My posts vary a lot in how many qualifiers I have. So much so that I get a lot of feedback in both directions: for some posts people fume at how “overconfident” I sound, for others people complain about how all the qualifiers make it very hard to read.
I’m sure a lot of this is just a skill issue on my part but I think a lot of what ppl see as tonal mismatches just comes from me being someone who legitimately ( “genuinely”) believes in the power of reason and evidence. It’s very hard to demonstrate this within a single post.
People (especially people who object-level disagree with me) often get mad at me for how unreasonably confident I might sound on surprising claims. But imo it’d be more performatively dishonest for me to add qualifiers when I don’t believe in them!
For those posts, I add a lot of caveats, to the point that people DM me complaints for how annoying it is to read. Again, definitely partially a skill issue on my part! But I don’t how else to gesture at actually being uncertain, especially given practical time constraints!
I think other writers (and AIs) are drawn to a kind of milquetoast “charity” or mid degree of across-the-board performative humility. Or like the performed humility is correlated more with whether a claim “sounds surprising” than what the balance of reason and evidence should lead you to believe, or what the author actually believes. Or how humble you sound comes across as instantiation of your personality or “brand” (real or affected) as a public intellectual.
I find this variously imperceptive (in the sense that the writer does not have enough trust, or justified trust, in their own assessment of the truth), or dishonest.
Hopefully it’s obvious, if your writing is widely read and says important things, there will always be people calling you overconfident. You cannot add enough qualifiers such that all people stop reading disagreements as your reasoning being overconfident.
feels like a case where a crux is “why are you writing?”, and based on the answer to that, “what is the impact to your objective, if any, of the specific reactions that specific people have to the relevant writings?” such that you can usefully assess which, if any, praxis opportunities the reaction-data reveals for you.
This is one thing that I disagree (implicitly perhaps) with Nate and other writers about. I do use epistemic statuses (including in an example above, which I linked just now). But I just don’t think epistemic statuses are sufficient for a lot of readers.
Signaling confidence all the way up when you’re object level uncertain comes across as a communicative error to me, in the sense that it predictably misleads most readers who aren’t carefully paying attention, and likely many readers who are.
People (especially people who object-level disagree with me) often get mad at me for how unreasonably confident I might sound on surprising claims. But imo it’d be more performatively dishonest for me to add qualifiers when I don’t believe in them!
It seems generally very doable to me to not overly hedge when you’re confident and also state your true beliefs at the same time. For example, instead of saying “My best guess is that the encylical written by AI, but I am not sure” (which is hedging too much), its possible to say something like “This will probably sound surprising, but I believe I have amassed significant evidence that the encylical was written by LLM. Judge for yourself”[1]. I also think this is not too time-consuming once you notice that something might sound overly confident.
In general, I think there’s a false tradeoff between honesty and persuasiveness and that its generally possible to be both if you phrase things well. But I do agree with your other point about hedging too much: it’s often hard to be honest about your unconfident views but also be easily readable.
For this example in particular you may have done the second phrasing, I don’t remember the post. I think the point stands regardless. I use this example only because you mention it.
My posts vary a lot in how many qualifiers I have. So much so that I get a lot of feedback in both directions: for some posts people fume at how “overconfident” I sound, for others people complain about how all the qualifiers make it very hard to read.
I’m sure a lot of this is just a skill issue on my part but I think a lot of what ppl see as tonal mismatches just comes from me being someone who legitimately ( “genuinely”) believes in the power of reason and evidence. It’s very hard to demonstrate this within a single post.
I just have a lot of confidence based on careful reasoning and evidence that some very ex ante surprising claims are true: the recent papal encyclical is with very high probability AI-assisted, or that you can get better at board games via paying attention to a single heuristic.
People (especially people who object-level disagree with me) often get mad at me for how unreasonably confident I might sound on surprising claims. But imo it’d be more performatively dishonest for me to add qualifiers when I don’t believe in them!
In contrast, I sometimes make arguments that I think are interesting and important but have a high chance of being wrong: anthropic explanation for the unreasonable effectiveness of mathematics, or anti-ageing hopes being obviated by basic evolutionary theory.
For those posts, I add a lot of caveats, to the point that people DM me complaints for how annoying it is to read. Again, definitely partially a skill issue on my part! But I don’t how else to gesture at actually being uncertain, especially given practical time constraints!
I think other writers (and AIs) are drawn to a kind of milquetoast “charity” or mid degree of across-the-board performative humility. Or like the performed humility is correlated more with whether a claim “sounds surprising” than what the balance of reason and evidence should lead you to believe, or what the author actually believes. Or how humble you sound comes across as instantiation of your personality or “brand” (real or affected) as a public intellectual.
I find this variously imperceptive (in the sense that the writer does not have enough trust, or justified trust, in their own assessment of the truth), or dishonest.
Hopefully it’s obvious, if your writing is widely read and says important things, there will always be people calling you overconfident. You cannot add enough qualifiers such that all people stop reading disagreements as your reasoning being overconfident.
True, though framing things in just the right way is often nonetheless a skill issue, can mean the difference between getting 1000 haters and 50.
feels like a case where a crux is “why are you writing?”, and based on the answer to that, “what is the impact to your objective, if any, of the specific reactions that specific people have to the relevant writings?” such that you can usefully assess which, if any, praxis opportunities the reaction-data reveals for you.
If it’s about the whole post, epistemic status declaration are can be helpful in that regard.
This is one thing that I disagree (implicitly perhaps) with Nate and other writers about. I do use epistemic statuses (including in an example above, which I linked just now). But I just don’t think epistemic statuses are sufficient for a lot of readers.
Signaling confidence all the way up when you’re object level uncertain comes across as a communicative error to me, in the sense that it predictably misleads most readers who aren’t carefully paying attention, and likely many readers who are.
That’s a known phenomenon.
fwiw I think this is a pretty different style than what I have/what I’m describing.
Perhaps Linch should’ve qualified that they weren’t sure if this had been explored before.
This was interesting, thanks!
It seems generally very doable to me to not overly hedge when you’re confident and also state your true beliefs at the same time. For example, instead of saying “My best guess is that the encylical written by AI, but I am not sure” (which is hedging too much), its possible to say something like “This will probably sound surprising, but I believe I have amassed significant evidence that the encylical was written by LLM. Judge for yourself”[1]. I also think this is not too time-consuming once you notice that something might sound overly confident.
In general, I think there’s a false tradeoff between honesty and persuasiveness and that its generally possible to be both if you phrase things well. But I do agree with your other point about hedging too much: it’s often hard to be honest about your unconfident views but also be easily readable.
For this example in particular you may have done the second phrasing, I don’t remember the post. I think the point stands regardless. I use this example only because you mention it.
Also true for honesty and kindness