It makes a big difference if the argument you're trying to steelman is logical but based on disputed assumptions or facts or whether the argument is illogical (or at least the logic is hard to parse).
It's really hard to do the latter, because you have to yada yada yada to get to the end position. In the former case it's helpful because it highlights that it's really the specific assumptions you're making that are at odds.
But then attempting to steelman the latter is a good exercise because it highlights where you have trouble making the logical leap. But it's the failure to steelman that you should note, not the success.
I agree 100% that Smith’s take here is somewhere between absurd and just stupid.
With that said, two additional comments on your take:
1) ‘Does every crank’s screwball theorizing need to be steelmanned? Maybe not,…”
Every competing theory need not be taken on, no. But if you DO address it, then yes you need to address the best reasonable version of it, not a strawmanned version.
2) “He also claims that trusting the marketplace of ideas is more humble than representing both sides of an issue yourself:”
IMO here you missed the most important point of all (though you allude to it earlier but don’t make it clear): exactly WHICH of the opinions in said marketplace of ideas on any given topic does he consider best?!?
The act of choosing said best one(s), done honestly, IS steelmanning!
The alternative of not choosing any at all and explicitly never mentioning the other side of the argument is to be an even more extreme version of the “role-playing attorney” he supposedly objects to himself!
If he wants to argue that he will always find and note said best alternatives in the marketplace of ideas instead of creating the steelman himself, I could even agree that that is a superior form of steelmanning to “rolling your own”. But of course he’s *not* making that argument, since that would be conceding that steelmanning is a good idea!
P.S. you *do* gotta give Smith props for the inspired Colossus pic he uses at the top of his piece.
I think you're right overall, with one reservation. Let me steelman the anti-steelman view. The context of Noah's article is a request to "steelman" Trump's policy ideas, and this is perhaps one area in which steelmanning doesn't make sense. If a policy is predicated on having a supposed effect and there's no reason to believe it would have that effect, then it would be untruthful to suppose for your opponent's sake that it really would do what they say it would (unless you're making a reductio ad absurdum argument).
This is a fair proviso. Some naïve opinions are both commonplace and based on flawed instincts, and there's probably a minimum threshold of logic and plausibility required for steelmanning to function. For the concept to be sensible, the steel man's constructor needs an assortment of counterarguments to choose from.
Yeah but… even if that was the “context”, Smith made plain he thought it applied “universally “, not just to the WaPo request for standalone steelmanned arguments.
Noah highlights one key point that has convinced me: in public square arguments are not really arguments. So using honest argumentation on behalf of the opponent, who, you suspect, is not going to pay back with the same respect, is giving up the game.
When there are emotionally charged sound-bite consumers, agreeing in public is dangerous and should be done with care.
It’s not great that that’s where we are, but one can easily lose having objectively better ground, because if emotional landscape we’re in.
In private clubs or scientific circles steelmanning is still a great and noble game, of course.
That's an interesting angle. I'm unsure whether steelmanning's benefits depend as much on a conducive atmosphere as you're suggesting, but I suppose I'm working from the concept of steelmanning as maneuver in an argument. It's def possible that steelmanning isn't a profitable tactic if you're really doing something else. The question is then whether that other thing is worth doing if it's incompatible with steelmanning.
Sorry, imo your argument *is* very fair in terms of what politicians themselves, or their surrogates in the run-up to an election, should do.
It is not reasonable at all, imo, for someone who holds themselves out as a public intellectual rather than as a pure partisan.
Of course, there are a few such “public intellectuals” out there - I’m thinking first and foremost of Paul Krugman - for whom being a public intellectual and being hyperpartisan are one and the same, but that is to the discredit of Krugman and decidedly not an argument in favor of simply arguing in a fundamentally partisan manner yourself.
Steel-manning is to actively attempt to understand a position in the light most favourable to the person who holds the position. It doesn't require anything from the steel-manner but an honest effort to be fair in their evaluation. It does not require that you fully understand or defend the position. That comes after they verify whether your evaluation is credible. You've straw-manned steel-manning.
It makes a big difference if the argument you're trying to steelman is logical but based on disputed assumptions or facts or whether the argument is illogical (or at least the logic is hard to parse).
It's really hard to do the latter, because you have to yada yada yada to get to the end position. In the former case it's helpful because it highlights that it's really the specific assumptions you're making that are at odds.
But then attempting to steelman the latter is a good exercise because it highlights where you have trouble making the logical leap. But it's the failure to steelman that you should note, not the success.
I agree 100% that Smith’s take here is somewhere between absurd and just stupid.
With that said, two additional comments on your take:
1) ‘Does every crank’s screwball theorizing need to be steelmanned? Maybe not,…”
Every competing theory need not be taken on, no. But if you DO address it, then yes you need to address the best reasonable version of it, not a strawmanned version.
2) “He also claims that trusting the marketplace of ideas is more humble than representing both sides of an issue yourself:”
IMO here you missed the most important point of all (though you allude to it earlier but don’t make it clear): exactly WHICH of the opinions in said marketplace of ideas on any given topic does he consider best?!?
The act of choosing said best one(s), done honestly, IS steelmanning!
The alternative of not choosing any at all and explicitly never mentioning the other side of the argument is to be an even more extreme version of the “role-playing attorney” he supposedly objects to himself!
If he wants to argue that he will always find and note said best alternatives in the marketplace of ideas instead of creating the steelman himself, I could even agree that that is a superior form of steelmanning to “rolling your own”. But of course he’s *not* making that argument, since that would be conceding that steelmanning is a good idea!
P.S. you *do* gotta give Smith props for the inspired Colossus pic he uses at the top of his piece.
I think you're right overall, with one reservation. Let me steelman the anti-steelman view. The context of Noah's article is a request to "steelman" Trump's policy ideas, and this is perhaps one area in which steelmanning doesn't make sense. If a policy is predicated on having a supposed effect and there's no reason to believe it would have that effect, then it would be untruthful to suppose for your opponent's sake that it really would do what they say it would (unless you're making a reductio ad absurdum argument).
This is a fair proviso. Some naïve opinions are both commonplace and based on flawed instincts, and there's probably a minimum threshold of logic and plausibility required for steelmanning to function. For the concept to be sensible, the steel man's constructor needs an assortment of counterarguments to choose from.
Yeah but… even if that was the “context”, Smith made plain he thought it applied “universally “, not just to the WaPo request for standalone steelmanned arguments.
Noah highlights one key point that has convinced me: in public square arguments are not really arguments. So using honest argumentation on behalf of the opponent, who, you suspect, is not going to pay back with the same respect, is giving up the game.
When there are emotionally charged sound-bite consumers, agreeing in public is dangerous and should be done with care.
It’s not great that that’s where we are, but one can easily lose having objectively better ground, because if emotional landscape we’re in.
In private clubs or scientific circles steelmanning is still a great and noble game, of course.
That's an interesting angle. I'm unsure whether steelmanning's benefits depend as much on a conducive atmosphere as you're suggesting, but I suppose I'm working from the concept of steelmanning as maneuver in an argument. It's def possible that steelmanning isn't a profitable tactic if you're really doing something else. The question is then whether that other thing is worth doing if it's incompatible with steelmanning.
Sorry, imo your argument *is* very fair in terms of what politicians themselves, or their surrogates in the run-up to an election, should do.
It is not reasonable at all, imo, for someone who holds themselves out as a public intellectual rather than as a pure partisan.
Of course, there are a few such “public intellectuals” out there - I’m thinking first and foremost of Paul Krugman - for whom being a public intellectual and being hyperpartisan are one and the same, but that is to the discredit of Krugman and decidedly not an argument in favor of simply arguing in a fundamentally partisan manner yourself.
Steel-manning is to actively attempt to understand a position in the light most favourable to the person who holds the position. It doesn't require anything from the steel-manner but an honest effort to be fair in their evaluation. It does not require that you fully understand or defend the position. That comes after they verify whether your evaluation is credible. You've straw-manned steel-manning.
Well I agree with all you wrote except perhaps the last sentence. At minimum you’re “You” subject there is vague.
B.P.S. Is not the one who has straw-manned steel-manning; Smith has.