How to Steelman an Opposing View

Most of us are experts at the straw man. We take the weakest, clumsiest version of an opponent's view, knock it over, and feel victorious. It is satisfying and almost completely useless. The straw man teaches you nothing because the argument you defeated was never the one anyone actually held.

The antidote is the steelman: the discipline of reconstructing the opposing position in its strongest, most charitable form before you respond to it. You build the best version of their case, sometimes a better one than they articulated themselves, and only then do you engage. Done honestly, it is the single most powerful habit a thinker can develop.

Why steelmanning works

Steelmanning is not about being nice. It is about being correct. If you can only defeat a weak version of a view, you have no idea whether the strong version is true. You may be holding your own belief for bad reasons and never find out.

• It pressure-tests your own position. The strongest counterargument is the only one worth answering. If your view survives it, you have earned your confidence.
• It earns trust. People stop defending and start listening the moment they feel genuinely understood. Nothing disarms quite like hearing your own argument stated back to you better than you made it.
• It surfaces hidden truth. Most disagreements contain a kernel the other side is right about. Steelmanning is how you find it.

How to actually do it

Steelmanning is a skill, not a slogan. Here is a practical sequence.

1. Restate before you rebut. Summarize their view until they say yes, that's it. If they correct you, you have learned something and you try again.
2. Find the strongest evidence. Ask: what would have to be true for a smart, honest person to believe this? Then assume those things are true and see where the argument leads.
3. Repair the weak links. If their reasoning has a gap, patch it on their behalf. Replace a bad example with a good one. Tighten a sloppy premise.
4. Locate the real crux. Once you have the strongest version, the disagreement usually shrinks to one or two genuine points of difference. That is where the actual conversation is.

A quick example

Suppose someone argues that social media should be heavily regulated. The straw man version is they want to ban free speech. The steelman is closer to this:

Attention is a finite resource, platforms are engineered to exploit it, and the externalities, polarization and harm to teenagers, are borne by people who never consented. We already regulate other industries with large negative externalities, so the burden is on the platforms to show why they are different.

You may still disagree. But now you are disagreeing with something worth disagreeing with, and your response will be far sharper for it.

The limits

Steelmanning has boundaries. You are not obligated to invent a brilliant case for a position that is simply bad faith or factually empty. And there is a failure mode called the weak man's steelman, where you build such a generous version that you are no longer addressing what your interlocutor believes. The goal is the strongest honest version of their view, not a different view entirely.

Like any rigorous habit, steelmanning improves with repetition against worthy opponents, which is exactly the kind of deliberate practice the Debate app is built around. Try it on your next disagreement: before you say a word in response, make their case for them. You will argue better, and occasionally you will change your own mind, which is the highest form of winning there is.