I’m not done with the post yet, but this part really jumped out at me.
Second point is a doozy, and it’s that you can’t look at intent when diagnosing frame control. As in, “what do they mean to do” should be held separate from “what are the effects of what they’re doing”—which I know is counter to almost every good lesson about engaging with people charitably.
I think you’re right in a narrow way but mostly wrong here. The narrow way in which you seem right is that (someone’s intent) and (someone’s impact) are indeed separate quantities. But someone having good intent—or someone seeming to have good intent, if one can generally discern this with above-random accuracy—means that their actions are more likely optimized to have good effects, and so these two quantities are generally correlated.
In this section, your language conflates two possible scenarios (by my reading). In the first, we condition on “leader X seems to have good intent.” In the second, we condition on “me and my friends are talking about how leader X is deeply flawed and perceptive, and the things he did that hurt people were either for their own good, or an unintentional byproduct of him genuinely trying to do good.” The second scenario is very different.
In the first scenario, we surely have P(bad frame-controller | leader X seems to have good intent) < P(bad frame controller | leader X does not seem to have good intent).
In the second scenario, P(bad frame-controller | leader X is flawed but seems well-meaning, [other red flags]) should still be less than P(bad frame-controller | leader X is flawed but does not seem well-meaning, [other red flags]), but I think that the probability of something bad going on should be high either way.
And so as far as I can tell, intent does matter for the beliefs you arrive at, and I am very very wary that this post claims otherwise in. “Frame control” has the potential to be an argumentative superweapon.
This all might sound pretty dark, like I’m painting a reality where you might go around squinting at empathetic, open, caring people who have zero ill intent whatsoever and trying to figure out how they are ‘actually bad.’ And this is kind of true, but if only because “I am an empathetic, open, caring person with zero ill intent” is exactly the kind of defense actual frame-controllers inhabit.
Surely we have P(awful frame control | They seem like an empathetic, open, caring person with zero ill intent) << P(awful frame control | They do not seem like an empathetic, open, caring person with zero ill intent)?
The vast majority of good people with good intent aren’t doing any significant kind of frame control; my point is just that “good person with good intent” should not be considered a sufficient defense if there seems to be other elements of frame control present.
But then you back off the original claim that you can’t look at intent, and merely say that good intent is not sufficient to conclude that they aren’t doing this horrible frame control. OK. I agree. But this feels like a motte and bailey.
Rereading the portion in question to make sure I’m not missing something, you write:
And so, when evaluating frame control, you have to throw out intent. The question is not “does this person mean to control my frame,” the question is “is this person controlling my frame?”. This is especially true for diagnosing frame control that you’re inside of, because the first defense a frame controller uses is the empathy you hold for them.
I think “be on the lookout for ways people can weaponize your empathy” seems wiser to me than “throw out intent.”
The way I understand the intent vs. effect thing is that the person doing “frame control” will often contain multitudes: an unconscious, hidden side that’s driving the frame control, and then the more conscious side that may not be very aware of it, and would certainly disclaim any such intent.
I’m not done with the post yet, but this part really jumped out at me.
I think you’re right in a narrow way but mostly wrong here. The narrow way in which you seem right is that (someone’s intent) and (someone’s impact) are indeed separate quantities. But someone having good intent—or someone seeming to have good intent, if one can generally discern this with above-random accuracy—means that their actions are more likely optimized to have good effects, and so these two quantities are generally correlated.
In this section, your language conflates two possible scenarios (by my reading). In the first, we condition on “leader X seems to have good intent.” In the second, we condition on “me and my friends are talking about how leader X is deeply flawed and perceptive, and the things he did that hurt people were either for their own good, or an unintentional byproduct of him genuinely trying to do good.” The second scenario is very different.
In the first scenario, we surely have P(bad frame-controller | leader X seems to have good intent) < P(bad frame controller | leader X does not seem to have good intent).
In the second scenario, P(bad frame-controller | leader X is flawed but seems well-meaning, [other red flags]) should still be less than P(bad frame-controller | leader X is flawed but does not seem well-meaning, [other red flags]), but I think that the probability of something bad going on should be high either way.
And so as far as I can tell, intent does matter for the beliefs you arrive at, and I am very very wary that this post claims otherwise in. “Frame control” has the potential to be an argumentative superweapon.
Surely we have P(awful frame control | They seem like an empathetic, open, caring person with zero ill intent) << P(awful frame control | They do not seem like an empathetic, open, caring person with zero ill intent)?
But then you back off the original claim that you can’t look at intent, and merely say that good intent is not sufficient to conclude that they aren’t doing this horrible frame control. OK. I agree. But this feels like a motte and bailey.
Rereading the portion in question to make sure I’m not missing something, you write:
I think “be on the lookout for ways people can weaponize your empathy” seems wiser to me than “throw out intent.”
The way I understand the intent vs. effect thing is that the person doing “frame control” will often contain multitudes: an unconscious, hidden side that’s driving the frame control, and then the more conscious side that may not be very aware of it, and would certainly disclaim any such intent.