The last lesson (13.1.1) argued that the mark of a mature sociological mind is the ability to argue every side of the great debates at its strongest, then judge which lens fits the case. That is easy to say and hard to do — so this lesson turns the ideal into a technique . It is the most practical lesson in the whole course: a method for genuinely inhabiting a position you do not hold, and a live demonstration of it on the debates themselves.
This matters because "argue all sides" is one of those phrases everyone nods at and almost no one can actually perform. Most people, asked to state an opposing view, produce a straw man — a weak, distorted version that is easy to knock down and that its actual holders would not recognise. Producing the steel man — the strongest, most sympathetic version, the one its ablest defender would embrace — is a real skill with real steps, and it is trainable. Learning it does something profound: it is the difference between an education that makes you more certain and more tribal, and one that makes you wiser — able to hold a firm view while genuinely understanding what the other side is protecting. This is the technique that turns knowledge into judgement.
The test that Mill set, and almost everyone fails.
The philosopher John Stuart Mill posed a challenge that quietly separates real understanding from the imitation of it. He wrote that anyone who knows only their own side of an argument knows little of even that — because they have never had to meet the strongest objections, never had to see why intelligent people believe the opposite, and so hold their view as a prejudice they happen to be right about rather than as a conviction they have earned.
Mill's test is brutal and simple: can you state your opponent's position so well, so fully and so fairly, that they would say "I could not have put it better myself"? Not a version they would wince at. A version they would thank you for.
Almost everyone fails this test, and here is why it matters. When you argue only against the weak version of a view — the straw man — you win easily and learn nothing, and you come away more convinced you are right and the other side is foolish. But you have defeated a scarecrow, not the actual argument, which stands untouched and believed by intelligent people you have just written off. The straw man is comforting and useless; the steel man is uncomfortable and transformative. Because the moment you can state the other side at its strongest, one of two things happens: either you discover it is stronger than you thought and you revise, or you defeat even its best version and now hold your own view with earned, tested confidence instead of borrowed prejudice. Both outcomes make you wiser. The straw man makes you neither.
This lesson is training for Mill's test — the technique of building steel men rather than straw men, applied to the debates that define sociology and to any disagreement you will ever face.
The plain-words core: straw man versus steel man
Everything in this lesson rests on one distinction, so let us make it precise.
A straw man is a weak, distorted, or extreme version of a position, built (often unconsciously) to be easy to attack. It usually takes the opponent's view, strips out its qualifications, ignores its best evidence, attaches its dumbest possible motive, and states it in a way no thoughtful holder would accept. Defeating it feels like winning and teaches nothing.
A steel man (or "steelman") is the strongest possible version of a position — the one you would build if you were its most intelligent, most informed defender. To steelman a view you must: state it in its most reasonable form, include its best evidence and sharpest arguments, grant it its strongest qualifications, attribute to its holders their most sensible motives, and remove every weakness that is not essential to it. You steelman a view when its actual holders would read your version and say "yes — that is exactly what I mean, and you've put it well."
The discipline is called the principle of charity : interpret others' positions in the most reasonable way their words can bear, not the least. Not because you are being nice, but because it is the only way to actually learn — you cannot test your view against an argument you have secretly weakened, and you cannot understand why intelligent people disagree with you until you can state their reason as they would.
The method: four steps to inhabit a position
Steelmanning is not a mood; it is a procedure. Here is the procedure, in four steps you can actually run on any position you do not hold.
Step one — Assume the holder is intelligent and informed. Begin from the premise that thoughtful, well-informed people hold this view for reasons , not from stupidity or malice. If your explanation of why someone believes X is "they're dumb" or "they're evil" or "they just want to protect their interests," you have not understood; you have dismissed. Find the version of them that is as smart as the smartest person you know who actually holds it.
Step two — Find the genuine truth it guards. Every serious position exists because it is protecting something real that a rival position neglects (13.1.1). Ask: what real thing is this view right about? what does it see that its opponents miss? Structure-theory guards the reality of constraint; agency-theory guards the reality of choice. Until you can name the truth a position guards, you cannot steelman it, because you do not yet know what it is for .
Step three — Build its best case in its own terms. State the position using its concepts, its strongest evidence, its framing — not yours. Argue it the way a believer would, from the inside, as if you had to persuade a skeptic and your reputation depended on it. Include the qualifications real defenders make; do not argue the crude version they would disown.
Step four — Only then, respond. Now — and only now — you may push back. But your objection now lands on the real argument, not a scarecrow, and it will be a far better objection for it, because you built it against the strongest version. Often you will find your objection needs to be more careful than you thought — and sometimes you will find it does not survive contact with the steel man, which is itself the most valuable discovery you can make.
Steelmanning is not the same as agreeing — and it does not mean every side is equally right.
Two misunderstandings must be cleared, because they scare people away from the technique.
First, steelmanning a view does not commit you to holding it. You can build the most powerful possible case for a position and still conclude it is wrong — indeed you are now far better placed to conclude that, because you have defeated its best form rather than its worst. Arguing all sides is not spineless relativism; it is the precondition of a well-founded judgement. The person who ends up disagreeing after steelmanning holds a far stronger position than the one who disagreed without it. You hold your view more firmly, not less — but now for reasons you have earned.
Second, "argue all sides" does not mean all sides are equally valid. Some positions really are better supported than others; some really are wrong; a few really are straw all the way through (there is no steel man of a claim that is simply false or made in bad faith, and pretending there is would be its own dishonesty). The principle of charity says build the strongest version that the position can honestly bear — not invent strength that is not there. Steelmanning a genuinely empty claim just reveals, rigorously, that it is empty. So the technique does not flatten all views into equal merit; it tests them fairly, which sometimes confirms that one side is simply stronger — a conclusion you can now trust because you reached it charitably rather than by rigging the fight. Fair testing and equal validity are different things; the technique gives you the first, not the second.
A live demonstration: one debate, both sides steelmanned
Method is empty without practice, so let us actually do it — take one debate from 13.1.1 and steelman both sides in turn, so you can feel what inhabiting a position is like.
Consensus vs conflict, argued from the inside — twice.
The consensus case, steelmanned (argued as its ablest defender): "Look at what societies actually do: millions of strangers cooperate every day, keep promises, obey rules no one is enforcing, and feel real solidarity with people they will never meet. This is astonishing and it is real — society is not a war of all against all but a moral community held together by genuinely shared values, which most people internalise and most of the time uphold (Part 10's blackout finding — order holds without the police). A view that sees only conflict cannot explain why society mostly works , why people mostly cooperate, why order is the norm and breakdown the exception. The shared values are not a trick; they are the real glue of social life, and a sociology that cannot see them is blind to the most basic fact about societies — that they cohere."
The conflict case, steelmanned (argued as its ablest defender): "Look at who those 'shared values' actually serve. The value that says 'work hard and accept your place' benefits those at the top; the 'consensus' is often the worldview of the powerful, internalised by those it disadvantages (the third face of power, 11.1.1). Beneath the surface calm run real, structured conflicts of interest — over the surplus, over whose story counts, over who rules whom — and the 'order' the consensus theorists admire is often domination that has succeeded so well it no longer needs force. A view that sees only shared values mistakes the quiet of the well-managed for the peace of the genuinely agreed, and cannot explain domination, exploitation, or why the 'consensus' so reliably favours some over others."
Notice what just happened. Each case was argued at full strength, in its own terms, and each is genuinely compelling — you can feel the pull of both. Neither is a scarecrow. And now, having built both steel men, you are finally equipped to judge: not "which team am I on?" but for this society, this question, which force is doing more work — and where exactly does each one's truth end? You cannot ask that question well until both sides stand at their full height in your mind. That is arguing all sides — and it is a technique you just watched performed, not a slogan.
The tell of a straw man — how to catch yourself doing it.
Steelmanning is a skill, and the first skill is catching yourself failing at it. Here are the tells that you are arguing a straw man rather than a steel man — watch for them in your own thinking, because they are almost invisible from the inside:
You feel contempt. If stating the other side makes you feel how stupid or bad they are, you are almost certainly arguing the straw version. The steel man does not feel contemptible; it feels troubling , because it has a real point.It was easy. If the opposing view fell over at the first push, you pushed on a scarecrow. Real positions, held by intelligent people for a century, do not collapse that easily.No real holder would recognise it. If an actual, thoughtful believer would say "that's not what I think at all," you have built a straw man, however accurately you feel you have described them.You attributed a dumb or evil motive. "They just believe it because they're selfish / naïve / brainwashed" is the signature of a straw man. The steel man attributes the best plausible motive and the smartest plausible reasoning. The corrective is Mill's test, run honestly: would the ablest holder of this view thank you for how you stated it? If not, you have more work to do — and the discomfort of doing that work, of making the other side strong in your own mind, is precisely the discomfort that separates real understanding from the tribal imitation of it. When it feels hardest to be fair is exactly when being fair teaches you the most.
Why this matters
Because the ability to argue all sides is not merely a scholarly nicety — it is, increasingly, one of the rarest and most needed capacities in public life, and sociology trains it as well as any discipline can. We live surrounded by the opposite: by straw-manning as a way of life, where every disagreement is met by attributing the stupidest motive and the weakest argument to the other side, where "understanding" the opposition means only knowing how to defeat its worst version, where being educated too often means being more fluent in contempt. Against all of that, the discipline of the steel man is a quiet radicalism. It says: before you may reject a view, you must first be able to hold it — fully, sympathetically, at its strongest — in your own mind. And it turns out that most of what passes for confident disagreement cannot survive that requirement.
For you personally, this is the technique that keeps an education from curdling into mere sophistication — the ability to argue cleverly for whatever you already believed. The steel man forces the opposite: it makes you inhabit what you do not believe, and in doing so it does two things at once. It makes you more honest , because you can no longer pretend the other side is foolish once you have argued it well. And it makes you more free , because a view you have tested against its strongest opposition is a view you actually own , rather than one you inherited or absorbed. The person who can argue every side is not a person without conviction; they are the only kind of person whose convictions are fully their own — earned, tested, and held with the humility of someone who knows exactly what the other side is right about. That is the deepest thing sociology's unresolved debates have to teach, and it reaches far past sociology into every disagreement you will ever have.
This closes Topic 13.1. The next topic (13.2) turns the same critical rigour outward — from arguing positions to reading and dismantling the actual work sociologists produce: how to read a theorist, an article, and a data study, and how to take an argument apart to see whether it holds.
Straw man — a weak or distorted version of a position, built to be easy to attack; defeating it feels like winning but teaches nothing.
Steel man — the strongest, most reasonable version of a position, including its best evidence and qualifications, that its ablest defender would embrace.
Principle of charity — the discipline of interpreting others' positions in the most reasonable way their words can bear, as the precondition of genuine understanding and fair testing.
Mill's test — the standard that you understand a view only if you can state it so well its holders would say you put it better than they could.
Inhabiting a position — arguing a view from the inside, in its own concepts and framing, as a committed defender would, rather than describing it from outside.
Fair testing vs equal validity — the distinction between assessing every position at its strongest (which the technique provides) and treating all positions as equally correct (which it does not imply).
Try it yourself:
One — Run Mill's test on a real disagreement. Take a view you firmly oppose (political, moral, sociological). Write its steel man using the four steps — intelligent holder, the truth it guards, its best case in its own terms. Then ask honestly: would its ablest defender thank you for this version? Revise until they would. Notice how the exercise changes your relationship to your own view.
Two — Catch a tell in yourself. Over the next week, notice one moment when you dismiss a view with contempt or an easy win. Stop and ask: am I arguing the straw man? Then build the steel man of that view on the spot. The contempt is the alarm bell; the steel man is the response.
Three — Steelman both sides of one live debate. Pick any debate you care about and argue both sides at full strength, in turn, as the consensus/conflict demonstration did — until you can feel the genuine pull of each. Only then form your judgement. You will find the judgement is better, and held more honestly, than any you could have reached without inhabiting both.
What's next: Topic 13.2 opens with 13.2.1, How to Read a Theorist, an Article, and a Data Study — we turn from arguing positions to reading the work itself, and learn that the three main kinds of sociological writing must each be read in a completely different way, with different questions and different traps to watch for.