The 'steelman challenge': what is it and how can you use it to disagree better in meetings

Most disagreements follow the same unhelpful pattern. The steelman technique disrupts this pattern and gets us back into healthy debate.

You have heard me talk many times about how healthy, helpful conflict is crucial for developing high quality outcomes in meetings and therefore in teams. The question I get all the time is HOW do we do this - in the real world, in a real meetings with 'tricky' people?

As a reminder, previous techniques I've shared include how to understand and diffuse conflict via the Drama Triangle, get better at rupture and repair, this 3-step conversation pattern for handling conflict and this way to turn unhelpful conflict into helpful conflict.  

Today, I want to introduce you a technique called 'steelman challenge'. 

It's something you can use yourself (without specifically naming it) or you can introduce it as an explicit Thing you are going to use in a project or team or whatever. 

What is the steelman challenge?

A classic disagreeing fallacy is the 'strawman challenge'. You take the weakest or most extreme version of what someone said, ignore the nuance and the reasoning behind it, and knock that down instead of engaging with the real argument. This is a bad faith argument.

Steel manning is the deliberate opposite. Before you counter someone's argument, you have to construct the strongest possible version of it and express it, crucially to their satisfaction. That means understanding not just what they are saying, but why they are saying it, what evidence or logic underpins it, and what legitimate concerns are driving their position. You then articulate all of that back to them, and they have to confirm that you have represented their argument fairly before you make your counter argument. This is a good faith argument.

This is a new name for an age old technique. The philosopher Daniel Dennett argued that you should be able to restate the other person's position so clearly and fairly that they would say "I wish I'd put it that way myself." 

Why it works

1. It forces genuine comprehension. You cannot steelman an argument you have not properly understood. That alone slows down reactive disagreement and improves the quality of thinking in the room.

2. It brings down defences. When someone feels their argument has been genuinely understood, they become much more open to hearing a counter-argument. Feeling misrepresented is one of the fastest routes to a conversation breaking down.

3. The process of steel manning almost always surfaces areas of agreement that would otherwise stay hidden. When you straw man, you strip out the parts of the argument that are reasonable because those parts make it harder to disagree. When you steelman, you have to include them. This makes the actual point of disagreement much clearer and usually much narrower than it first appeared.

The disagreement stops being "I think you are wrong" and becomes something much more specific, like "I agree with your analysis of the problem but I think you are underestimating the cost of this solution" or "we want the same outcome but we are prioritising different risks."

THIS is how we disagree well at work.

If you're not disagreeing in meetings, you should be

I've worked with 100s of teams and organisations and high conflict people are the exception not the norm in most knowledge-based workplaces. I see lots of nice, smart, principled people doing their best. This doesn't mean they agree on everything but often those disagreements don't truly surface in meetings. 

The problem is, they will find somewhere else to emerge, whether that is the water cooler in the good old days or more likely the DMs these days. The straw man challenge is more likely happening inside people's heads or in back channels, and this matters a lot. We have a lot of bad faith thoughts under the surface of good faith meetings!

The steelman challenge is a way to bring that disagreement into the open in a way that low-conflict groups can tolerate. 

How to use it

Let's explore how to steelman challenge in the real world. You have two main options:

1. You can use it yourself without referring to the name or even specifically calling out what you're doing.

"Before I push back, let me make sure I've understood your argument. What you're saying is [X], because [Y], and the thing you're most concerned about is [Z]. Is that right?"

"I want to acknowledge what's strong about this position before I offer a different view. It sounds like XYZ - is that right?"

The test is always the same: have you summarised their argument to their satisfaction? If the answer is no, you are not ready to disagree yet.

2. You can introduce it as a new norm for groups to use when they need it - this time, referring to it by name. 

Name the problem first: Say something like: "I think we are sometimes not very good at disagreeing - or we don't disagree because we don't want to offend each other?" Let people respond. You want them to acknowledge the problem before you offer the tool.

Explain the concept simply: We want to speak up and counter arguments more but first we will summarise the other person's position to their satisfaction - this is called a 'steelman challenge'. 

Practice it before you need it: The first time people try this should not be in a live, high-stakes disagreement. Run a quick practice exercise on a low-stakes topic first, so people know what it feels like before they need it for real.

Agree the signal: "Can we try a steelman summary before we share a different view?" "Can you say back what you think they're arguing first?"

Model it yourself: The next time you disagree with something, say out loud: "Before I share my thoughts, let me make sure I've understood the argument." People learn this norm by watching it work.

But what then?

You've summarised the other person's argument to their satisfaction and you shared yours. Now what?! How does that actually help you move forward? 

Find the overlap: "So where do we actually agree?" Say it out loud and list those things. This narrows the disagreement down to what it actually is, which is almost always smaller than it first appeared.

Name the real point of difference: Most disagreements, once the straw man is gone, turn out to be one of these:

- We agree this is true but disagree on whether it is important enough to act on.

- We agree on the goal but disagree on the best route there.

- We agree on the problem but disagree on who should own it.

"So it sounds like we actually agree on the problem. The disagreement is about priority, is that right?"

What kind of decision is it? Some disagreements need resolving right now. Others can be parked, tested, or decided by one person with the mandate to do so. Ask: "Does this need to be resolved today, or can we agree a way to test it and come back?"

Whatever the outcome, close the loop explicitly. "We've agreed X. We're going to test Y. Z is owned by this person."