But I was not trying to make an argument from authority. Apparently I should be clearer in the future since that’s how it came across. I was trying to make a purely logic-based (as in, not evidence-based or authority-based, etc.) argument and cited the SSC rule as a shortcut to what I was trying to explain. Rather than write a long post explaining exactly why I felt the post was unjustified given its expected impact, I thought that citing something most people here are familiar with would be an easier/faster explanation.
Indeed. This clarifies things, thanks. I agree that it would be helpful to be more clear about this in the future—though I question the wisdom of making reference to such rules at all, due to what I said about “privileging the hypothesis”.
I meant my original point to read something like: I expect this post to be harmful to the discourse here for reasons most easily summarized by saying that it fails the SSC 2-out-of-3 rule.
But you see, this is precisely the problem. What are the reasons (for why a post might be harmful to the discourse) that are summarized by saying that something fails the SSC rule? It’s not clear that we know—certainly I don’t!
Or, put it this way: since you assumed that your interlocutor (and, possibly, other readers of Less Wrong) would know what those reasons are (since you expected that your shorthand would be decoded properly), it must therefore be true (in your view) that these reasons have been discussed and elucidated, prominently and publicly. Could you provide some links to such prominent and public discussions?
… it would be like bringing up a fake framework …
It will probably not surprise you to learn that I take rather a dim view of “fake frameworks”.
it must therefore be true (in your view) that these reasons have been discussed and elucidated, prominently and publicly
If I wanted to ensure that every interlocutor understood (or could easily understand given a small amount of time) my point, then this would have to be true. You can’t decode the message if you don’t have the key, so if you want everyone to decode the message then the key must be as public and prominent as possible.
But I didn’t want to ensure that because doing so would take a lot more effort than I spent on that first comment. I am totally OK with only some people decoding the message if it reduces what would have been a significant writing effort into a quick comment. Please correct me if I’m mistaken, but you seem to think that sending messages like this—messages that are easier to decode for some than others—harms the discourse. But couldn’t that argument be made for any message that requires investment to understand?
For instance, LW contains many posts that assume a solid understanding of linear algebra, something that very few people (out of, say, all people who can read English and access the internet) have. To those unfamiliar with linear algebra, most of those LA-heavy posts are unintelligible. Should we avoid posting LA-heavy posts?
It’s actually a little funny because thinking through this has made me realize that a math-heavy post creates a polarized response (those who understand the math enough to get something out of the post vs those who don’t) in the same way that a political post does. And by your argument, referencing a framework that not everyone understands / agrees with / believes is useful also polarizes the response in a similar way.
Given that I see no problem with referencing math or disputed frameworks as a means of communicating with some people at the expense of losing others, this makes me much less confident that posts expected to produce politically polarized responses are problematic. If people unfamiliar with LA can skip LA-heavy posts, people with partisan biases/feelings can skip posts that touch on those.
Please correct me if I’m mistaken, but you seem to think that sending messages like this—messages that are easier to decode for some than others—harms the discourse.
Well… no, I don’t think I’d say that. It’s just that, if what you say is “hard to decode” in a certain way that will result in some people misunderstanding you as saying something which you are not actually saying, then, if you say something that—interpreted as intended—does not “harm the discourse”, nevertheless some people may misinterpret you as saying something different—something which does “harm the discourse”.
Which seems to be what happened here. (I think? There seem to have been several layers of misinterpretation, and I confess to having somewhat lost track. Which is, of course, itself a problem…)
I… think… I am not sure, but I think that, with all the corrections and clearings-up-of-misinterpretations that we’ve managed to muddle through, I no longer have any problem with your intended meaning. (My confidence in this is not high at all, but I’d judge it as more likely than the opposite.)
For instance, LW contains many posts that assume a solid understanding of linear algebra, something that very few people (out of, say, all people who can read English and access the internet) have. To those unfamiliar with linear algebra, most of those LA-heavy posts are unintelligible. Should we avoid posting LA-heavy posts?
This analogy is fundamentally flawed, and here is why.
As I’ve said before, the problem is not opacity, or the possibility of misunderstanding, per se; the problem is the double illusion of transparency. Consider that if I read a post filled with talk of linear algebra, then I know perfectly well that I don’t understand it. I don’t know much of linear algebra, so if you start saying things about vector spaces, then it’s clear to me that I don’t know what you’re talking about. And so it’s clear enough that a post like this one is simply not aimed at me.
Whereas when I read a comment that is not math-heavy, and is phrased in what seems to be perfectly ordinary plain English, and makes reference to ideas and words and phrases with which I am familiar, and does not seem confusing, then… how am I to know that the comment is actually not aimed at me? Why would I assume that it isn’t?
What’s more, consider the consequence of a norm that approved of communicating in this way. It would no longer be required of participants in Less Wrong, that they make their writing comprehensible. Why would they? Should anyone question them, they simply respond that their comment was aimed only and exclusively at those who already understand them. You would turn us all into Zizek, or perhaps Nasruddin! (Admirable figures, the both of them, each in his own way… but hardly role models for the Less Wrong commenter!)
Whereas when I read a comment that is not math-heavy, and is phrased in what seems to be perfectly ordinary plain English, and makes reference to ideas and words and phrases with which I am familiar, and does not seem confusing, then… how am I to know that the comment is actually not aimed at me? Why would I assume that it isn’t?
Is that what happened here, though? I posted a comment referencing the SSC rule and you objected to its use in that context. We both knew what I was referring to. The confusion seems to have arisen because I was intending the reference as a shortcut through my reasoning and you interpreted it as me smuggling a foreign norm into the discourse as if it were already widely accepted.
If I had been clearer about how I was using the reference, would there by any illusion of transparency, much less a double illusion? I didn’t expect anyone unfamiliar or not on board with the reference to understand my logic and you didn’t think you understood my point while actually misunderstanding it—in fact, you very clearly expressed that you didn’t understand my point. So both of us were aware of the lack of transparency from the get-go, I think.
It would no longer be required of participants in Less Wrong, that they make their writing comprehensible.
I mean this in the least sarcastic way possible: to 99% of people I talk to, LW writing is incomprehensible. I have tried many times to introduce LW-related concepts to people unfamiliar with LW and, in my experience, it’s insanely difficult to export anything from here to the rest of the world. Obviously my success also depends on how well I explain things, but the only subjects I have similar difficulty explaining to people are very technical things from my own field.
To be clear, I’m not saying “well everything’s already too hard to explain so let’s go full Zizek!” It is always better to be more comprehensible, all else equal. But all else is not equal—unwrapping explanations to the extent that they are understandable to someone with no familiarity with the subject comes at a great cost. It’s good to have develop jargon and shorthand to expedite communication between people in the know, and when the jargon is explicitly jargon (e.g., “the 2-out-of-3 rule from SSC”), I don’t think there is any illusion of transparency.
Sorry for the wall of text—I’m trying to keep these responses as short as possible but I also want to be clear. One more thing:
Should anyone question them, they simply respond that their comment was aimed only and exclusively at those who already understand them
If I’m trying to explain X but end up only explaining it to people who understand X, then yes, this is pointless and silly. But if I’m trying to explain Y and end up only explaining it to people who understand X, that is useful, especially when many people understand X.
[Brief mod note: I don’t think there’s anything wrong with the thread so far, but it’s approaching a point where I’m somewhat worried, mostly for high-context reasons relating to past threads involving site norms. I don’t have a specific ask at this point but am trying to err in the direction of publicly flagging when I feel worried earlier than I have in the past]
Indeed. This clarifies things, thanks. I agree that it would be helpful to be more clear about this in the future—though I question the wisdom of making reference to such rules at all, due to what I said about “privileging the hypothesis”.
But you see, this is precisely the problem. What are the reasons (for why a post might be harmful to the discourse) that are summarized by saying that something fails the SSC rule? It’s not clear that we know—certainly I don’t!
Or, put it this way: since you assumed that your interlocutor (and, possibly, other readers of Less Wrong) would know what those reasons are (since you expected that your shorthand would be decoded properly), it must therefore be true (in your view) that these reasons have been discussed and elucidated, prominently and publicly. Could you provide some links to such prominent and public discussions?
It will probably not surprise you to learn that I take rather a dim view of “fake frameworks”.
If I wanted to ensure that every interlocutor understood (or could easily understand given a small amount of time) my point, then this would have to be true. You can’t decode the message if you don’t have the key, so if you want everyone to decode the message then the key must be as public and prominent as possible.
But I didn’t want to ensure that because doing so would take a lot more effort than I spent on that first comment. I am totally OK with only some people decoding the message if it reduces what would have been a significant writing effort into a quick comment. Please correct me if I’m mistaken, but you seem to think that sending messages like this—messages that are easier to decode for some than others—harms the discourse. But couldn’t that argument be made for any message that requires investment to understand?
For instance, LW contains many posts that assume a solid understanding of linear algebra, something that very few people (out of, say, all people who can read English and access the internet) have. To those unfamiliar with linear algebra, most of those LA-heavy posts are unintelligible. Should we avoid posting LA-heavy posts?
It’s actually a little funny because thinking through this has made me realize that a math-heavy post creates a polarized response (those who understand the math enough to get something out of the post vs those who don’t) in the same way that a political post does. And by your argument, referencing a framework that not everyone understands / agrees with / believes is useful also polarizes the response in a similar way.
Given that I see no problem with referencing math or disputed frameworks as a means of communicating with some people at the expense of losing others, this makes me much less confident that posts expected to produce politically polarized responses are problematic. If people unfamiliar with LA can skip LA-heavy posts, people with partisan biases/feelings can skip posts that touch on those.
Well… no, I don’t think I’d say that. It’s just that, if what you say is “hard to decode” in a certain way that will result in some people misunderstanding you as saying something which you are not actually saying, then, if you say something that—interpreted as intended—does not “harm the discourse”, nevertheless some people may misinterpret you as saying something different—something which does “harm the discourse”.
Which seems to be what happened here. (I think? There seem to have been several layers of misinterpretation, and I confess to having somewhat lost track. Which is, of course, itself a problem…)
I… think… I am not sure, but I think that, with all the corrections and clearings-up-of-misinterpretations that we’ve managed to muddle through, I no longer have any problem with your intended meaning. (My confidence in this is not high at all, but I’d judge it as more likely than the opposite.)
This analogy is fundamentally flawed, and here is why.
As I’ve said before, the problem is not opacity, or the possibility of misunderstanding, per se; the problem is the double illusion of transparency. Consider that if I read a post filled with talk of linear algebra, then I know perfectly well that I don’t understand it. I don’t know much of linear algebra, so if you start saying things about vector spaces, then it’s clear to me that I don’t know what you’re talking about. And so it’s clear enough that a post like this one is simply not aimed at me.
Whereas when I read a comment that is not math-heavy, and is phrased in what seems to be perfectly ordinary plain English, and makes reference to ideas and words and phrases with which I am familiar, and does not seem confusing, then… how am I to know that the comment is actually not aimed at me? Why would I assume that it isn’t?
What’s more, consider the consequence of a norm that approved of communicating in this way. It would no longer be required of participants in Less Wrong, that they make their writing comprehensible. Why would they? Should anyone question them, they simply respond that their comment was aimed only and exclusively at those who already understand them. You would turn us all into Zizek, or perhaps Nasruddin! (Admirable figures, the both of them, each in his own way… but hardly role models for the Less Wrong commenter!)
Is that what happened here, though? I posted a comment referencing the SSC rule and you objected to its use in that context. We both knew what I was referring to. The confusion seems to have arisen because I was intending the reference as a shortcut through my reasoning and you interpreted it as me smuggling a foreign norm into the discourse as if it were already widely accepted.
If I had been clearer about how I was using the reference, would there by any illusion of transparency, much less a double illusion? I didn’t expect anyone unfamiliar or not on board with the reference to understand my logic and you didn’t think you understood my point while actually misunderstanding it—in fact, you very clearly expressed that you didn’t understand my point. So both of us were aware of the lack of transparency from the get-go, I think.
I mean this in the least sarcastic way possible: to 99% of people I talk to, LW writing is incomprehensible. I have tried many times to introduce LW-related concepts to people unfamiliar with LW and, in my experience, it’s insanely difficult to export anything from here to the rest of the world. Obviously my success also depends on how well I explain things, but the only subjects I have similar difficulty explaining to people are very technical things from my own field.
To be clear, I’m not saying “well everything’s already too hard to explain so let’s go full Zizek!” It is always better to be more comprehensible, all else equal. But all else is not equal—unwrapping explanations to the extent that they are understandable to someone with no familiarity with the subject comes at a great cost. It’s good to have develop jargon and shorthand to expedite communication between people in the know, and when the jargon is explicitly jargon (e.g., “the 2-out-of-3 rule from SSC”), I don’t think there is any illusion of transparency.
Sorry for the wall of text—I’m trying to keep these responses as short as possible but I also want to be clear. One more thing:
If I’m trying to explain X but end up only explaining it to people who understand X, then yes, this is pointless and silly. But if I’m trying to explain Y and end up only explaining it to people who understand X, that is useful, especially when many people understand X.
[Brief mod note: I don’t think there’s anything wrong with the thread so far, but it’s approaching a point where I’m somewhat worried, mostly for high-context reasons relating to past threads involving site norms. I don’t have a specific ask at this point but am trying to err in the direction of publicly flagging when I feel worried earlier than I have in the past]