“To actually believe that you’ve made a steel man, not a straw man, the person you’re arguing with would have to admit that you’ve created a stronger argument for their own position than they could.”
That’s rather obviously false. But let’s consider the steel man where “could” is replaced with “happened to have done in the particular case being replied to” which makes the claim fairly straightforward and true rather being nearly exactly backwards.
As Hugh observes in his reply it is almost always pointless to use the steel man concept in adversarial debate. Even in the rare case where the person agrees they still find it condescending (because it approximately as close to the literal meaning of condescending as it is possible to be). The best uses of steel men that I have seen is when someone takes an argument that is for some reason important or valuable and presents it to an audience in an improved version and then responds to the improved argument. The original arguer need not be involved at all.
Do note that my reply was not intended as a steel man at all. It was made as a response to mwengler with the intended audience of mwengler and anyone who, like myself, would read the quote mwengler made more charitably in the expanded context. I assigned a probability of about 0.65 that your intent was at least partially influenced by the additional details in the example beyond the need for learning. (p = 0.65 is pretty damn high for confidence assigned by me to for an interpretation task.)
(I disagree that just because someone happens to give me a boner it’s a bad idea to be friends with them, so long as we get along well.)
I didn’t claim that either (ie. that is a straw man). I have plenty of friends who have happened to “give me a boner” at some point and there is no particular problem with that. In fact I merely weakened the claim you made (or implied). Consider:
IF “ey needs to figure out how to be friends with her” THEN odds are there are other people who would make better friends than her.
If “ey needs to figure out how to be friends with her” AND “there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama” THEN odds are there are other people who would make better friends than her.
If “ey needs to figure out how to be friends with her” AND “there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama” AND “she is an accountant” THEN odds are there are other people who would make better friends than her.
In other words while I don’t agree fully with the advice to avoid social relationships that require work (for the reasons mwengler has explained) I do agree that the principle applies in many cases. In particular, for those people that mwengler is talking about—those for whom ALL friendships take effort due to weaker social skills, etc—there will most likely be alternative effortful friendship opportunities that at least don’t have the additional overhead of “(sexual) relationship drama without (sexual) relationship”.
But… Is it harder to be friends with people one is attracted to than with people one is not attracted to, in the real world (as opposed to stereotypes and Hollywood movies)? ISTM that, if anything, IME it’s the other way round (though the effect is smaller when controlling for age and gender), which is what I’d theoretically expect given that there is such a thing as the halo effect.
If “ey needs to figure out how to be friends with her” AND “there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama” AND “she is an accountant” THEN odds are there are other people who would make better friends than her.
Huh? The conjunction fallacy doesn’t apply to the right of the pipe—whereas P(AB|C) cannot possibly be greater than P(A|C), P(C|DE) can be less than, equal to, or greater than P(C|D). Am I missing something?
(In this particular example, I’d guess (with low confidence) that for A=“there are other people who would make better friends than her”, B=“ey needs to figure out how to be friends with her”, C=“there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama”, and D=“she is an accountant”, P(A|BCD) is slightly but not terribly lower than P(A|BC), by a reasoning that would be politically incorrect to fully explain but involves, among other things, looking at where “Accounting occs” are on this chart and wild-ass extrapolation from my personal experiences. :-))
The conjunction fallacy doesn’t apply to the right of the pipe
The intended meaning of the link was “generalised lesson of reasoning with conjunctions”. Since it is indeed possible to reformulate the message from the “IF THEN” format to probability assignments I can see how this could be misleading.
(I removed the link and now endorse the unadorned text.)
That’s rather obviously false. But let’s consider the steel man where “could” is replaced with “happened to have done in the particular case being replied to” which makes the claim fairly straightforward and true rather being nearly exactly backwards.
As Hugh observes in his reply it is almost always pointless to use the steel man concept in adversarial debate. Even in the rare case where the person agrees they still find it condescending (because it approximately as close to the literal meaning of condescending as it is possible to be). The best uses of steel men that I have seen is when someone takes an argument that is for some reason important or valuable and presents it to an audience in an improved version and then responds to the improved argument. The original arguer need not be involved at all.
Do note that my reply was not intended as a steel man at all. It was made as a response to mwengler with the intended audience of mwengler and anyone who, like myself, would read the quote mwengler made more charitably in the expanded context. I assigned a probability of about 0.65 that your intent was at least partially influenced by the additional details in the example beyond the need for learning. (p = 0.65 is pretty damn high for confidence assigned by me to for an interpretation task.)
I didn’t claim that either (ie. that is a straw man). I have plenty of friends who have happened to “give me a boner” at some point and there is no particular problem with that. In fact I merely weakened the claim you made (or implied). Consider:
IF “ey needs to figure out how to be friends with her” THEN odds are there are other people who would make better friends than her.
If “ey needs to figure out how to be friends with her” AND “there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama” THEN odds are there are other people who would make better friends than her.
If “ey needs to figure out how to be friends with her” AND “there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama” AND “she is an accountant” THEN odds are there are other people who would make better friends than her.
In other words while I don’t agree fully with the advice to avoid social relationships that require work (for the reasons mwengler has explained) I do agree that the principle applies in many cases. In particular, for those people that mwengler is talking about—those for whom ALL friendships take effort due to weaker social skills, etc—there will most likely be alternative effortful friendship opportunities that at least don’t have the additional overhead of “(sexual) relationship drama without (sexual) relationship”.
But… Is it harder to be friends with people one is attracted to than with people one is not attracted to, in the real world (as opposed to stereotypes and Hollywood movies)? ISTM that, if anything, IME it’s the other way round (though the effect is smaller when controlling for age and gender), which is what I’d theoretically expect given that there is such a thing as the halo effect.
Huh? The conjunction fallacy doesn’t apply to the right of the pipe—whereas P(AB|C) cannot possibly be greater than P(A|C), P(C|DE) can be less than, equal to, or greater than P(C|D). Am I missing something?
(In this particular example, I’d guess (with low confidence) that for A=“there are other people who would make better friends than her”, B=“ey needs to figure out how to be friends with her”, C=“there is evidently a significant level of impotent sexual interest involved that is already creating relationship drama”, and D=“she is an accountant”, P(A|BCD) is slightly but not terribly lower than P(A|BC), by a reasoning that would be politically incorrect to fully explain but involves, among other things, looking at where “Accounting occs” are on this chart and wild-ass extrapolation from my personal experiences. :-))
The intended meaning of the link was “generalised lesson of reasoning with conjunctions”. Since it is indeed possible to reformulate the message from the “IF THEN” format to probability assignments I can see how this could be misleading.
(I removed the link and now endorse the unadorned text.)