Press T for contents

Part III: The Art of Argument

15

STEELMANNING

Strengthening your opponent's case

17 min read

Art Deco illustration for Chapter 15

What if winning an argument meant first helping your opponent make theirs? It sounds backward—maybe even self-defeating. Why would you strengthen the position you're trying to defeat? But this counterintuitive move is one of the most powerful techniques available to anyone who cares about truth more than triumph.

We've all seen the opposite approach. In Chapter 13, we examined the straw man fallacy: misrepresenting an opponent's position to make it easier to knock down. Strawmanning is tempting because it works—at least in the short term. Distort what someone said, attack the distortion, declare victory. Audiences who aren't paying close attention might not notice. But anyone who does notice will lose respect for you. And more importantly, you haven't actually engaged with the real argument. You've shadow-boxed with a phantom.

Steelmanning is the opposite. Instead of weakening your opponent's argument, you strengthen it. You articulate their position better than they did. You fill in gaps they left, remove ambiguities they created, supply evidence they forgot. You build the strongest possible version of the case against your own position—and then you try to defeat that.

Steelman STEEL-man

The strongest possible version of an argument, constructed in good faith. To steelman an opponent's position means to articulate it more clearly, charitably, and persuasively than they have—then address that improved version rather than a weaker original. The term arose in contrast to "strawman."

WHY STRENGTHEN WHAT YOU'RE TRYING TO DEFEAT?

Three reasons. First, if you can defeat the strongest version of an argument, you've really proven something. Beating a weak version proves nothing except that your opponent argued poorly. Beating a strong version—the best case that could be made for the other side—proves the position itself has problems. The victory is real.

John Stuart Mill understood this profoundly. In On Liberty, he argued that we don't really understand our own positions unless we can articulate the case against them:

Second, steelmanning builds credibility. When you demonstrate that you understand your opponent's position—really understand it, well enough to state it better than they did—you earn respect. Your audience sees that you're engaging seriously, not playing tricks. Your opponent sees that you're treating them fairly. This is ethos in action, as we explored in Chapter 5: demonstrating intellectual virtue by how you argue, not just what you argue.

Third, steelmanning protects you from self-deception. It's easy to believe you've refuted a position when you've only refuted a caricature. The exercise of building the strongest opposing case forces you to confront your blind spots. Sometimes—often—you'll discover that the opposing view is stronger than you thought. That's uncomfortable, but it's also valuable. Better to discover the weakness in your position now than to have it exposed later in front of an audience.

THE PRINCIPLE OF CHARITY

Steelmanning operationalizes what philosophers call the "principle of charity": the practice of interpreting others' statements in the most reasonable way possible. When someone says something ambiguous, assume they meant the more sensible interpretation. When their argument has a gap, consider what they might have intended to fill it. Give them the benefit of the doubt.

Principle of Charity PRIN-sih-pul of CHAR-ih-tee

An interpretive guideline: when engaging with an argument, assume the most reasonable interpretation of what the speaker meant. Fill in unstated premises generously, resolve ambiguities in the speaker's favor, and address the strongest version of their position. This builds trust and ensures genuine engagement.

This isn't naive. The principle of charity doesn't require you to accept bad arguments or ignore genuine flaws. It requires you to distinguish between stated arguments (which might be poorly expressed) and intended arguments (which might be more reasonable than the words suggest). Attack the intended argument, not the stated one—unless you genuinely can't tell what was intended.

The alternative is interpretive hostility: assuming the worst possible reading of everything your opponent says. This happens constantly in political discourse. Someone makes a nuanced point; their opponents extract the most extreme interpretation and attack that. The original speaker protests that they were misunderstood; the opponents accuse them of backtracking. Everyone ends up more polarized, and nothing has been learned.

Daniel Dennett, the philosopher, suggested a rule: before criticizing someone, you should be able to state their position so well that they say "Thanks, I wish I'd thought of putting it that way." Only then have you earned the right to critique. This is a high bar. But it's the bar that separates serious intellectual engagement from scoring points.

HOW TO STEELMAN IN PRACTICE

Steelmanning is a skill, and like any skill, it improves with practice. Here's a process that works:

Start by listening. Really listening—not waiting for your turn to talk, not mentally composing rebuttals, but attending to what your opponent is actually saying. This is harder than it sounds. Your brain generates counterarguments automatically; that's how minds work when they encounter threatening information. So you need techniques to interrupt that process.

Here are three that work. First, take notes in their language, not yours. Write down their exact phrases rather than your paraphrases. "They believe X is caused by Y" is your framing; their actual words might reveal something different. The mechanical act of transcription occupies the part of your brain that would otherwise be composing rebuttals. Second, ask questions that can only help their argument. Not "But what about Z?" (a disguised objection) but "What's the strongest evidence for your view?" or "What would change your mind?" These questions train your attention on understanding, not winning. Third, notice and name your resistance. When you feel a rebuttal forming—and you will—mentally label it: "There's a counterargument arising." Don't suppress it; that doesn't work. Just notice it, set it aside, and return to listening. You can retrieve it later. This is borrowed from meditation practice, and it's the closest thing to a cognitive trick for genuinely hearing views you find threatening.

Then articulate their position back to them. "So if I understand you correctly, you're saying that..." This step does two things: it checks your understanding (they can correct you if you're wrong), and it demonstrates respect (you've bothered to get it right). Often your opponent will add nuances when they hear their position reflected back—nuances that make the position stronger and harder to refute.

Now strengthen the argument. Where are the gaps? What's the best evidence they could have cited but didn't? What's the strongest philosophical foundation for their view? Imagine you were their debate coach—how would you improve their case? This is the hard part, because it requires genuinely trying to help the other side.

Only then do you respond. And when you respond, you respond to the steelmanned version—not the weaker original. "The strongest version of your argument, as I understand it, is... And here's why I still disagree..." This framing shows that you've done the work. Your opponent can't accuse you of misrepresentation. Your audience can see that you're engaging seriously.

Bryan Caplan, an economist known for heterodox views, developed what he calls the "Ideological Turing Test." The idea is simple: can you articulate your opponent's position well enough that a neutral observer couldn't tell whether you believe it or not? If you pass the test, you understand the view. If you fail—if your summary is obviously hostile or uncharitable—you haven't done the intellectual work.

THE INTELLECTUAL HONESTY DIVIDEND

Steelmanning pays dividends beyond any single argument. It builds a reputation for fairness that follows you across conversations, institutions, and years. When you consistently treat opposing views charitably, people trust you more—even people who disagree with you.

Consider the opposite. We all know people who can't state an opposing view without mocking it. Every summary comes loaded with sarcasm, every characterization drips with contempt. These people might win arguments in the short term through sheer aggression. But they don't persuade. They don't learn. And they don't earn the respect of anyone whose opinion is worth having.

Abraham Lincoln was a master steelmanner, though the term didn't exist in his time. Before the Cooper Union speech that launched his presidential campaign, he spent months researching the views of the Founding Fathers on slavery—not to misrepresent them, but to understand them precisely. When he argued that the federal government could restrict slavery in the territories, he did so by engaging with the strongest counterarguments, demonstrating he understood them, and then showing why they failed. His audience knew they were hearing an honest thinker, not a partisan hack.

The intellectual honesty dividend also compounds internally. When you practice steelmanning regularly, you get better at noticing your own weak arguments before you make them. You become more rigorous, more careful, more likely to change your mind when the evidence warrants it. This isn't comfortable—changing your mind rarely is—but it's how you get closer to truth.

Consider how skilled interviewers model this. Ezra Klein, before critiquing a guest's position, often restates it more clearly than they expressed it: "So your argument is that..." followed by a sharpened version. The guest typically responds with gratitude—"Yes, exactly!"—and the conversation advances from a higher baseline. This isn't just politeness. It's strategy. By demonstrating that he's understood the strongest version of their view, Klein earns the right to push back. His critiques land because they can't be dismissed as misunderstanding. The same technique works in any domain: job interviews, negotiations, difficult family conversations. Understanding someone else's position better than they do is the ultimate form of respect—and the ultimate preparation for persuading them otherwise.

WHEN STEELMANNING FAILS

Steelmanning isn't always appropriate. Like any technique, it has limits—and knowing those limits is as important as knowing the technique itself.

Some positions don't deserve steelmanning. If someone argues for genocide, slavery, or deliberate cruelty, you're under no obligation to help them make a better case. But the hard cases aren't these obvious ones. What about homeopathy? Flat-Earth beliefs? Vaccine hesitancy? Here's a decision framework: Ask whether the position, if widely adopted, would cause serious harm. Ask whether the person holding it seems genuinely open to evidence. Ask whether your steelmanning would give the position undeserved legitimacy. If harm is high, openness is low, and legitimacy risk is real, rejection may be the right response. Conspiracy theories about school shootings being staged? Don't steelman. Concerns about vaccine side effects from a worried parent? Steelman, then respond with evidence. The difference is harm potential, epistemic openness, and whether your engagement elevates or exposes the view.

Some interlocutors don't deserve steelmanning. If your opponent consistently argues in bad faith—misrepresenting your views, refusing to engage with your points, changing the subject whenever cornered—your charitable interpretation is being exploited. But how do you know it's bad faith and not just incompetence or different communication styles? Look for patterns: Does the person ever concede any point, no matter how minor? Do they accurately represent your position when summarizing it? Do they respond to your strongest arguments or only your weakest ones? Do their goalposts stay fixed or constantly move? Bad faith reveals itself through consistency—the person who never, ever updates, no matter what evidence appears, is probably not seeking truth. Incompetence looks different: the incompetent arguer sometimes concedes, sometimes gets your position right, sometimes engages with strong points. They're trying and failing, not refusing to try.

Some contexts don't reward steelmanning. In a courtroom, your job is to advocate for your client, not help the prosecution. In certain competitive debates, judges score on points won, not on intellectual virtue. In negotiations, strengthening the other side's position can cost you money. These are adversarial contexts where the rules are different. You should still avoid straw men—misrepresentation is always wrong—but you're not obligated to actively strengthen the opposing case. The key distinction: Are you seeking truth together, or competing for a prize? Steelmanning belongs to truth-seeking contexts.

And sometimes steelmanning can be a trap. If you spend all your time improving your opponent's arguments, you might never get around to making your own. Analysis paralysis sets in; you keep finding new nuances, new considerations, new ways the opposing view might be correct. At some point you need to commit. Steelmanning is preparation for argument, not a substitute for it.

THE PERSPECTIVE-TAKING EXERCISES

Competitive debaters have developed specific exercises for building the steelmanning muscle. In the final moments before stepping on stage—after weeks of research and preparation—experienced debaters pause their certainty. They take out a fresh sheet of paper and write down the four strongest arguments for the opposing side. Then they look over their own case through hostile eyes, identifying every flaw and weakness an opponent might exploit. Finally, they imagine a world in which they lose the debate. What would the judge say went wrong?

These "side-switching" exercises serve multiple purposes. They catch last-minute weaknesses in your own case. They prepare you for arguments you might otherwise be blindsided by. But most importantly, they create what practitioners call "wiggle room"—a small space of uncertainty that allows for flexibility, for listening, for the possibility that you've missed something. The exercises don't produce humility or empathy directly. They create conditions under which humility and empathy can emerge.

You don't need to be a competitive debater to practice this. Before any important argument—a difficult conversation with a partner, a proposal to your boss, a contentious family dinner—take five minutes for perspective-taking. Write down the strongest objections to your position. Look at yourself through your opponent's eyes. Imagine losing the argument and ask why. This preparation costs almost nothing and transforms how you show up in the conversation. You'll be less defensive, more responsive, and paradoxically more persuasive—because you've already grappled with the challenges your opponent might raise.

STEELMANNING YOURSELF

The deepest application of steelmanning is turning it on your own positions. What would the smartest, most informed critic say about your views? Not the loudest critic, not the meanest critic—the best one. What evidence would they cite? What logical problems would they identify? What blind spots would they expose?

This is excruciating. It means actively looking for reasons you might be wrong—when what you want is confirmation that you're right. It means taking seriously the possibility that your cherished beliefs are mistaken, your hard-won conclusions are flawed, your identity-defining positions are untenable. Most people can't do it. They don't even try.

But the people who can do it—who regularly subject their own views to the same scrutiny they apply to others—are the people worth listening to. They've earned their positions through rigorous self-examination. They've already encountered the best counterarguments and found their views could withstand them. When they speak, they speak with the confidence that comes from having genuinely tested their ideas.

Charlie Munger, Warren Buffett's longtime partner, famously said: "I never allow myself to have an opinion on anything that I don't know the other side's argument better than they do." That's a demanding standard. But it's the standard that separates genuine thinkers from people who merely have opinions.

This is where we've been heading all along—toward the idea that good argument isn't about victory but about truth. Steelmanning is the practice that embodies this ideal. You help your opponent because you want to defeat the best version of their case, not the worst. You help yourself by imagining the best criticisms of your own position. And in doing both, you move closer to understanding what's actually true—regardless of which side that truth favors.

Quick Tactic
The Steelman Starter

Before responding to any disagreement, complete these three steps:

  1. Reflect it back: "So if I understand you correctly, you're saying..." State their position until they say "Yes, exactly."
  2. Find the reasonable core: Ask yourself: "What would a smart, well-intentioned person have to believe to hold this view?" There's almost always something.
  3. Improve it: Before critiquing, offer: "And the strongest version of that argument might be..." They'll either accept your improvement or clarify further—both outcomes help.

Ninety seconds of steelmanning saves ninety minutes of talking past each other.

Quick Tactic
Steelmanning in SuperDebate Rebuttals

In competitive debate, steelmanning isn't just ethical—it's strategic. Judges reward debaters who engage with the strongest version of opposing arguments. Here's how to use it in your 3-minute rebuttal:

  1. Open with the steelman: "My opponent's best argument is [X]. Let me address that directly." This signals confidence.
  2. Grant what you can: "They're right that [narrow point]. But that actually supports my case because..."
  3. Attack the strongest version: "Even in its strongest form, this argument fails because [reason]."
  4. The killer line: "I've addressed their best argument. The rest of their case is even weaker."

Judges have seen debaters dodge, deflect, and strawman. When you steelman, you stand out. See Live Response Guide for opponent-type strategies and Dialogue Scripts for rebuttal examples.

In Chapter 20, we'll explore what it means to pursue this kind of victory—the philosopher's victory, where winning means getting closer to truth even if it means changing your mind. Steelmanning is the preparation. It trains you to value understanding over winning, evidence over ego, dialogue over dominance. It's not the easy path. But it's the path that leads somewhere worth going.

With steelmanning, we've completed Part III's exploration of argument technique—the practical arts of constructing cases, dismantling opposing ones, avoiding logical traps, questioning skillfully, and strengthening your opponent's position before engaging it. But technique alone isn't enough. The ancient skills must be applied in modern contexts: workplaces with their hierarchies and politics, classrooms with their power dynamics, digital platforms with their distortions, civic spaces where democracy itself hangs in the balance. Part IV turns to these applications—where the rubber of Greek wisdom meets the road of contemporary life.

Defeating a weak version of your opponent's argument proves nothing. Only the strongest version is worth engaging.

EXERCISES

Reflection

Your Unfair Summary

Think of a political or social view you strongly oppose. Write down your usual summary of that view—the one you'd give to friends who agree with you. Now look at it critically: is it a steelman or a strawman? What would supporters of that view say you're missing or distorting?

Practice

The Ideological Turing Test

Choose a position you disagree with and write a one-page defense of it—not a parody, but a genuine attempt to argue for it as persuasively as you can.

Success criteria: The original test requires showing it to someone who holds the view—but that's not always practical. Here's a self-check version: (1) Does your argument contain phrases like "they believe" or "supporters claim"? If so, you've failed—a true believer wouldn't hedge that way. (2) Have you included any qualifiers that only an outsider would add ("despite the obvious problems with...")? Remove them. (3) Could this argument appear, unedited, on a website that genuinely advocates this position? If the answer is yes, you've passed. If you can find an actual advocate to review it, even better—ask them: "Does this sound like something our side would write?" Their reaction tells you more than their words.

Challenge

Steelman Yourself

Pick a position you hold confidently. Now write the strongest possible critique of that position—the argument that would most shake your confidence if someone made it. Sit with the discomfort. Has your view survived intact? Changed? Become more nuanced? Write down what you learned.

Test Your Knowledge

Chapter 15 Quiz

Review what you've learned about steelmanning

Questions
To Pass
Your Best
Previous Next Esc Contents