man: “You think AI might have some benefits? So you’re saying it literally doesn’t matter if the government uses it to kill people?”

A few years ago, I had a big argument with one of my relatives. She had suggested that cultural differences between social groups might partially explain other differences between them — for instance, mental health outcomes and wealth. This is obviously true. Why are Jews so disproportionately successful compared to other religious groups? Presumably, something useful trickled down from five thousand years of being the “people of the book.”

Unpublishable Papers is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

If she had used Jewish success as an example, we probably wouldn’t have had much to argue about. But instead, she used crime rates among African-Americans. There is no reason to reject this argument without seeing where it goes: if we can attribute at least some portion of a social group’s success to cultural differences, then it is not prima facie wrong to attribute at least some portion of their problems to those same differences.

Alas, I was an English major. So I committed a mortal sin of argumentation — I allowed myself to follow a chain of preexisting assumptions about what her point implied. She said this about crime rates and cultural differences —— which means that she thinks structural barriers don’t matter —— which means that she doesn’t believe in systemic oppression —— which means that she thinks the plight of African-American people is entirely their own fault and that we owe them no care or support.

If I had just asked her to elaborate, instead of following these associations toward an inaccurate interpretation of her argument, then the next three hours of useless yelling probably could have been avoided.

I call this cascading: I’ll define it as “the process of converting someone’s defensible point into an indefensible point based on preexisting assumptions about their views.”1 I want to get better at avoiding it, and I want other people to get better at avoiding it too.

Unfortunately, it is difficult to avoid: cognition loves a cascade, because cascading is a very efficient way to extract useful models of the world. Look at the pictures below. You see faces. Why? There aren’t any faces here. But that doesn’t matter. Your brain sees shapes arranged in the way that facial features tend to be arranged, so it cascades toward the conclusion that this array of pickles, cones, purses, and kayaks all possess faces.

For those curious, this phenomenon is called pareidolia.

For vision, cascading is just fine — you can’t rid yourself of the face illusion, but you’re smart enough to know that you shouldn’t feel bad about eating the pickles just because they’re apparently smiling.

For argumentation, however, it is a mortal sin. In my experience — both as a sinner and a victim of the sin — rhetorical cascading tends to shut down conversations before they can start. It limits our ability to learn from each other, maintain mutual respect, and arrive at more objective views of ourselves and the world around us.

So how do you stop yourself from cascading during debates? The crucial step, I think, is to notice what it feels like when you’re doing it.2 Here is a non-exhaustive diagnostic checklist — any one of the symptoms should signal that you’re probably about to cascade.

1. You’re preparing to reply before the other person has finished their current point.
2. You’re primarily thinking of rebuttals rather than questions.
3. Perhaps it is, in fact, time for a rebuttal — but you find yourself thinking mainly of buzzwords rather than arguments.
4. You do not feel like you could repeat back what the other person just said in a way that they would agree with.

If you’re in a debate and you’re displaying some or all of these symptoms, then you’re almost certainly about to ruin the conversation by jumping down a rhetorical waterfall.

Admittedly, I was mostly motivated to write about this by a few recent conversations where I’ve felt like I was the target of a cascade. It is disturbing to watch someone take a point you’ve thought carefully about and twist it into a completely different (and patently nonsensical) cliche. I could see the river rushing over the cliff, and there was nothing I could do to stop it.

POV: You’re about to be called a fascist.

Or perhaps there was! Cascades beget cascades — I suspect that in these conversations, I allowed the waterfalls of my interlocutors to shatter my own poorly constructed dam. Deciding not to initiate cascades is a virtue; getting smacked in the face by one and still managing to hold your ground is next to godliness.

So I urge you to be better than me. Yes, some people believe awful and ridiculous things. But before you jump to that conclusion in the course of a conversation, it’s best to hold back the flood and ask a few questions first.

Subscribe to Unpublishable Papers? Oh so you’re saying this is literally the only good Substack and all the other ones should get banned from the Internet?

1

I understand that this is very similar to “straw-manning” — I think this verb is useful, though, because it better captures the sensation of what it feels like to find yourself magnetized toward a straw man argument.

2

Depending on your worldview, you might find it useful to reflect upon the chain of associations that was triggered by the phrase “crime rates among African-Americans.”