I like Michael’s distinction between the style and core of an argument. I’m editing this paragraph to clarify the way in which I’m using a few words. When I talk about whether an argument is actually combative or collaborative, I mean to indicate whether it is more effective at goal-oriented problem-solving or at achieving political ends. By politics, I mean something like “social maneuvers taken to redistribute credit, affirmation, etc. in a way that is expected to yield selfish benefit”. For example, questioning the validity of sources would be combative if the basic points of an argument held regardless of the validity of those sources.
Claims like “EA would attract many additional high quality people if women were respected” or “social justice policing would discourage many good people from joining EA” are, while true, basically all combative, and the framing of effectiveness is just helping people self-deceive into thinking they’re motivated by impact or truth. They’re using a collaborative style (the style of caring about impact/truth) to do a combative thing (politics, in the wide definition of that word).
Ultimately, I can spin the observation that these things are combative into a stylistically collaborative but actually combative argument for my own agenda, so everything I’m saying is suspect. To illustrate: the EA phrase “politics is the mindkiller” is typically combatively used in this way, and I have the ability to do something similar here. “Politics is the mindkiller” is the mindkiller, but recognizing this won’t solve the problem, in the same way recognizing politics is the “mindkiller” doesn’t.
People can smell this, and they’d be right to distrust your movement’s ability to think clearly about impact, if you’re using claims of impact and clearer thinking to justify your own politics. People who are bright enough to figure this out are typically the ones I’d want to be working with.
Yeah, you all have a problem with how you treat women and other minority groups. Kelly did a lot of work in order to point out a real phenomenon, and I don’t see anyone taking her very seriously. You let people who want to disparage women get away with doing so by using a collaborative “impact and truth” discussion style to achieve combative, political aims. That’s just the way the social balance of power lies in EA. People would use “impact and inclusivity” as a collaborative style to achieve political aims if the balance of power were flipped. Plausibly there’s an intermediate sweet spot where this happens less overall, though shifting the balance of power to such spots is never a complete solution. I suspect a better approach would be to get rid of the politics first; this will make it easier to make progress on inclusivity.
The norm of letting people stylize politics with talk of impact and truth is deeply ingrained in EA. It’s best to work outside the social edifice of EA, if you want to think clearly about impact and truth. Which feels like a shame, but isn’t too bad if you take the view that good people will eventually be drawn to you if you’re doing excellent work. That was GiveWell’s strategy, and it worked.
“Kelly did a lot of work in order to point out a real phenomenon, and I don’t see anyone taking her very seriously.”—Kelly put in a lot of work, but there were a lot of issues with the original post. I think this was inevitable to an extent, unless you’re already a policy expert producing high quality work really needs a group of people or multiple feedback cycles. I think that it is especially important to maintain high standards of evidence in regards to this issue because the increasing political polarisation means that both sides of the spectrum are dropping their own standards.
I don’t see many people who want to figure out how much of a problem there is, and then apply e.g. utilitarianism to decide what to do about that. That would count as acting seriously.