Thanks for writing this! It definitely clarified for me where you’re coming from on this.
I’m now reasonably confident that I disagree with you (about whether this proposal will be helpful, and in what cases, and how/why).
The key thing here, I think, is that there is (it seems to me) a fairly fundamental disconnect between us about what public conversations are, and what they’re for.
As I read your comment, and thought about your original post and your other comments on this, I noticed what I think is an important clue: you repeatedly mention the possibility of other people, third parties, injecting themselves into a tricky or already-heated thread. For you, it seems, third parties / bystanders are important to how a thread develops due to their (potential) participation.
But I think that’s missing the point. The presence of third parties who might participate in a thread is nearly immaterial. What matters is that third parties are watching the thread. Insofar as they might participate, the sort of participation that matters is not contributions to the argument proper (whether they be constructive or otherwise), but comments that demonstrate and constitute “the presence of the audience”—expressions of support for one side or the other, condemnation or praise, opinions about how the discussion is going (and whether it ought to continue), calls for moderation, nitpicking, kibitzing, and other miscellaneous commentary.
You seem to be (correct me if I’m mistaken here) treating discussions on a forum such as Less Wrong as, more or less, conversations between people interested in a topic… that just so happen to be taking place in a publicly accessible place. This latter fact means that other people can interject themselves into a conversation uninvited, and drive it off track… which would be unfortunate, clearly, just as it would be unpleasant for someone to walk up and interrupt a conversation you were having with someone at a party.
But public discussions and debates are inherently performative. They can be less so or more so, but never not so.
In your model, a conversation is between two people (or possibly three, four, etc.—however many are productively participating in a thread). Thus a comment thread is either collaborative—with each participant basically wanting the same thing as the other one does (i.e., to figure out the truth and/or convince the other of a truth they already know) or there is, in an important sense, no point. Consequently, you treat such things as “tribal overtones”, or the sense that you have to defend yourself, or utter a clever comeback, as edge cases—indicators of something going astray.
But it seems to me that much, perhaps almost all, of the time, what is said, is said not (or at the very least, not just) for the benefit of one’s interlocutor, but for the audience. Viewed through this lens, much of what you write in your comment seems fundamentally mis-aimed.
(I do not think this is a bad thing, by the way; quite the opposite. This comment is already long, so I won’t go into why I think that—and perhaps it doesn’t need any explanation.)
The question, I guess, is: do you disagree with my characterization? Or do you think that I am correct descriptively but not normatively (and that this is true in general but should be discouraged on Less Wrong so strongly that we need not even take it into account when deciding how to deal with the discussions we do have)? Or something else?
This all sounds approximately right (point #6 is where I touch upon this aspect of the model. I didn’t dwell on it since there were a lot of other things to dwell on)
My claim is something like “yes, performative conversation is often the default, but it performative conversation makes it harder to find and agree on truth. So if that is your goal, taking it private it will help. If that’s not your goal, taking it private may not help.”
(meta side: I notice you splitting comments due to length, and not sure if that’s due to aesthetic or because of something about the site preventing long comments. I’ve been able to type long comments without issue so wasn’t sure)
I think it’s been mentioned a couple times that the site does not have any limits on comment length. If you’re having trouble posting long comments, can you elaborate on what happens when you try?
From what I understand Said is currently posting comments through greaterwrong.com, which is a site that uses our API to provide their own view on the LesserWrong.com content. Our API currently has a character limit for posting markdown directly into comments (which is an accidental result from some of the frameworks we are using), and I think greaterwrong.com is running into that problem.
Hmm. I feel like I might not have quite gotten my point across (which is possibly because it was nearly 5 AM when I posted that comment). I can’t yet tell if we disagree or if I simply haven’t made clear what I’m saying, so let me try to expand a bit on this.
You say:
yes, performative conversation is often the default, but it performative conversation makes it harder to find and agree on truth.
This seems to suggest a model where two people are engaging in collaborative truth-seeking, but—because they’re doing this in public—performativeness is a quality that their conversation ends up having, which interferes with their goal.
I, on the other hand, am suggesting a model where the performative aspect is inseparable from the goal, where it, in a serious sense, is [a large part of] the goal.
Now, maybe it’s just that we differ in our estimation of how prevalent this is (or how prevalent it is here). But… it seems to me to be a fairly safe supposition that even if conversation-as-performance[1] is less common on Less Wrong than elsewhere (relative to “conversation as collaborative truth-seeking”), it is probably almost always what’s going on in what you call “demon thread”.
But this means that taking the conversation private will basically never help.
I’m fairly confident that we’re (roughly) understanding each other, but have some underlying differences on a combination of a) how the world currently is, b) how the LW world is right now, c) what’s desireable and achievable for LW culture.
(Actually I think we probably agree on how the world in general is).
I think that’s beyond the scope of the conversation I want to have on this post though.
Fair enough, and I tentatively agree with your evaluation.
I do think that this broader conversation is important to have at some point (though, indeed, this post is not the place for it)—because whether this (or, indeed, any other) scheme succeeds, depends on its outcome.
Agreed. For now, I just want to be clear that I think the tactic outlined in this post only makes sense if you’re using the overall strategy listed in this parent comment, and I think whether that strategy makes sense depends on whatever your current situation is.
Thanks for writing this! It definitely clarified for me where you’re coming from on this.
I’m now reasonably confident that I disagree with you (about whether this proposal will be helpful, and in what cases, and how/why).
The key thing here, I think, is that there is (it seems to me) a fairly fundamental disconnect between us about what public conversations are, and what they’re for.
As I read your comment, and thought about your original post and your other comments on this, I noticed what I think is an important clue: you repeatedly mention the possibility of other people, third parties, injecting themselves into a tricky or already-heated thread. For you, it seems, third parties / bystanders are important to how a thread develops due to their (potential) participation.
But I think that’s missing the point. The presence of third parties who might participate in a thread is nearly immaterial. What matters is that third parties are watching the thread. Insofar as they might participate, the sort of participation that matters is not contributions to the argument proper (whether they be constructive or otherwise), but comments that demonstrate and constitute “the presence of the audience”—expressions of support for one side or the other, condemnation or praise, opinions about how the discussion is going (and whether it ought to continue), calls for moderation, nitpicking, kibitzing, and other miscellaneous commentary.
You seem to be (correct me if I’m mistaken here) treating discussions on a forum such as Less Wrong as, more or less, conversations between people interested in a topic… that just so happen to be taking place in a publicly accessible place. This latter fact means that other people can interject themselves into a conversation uninvited, and drive it off track… which would be unfortunate, clearly, just as it would be unpleasant for someone to walk up and interrupt a conversation you were having with someone at a party.
(comment split)
(comment continues)
But public discussions and debates are inherently performative. They can be less so or more so, but never not so.
In your model, a conversation is between two people (or possibly three, four, etc.—however many are productively participating in a thread). Thus a comment thread is either collaborative—with each participant basically wanting the same thing as the other one does (i.e., to figure out the truth and/or convince the other of a truth they already know) or there is, in an important sense, no point. Consequently, you treat such things as “tribal overtones”, or the sense that you have to defend yourself, or utter a clever comeback, as edge cases—indicators of something going astray.
But it seems to me that much, perhaps almost all, of the time, what is said, is said not (or at the very least, not just) for the benefit of one’s interlocutor, but for the audience. Viewed through this lens, much of what you write in your comment seems fundamentally mis-aimed.
(I do not think this is a bad thing, by the way; quite the opposite. This comment is already long, so I won’t go into why I think that—and perhaps it doesn’t need any explanation.)
The question, I guess, is: do you disagree with my characterization? Or do you think that I am correct descriptively but not normatively (and that this is true in general but should be discouraged on Less Wrong so strongly that we need not even take it into account when deciding how to deal with the discussions we do have)? Or something else?
This all sounds approximately right (point #6 is where I touch upon this aspect of the model. I didn’t dwell on it since there were a lot of other things to dwell on)
My claim is something like “yes, performative conversation is often the default, but it performative conversation makes it harder to find and agree on truth. So if that is your goal, taking it private it will help. If that’s not your goal, taking it private may not help.”
(meta side: I notice you splitting comments due to length, and not sure if that’s due to aesthetic or because of something about the site preventing long comments. I’ve been able to type long comments without issue so wasn’t sure)
It’s the latter (and I am told it’s being worked on, which is why I haven’t posted to complain about it).
I think it’s been mentioned a couple times that the site does not have any limits on comment length. If you’re having trouble posting long comments, can you elaborate on what happens when you try?
From what I understand Said is currently posting comments through greaterwrong.com, which is a site that uses our API to provide their own view on the LesserWrong.com content. Our API currently has a character limit for posting markdown directly into comments (which is an accidental result from some of the frameworks we are using), and I think greaterwrong.com is running into that problem.
Hmm. I feel like I might not have quite gotten my point across (which is possibly because it was nearly 5 AM when I posted that comment). I can’t yet tell if we disagree or if I simply haven’t made clear what I’m saying, so let me try to expand a bit on this.
You say:
This seems to suggest a model where two people are engaging in collaborative truth-seeking, but—because they’re doing this in public—performativeness is a quality that their conversation ends up having, which interferes with their goal.
I, on the other hand, am suggesting a model where the performative aspect is inseparable from the goal, where it, in a serious sense, is [a large part of] the goal.
Now, maybe it’s just that we differ in our estimation of how prevalent this is (or how prevalent it is here). But… it seems to me to be a fairly safe supposition that even if conversation-as-performance[1] is less common on Less Wrong than elsewhere (relative to “conversation as collaborative truth-seeking”), it is probably almost always what’s going on in what you call “demon thread”.
But this means that taking the conversation private will basically never help.
I’m fairly confident that we’re (roughly) understanding each other, but have some underlying differences on a combination of a) how the world currently is, b) how the LW world is right now, c) what’s desireable and achievable for LW culture.
(Actually I think we probably agree on how the world in general is).
I think that’s beyond the scope of the conversation I want to have on this post though.
Fair enough, and I tentatively agree with your evaluation.
I do think that this broader conversation is important to have at some point (though, indeed, this post is not the place for it)—because whether this (or, indeed, any other) scheme succeeds, depends on its outcome.
Agreed. For now, I just want to be clear that I think the tactic outlined in this post only makes sense if you’re using the overall strategy listed in this parent comment, and I think whether that strategy makes sense depends on whatever your current situation is.