Good Arguing: How to steelman (and why it’s hard)

In the last couple of posts I’ve been exploring different ways in which it’s possible to make bad arguments against someone’s position by ignoring the substance of their argument in favor of some distraction from it or distortion of it, even a completely fictional version of it. The tactic of constructing an inaccurate version of an argument in order to demolish it is called strawmanning.

Strawmanning is easy to do, and advantageous when the only people you care about impressing are people who already agree with you, and who also aren’t particularly concerned about you representing your opponent fairly– they just want to see you rip him/her apart, or at least a sufficiently convincing facsimile thereof. And since it’s very likely that the image of your opponent is already more of a caricature in the eyes of those who agree with you (that’s tribalism, in a nutshell), the chances are relatively low that someone on your side is going to pull back from basking in the warmth and comforting glow of the effigy which you’ve just set ablaze to tug at your sleeve and point out– hey man, that’s an effigy.

A strawman version of your opponent’s argument is easier to demolish for precisely the same reasons that the first little pig’s straw house was easy for the big bad wolf to demolish– it’s flimsy. It was constructed in haste with little thought put into it (who lives in a house made of straw, anyway?), and takes but a few forceful huffs and puffs and logic to blow it to smithereens. So if you, rhetorical big bad wolf that you are, could actually choose to have the person you’re arguing against live in a straw house rather than something sturdier, you would, wouldn’t you? It makes everything so. Much. Easier. And you’re angry, because damn that pig for having the gall to say…whatever horrible thing pigs say. Why should he get the benefit of a charitable, sturdy interpretation of his house I mean, argument?

Well, because that’s what logic– and fairness– demand. You want your opponent to engage the argument you’re actually making, rather than some shoddy imitation that’s easier to dismantle, so shouldn’t you extend the same consideration? And if his/her argument is really so pernicious and threatening, doesn’t that make it especially important to make sure that you’re addressing it accurately, in order to publically demonstrate its problems to every witness, so that they can avoid being taken in by it?  Does the group of people you care about convincing of the problems with your opponent’s argument include the opponent him/herself? And if not, shouldn’t it?

This is why steelmanning is so important. And so difficult. And so important.

Steelmanning is exactly what it sounds like– you turn the analogy of the strawman on its head, and imagine constructing a stronger, better version of your opponent’s argument. Perhaps even better than the one he/she initially constructed. You take the time to contemplate your opponent’s concerns, including the unspoken ones, and address them. You create the most convincing, best possible version of your opponent’s argument, and you lay it out for everyone to see. And then– only then– do you you show why it’s wrong.

To the best of my knowledge, use of the term “steelmanning” to refer to this practice originated with Chana Messinger. To quote her on the subject:

But Chana, you might say, I’m actually trying to get something done around here, not just cultivate my rationalist virtue or whatever nonsense you’re peddling. I want to convince people they’re wrong and get them to change their minds. Well, you, too, have something to gain from steelmanning. First, people like having their arguments approached with care and serious consideration. Steelmanning requires that we think deeply about what’s being presented to us and find ways to improve it. By addressing the improved version, we show respect and honest engagement to our interlocutor. People who like the way you approach their arguments are much more likely to care about what you have to say about those arguments. This, by the way, also makes arguments way more productive, since no one’s looking for easy rebuttals or cheap outs. Second, people are more convinced by arguments which address the real reason they reject your ideas rather than those which address those aspects less important to their beliefs. If nothing else, steelmanning is a fence around accidental strawmanning, which may happen when you misunderstand their argument, or they don’t express it as well as they could have. Remember that you are arguing against someone’s ideas and beliefs, and the arguments they present are merely imperfect expressions of those ideas and beliefs and why they hold them. To attack the inner workings rather than only the outward manifestation, you must understand them, and address them properly.

Now, of course, the concept of taking on the most robust version of your opponent’s argument, even if you have to construct it yourself, has been around a lot longer than the term “steelmanning” itself. You could simply call it arguing charitably. You could, as philosopher Daniel Dennett has been known to do, actually insert a stand-in for your opponent in the text of your own elucidation of your position, to fire objections and criticisms of that position in “real time,” giving you the opportunity to answer those criticisms. Of course, when you have multiple opponents, this means you probably won’t have the time and space to answer all of their potential criticisms. But again, you can choose the best of these and answer them– or at least, the best of them so far as you can honestly assess.

Dennett outlines the practice of charitable criticism in his recent book Intuition Pumps and Other Rules for Thinking, attributing it to Russian-American psychologist Anatol Rapoport:

Anatol Rapoport… once promulgated a list of rules for how to write a successful critical commentary on an opponent’s work. First, he said, you must attempt to re-express your opponent’s position so clearly, vividly and fairly that your opponent says “Thanks, I wish I’d thought of putting it that way.” Then, you should list any points of agreement (especially if they are not matters of general or widespread agreement), and third, you should mention anything you have learned from your opponent. Only then are you permitted to say so much as a word of rebuttal or criticism. I have found this a salutary discipline to follow– or, since it is challenging, to attempt to follow. When it succeeds, the results are gratifying: your opponent is in a mood to be enlightened and eagerly attentive.

Sounds good, right? Sounds like a total “best practice” for argumentation. This is something everybody should be doing right? So….why is, when we look around, we see so few people actually doing it? So few people, when deciding how to depict a position they oppose, selecting materials by reaching immediately for the straw rather than the steel?

Well, I know one thing with certainty– it’s not because they’re incapable.

There is no level of intelligence or education at which a person moves beyond having the incentive to strawman. The incentives, as I’ve described, include that that it’s easier and faster, but also there is the fact that it’s simply more satisfying to pin down and torture a good straw man when you’re angry, and when you’re speaking to people who are already angry for the same reason that you are, or whom you would like to make angry for the same reason.

A rhetorical crime has been committed, and by golly we want someone to answer for it. We want to haul in some guilty party and hold them to account, and when the guilty party is an argument, the penalties for getting the wrong man tend to be few. Violation of due process of the laws of logic for suspect arguments is not an offense for which most really suffer. We’re biased in a multitude of ways, perhaps most predominantly in favor of our own sense of being right. Being right feels good. Righteous indignation feels good. Watching people whose righteous indignation you share royally trouncing an argument that you find offensively wrong?  Gosh, that’s nice. That’s why we value an intelligent, caustic, sardonic ranter on our side so highly. Perhaps more than is really healthy on a sociological level, we value these people. There’s a reason for that.

But there ways to make steelmanning a great deal easier and more likely. Here are some I can identify:

1. A polite disagreement, where passions are low. 
2. Time is not a highly significant factor. This suggests that strawmanning is much more likely in verbal debates than in print.
3. Opponents know each other. It’s much easier to represent your opponent’s position charitably when you’re familiar with his/her views on other things which aren’t directly related to the topic of contention. 
4. Space, or rather the lack thereof, is not a significant factor. If you take the time to recreate a better version of your opponent’s argument before answering it with your own, there had better be some room to do it. Which means that you’re more likely to find steelmanning in a book than an essay or blog post. A blog post or essay than a verbal argument. A verbal argument than a sound bite. 
5. A reasonable expectation of continued interaction, on some level. 

Steelmanning is possible for all of us, though. It’s a best practice for all of us. We’re not terrible people if we fail to do it, but it’s something to aim for. It’s good arguing.