Person A : Person B, my model predicted Y because of evidence X. But your experience sounds like ~Y, so I was surprised and want to update. Tell me more about your ~Y experiences!
In other words, consider that the other person possesses evidence that you do not, and invite them to update you instead of trying to update them.
A non-gender example:
Atheist: Pentecostal, my model predicted that people would go home from church feeling bored, guilty, or self-righteous, because former church people I know talk about those experiences, and church people who are active in politics seem to be big on guilt and self-righteousness. But your experience sounds like church is a fun party, that you go home from feeling giddy and high. I was surprised and want to update. Tell me more about your religious experiences!
In other words, consider that the other person possesses evidence that you do not, and invite them to update you instead of trying to update them.
My communicating my differing perception to the other person in Option 1 is my invitation to have them update me.
Going through the song and dance of your third option is not required with some people, making them more efficient partners at finding the truth. I find people who require constant ego stroking in this manner, or who give it, literally tiresome in an intellectual endeavor.
It seems to me that flat contradiction without any communication of being open to being convinced is a strongly suboptimal invitation to update the speaker. This is especially so in cases of strongly asymmetric information (either direction).
‘Song and dance’ appears to me to be a dysphemism (perhaps unintentional) for ‘communicating what you mean’ as opposed to ‘indicating something in the general vein and hoping the receiver figures out what you meant’.
Edited to add: option A is much more reasonable than I credited it, so while I’ll stand by my first paragraph above, it’s not particularly relevant to the post above. And yes, option 3 could be streamlined.
without any communication of being open to being convinced
For me, you can take that I’m open to being convinced as the null hypothesis. Most civilized people are. Aren’t you?
dysphemism
Thank you! I’ve been looking for that word forever.
‘Song and dance’ appears to me to be a to be a dysphemism (perhaps unintentional) for ‘communicating what you mean’
Not really, because ‘communicating what you mean’ was not what I meant. I was referring to kabuki dance of your ritualized formula for disagreement to stroke a person’s ego so that he doesn’t feel a threat to his status by my disagreeing with him.
I don’t think the fellow is really confused about whether I’m open to being convinced of the error of my ways. If I say “I think you’re wrong because of X”, does not the human impulse to reciprocity sanction and invite him to respond in kind?
Does that fellow really need it explained to him that if I disagree with him on when the bus is coming, that he is free and invited to disagree with me right back? I don’t think so.
He: The bus is coming at 3:00. Me: No, it’s coming at 3:10; that’s when I caught it yesterday. He: But yesterday was Friday. Saturday has a different schedule.
That seems like an everyday, ordinary human conversation to me, that no one should get all excited or offended about.
I strongly suspect that tone and body language are a key component in whether the statement “that’s not right” is interpreted as “I disagree, let’s talk about it” or “shut up and think what I think”.
I further suspect that a tendency to interpret ambiguous or missing subtext in a negative or overly critical way correlates strongly with being “thin-skinned”. This is partly based on having both of these characteristics myself. A potential counter-argument here is that it is not “rational” or useful to always assume the worst in personal interactions if you have evidence to follow instead (Have people generally meant the worst things possible when I have been unsure in the past?), but the important thing to remember here is that we are not dealing with people who have had time to be trained in that way. A martial arts master does not go all out against a beginner knowing that they will one day be able to handle it.
It would be unwise to alienate a group of potential rationalists if there is a relatively simple way to avoid it. If it would cripple the discourse or otherwise be quite detrimental to implement any sort of fix, then I would not advocate that course of action. However, I believe that to not be the case.
At this time, I would like to agree with RichardKennaway’s observation that Plasmon’s option A was quite different from the situation posited by Submitter B, and further agree with his hypothesis that even option A is some sort of improvement (largely due to the word “may”).
My conclusion is that a few changes of word choices would be a low-cost, medium-reward first step in the right direction. This would include using words such as “may”, particularly in the context of someone’s perceived domain of expertise or cherished belief. Also, explicitly starting an evidence based conversation while voicing your disagreement.
Example: I disagree with your statement that “Most civilized people are [open to being convinced]”. As (anecdotal) evidence, I submit the large number of Americans who are closemindedly religious.
For me, you can take that I’m open to being convinced as the null hypothesis. Most civilized people are.
If one considers sufficiently impersonal topics like bus schedules? Yes, for the most part.
Microcultures with strong elements of authority will have a much harder time with this assumption, even in horizontal interactions. I would not call all of these uncivilized, though I’m not a fan of them.
It’s not complicated to frame a conversation as a search for truth as opposed to a vs. argument. Many people go overboard in this. I agree that this is obnoxious. I maintain that a flat contradiction is in many cases insufficient, especially in those cases where the matter at hand is contentious or personal, or there is any degree of hostility or unease between the conversants.
Option A wasn’t a flat contradiction only. In fact, the original person wrote it up in a more pussy footing way than I would.
Flat contradiction would be “you’re wrong”. I agree that’s not an invitation to further discussion.
My usual comment would be of the form:
“That’s wrong. Blah di blah isn’t blah di blee, it’s hooty hooty.”
It’s “you’re wrong” plus some evidence on which I based my disagreement. Would that be unclear to you personally, that you’re welcome to disagree and cite evidence for your disagreement in turn?
Maybe we could try an example so that we’re talking about something concrete. I just don’t think it’s a mystery. I think that a great many people get very touchy when it comes to being disagreed with. I’m of another species that likes to be disagreed with, because then we have a contradiction to resolve, and that’s fun and potentially productive.
I’m sorry: for reasons I do not understand, I misunderstood what you were referring to with ‘option A’. Your response made perfect sense and mine did not.
That didn’t seem like an accurate characterization of option A to me, so I gave a concrete example:
Flat contradiction would be “you’re wrong”.
and a concrete example of the option A alternative:
“That’s wrong. Blah di blah isn’t blah di blee, it’s hooty hooty.”
It would have been better to be more concrete.
Was that the issue?
I feel that in these more personal discussions abstract terms gets used, and each side is picturing a very different part of the spectrum for their concretes.
I think it was that I kind of short-circuited ‘option 1’ into meaning ‘the first option mentioned’, and from there ‘what the guy said in the first place’. This is not what you were referring to by ‘option 1’, and even though it’s an understandable error, I still should have been able to pick up on it from the context of the parent and grandparent comment to yours.
My communicating my differing perception to the other person in Option 1 is my invitation to have them update me.
Well, except that you would not be actually stating an invitation or request for more information. You would be assuming that the other person will interpret contradiction as an invitation for further discussion rather than as a dismissal, insult, threat, or other sort of speech act.
(Humans use language for a lot of other purposes besides the merely indicative, after all.)
If you say, “I’m having a party on Saturday,” some people in some situations will take this to mean that you are thereby inviting them to come to the party. Others will think that you are merely stating a fact about your own social life. Still others will think that you are excluding them, just as if you had added, ”… and you’re not invited, you disgusting worm!”
Some people hear an invitation. Some hear a statement of fact. Some hear an exclusionary insult.
If you want to make it clear that you are inviting them, you say, “I’m having a party on Saturday, would you like to come?” or ”… and you’re invited!”
This is not bullshit song-and-dance ego-stroking. It is clear communication, and in particular a way to address people’s differing priors about what your communication could mean. It probably depends on recognizing that people have different priors, and that they arrived at those priors legitimately.
(For that matter, if expressing curiosity about other people’s experiences is an effective way to get data from them, then rationalists should practice doing it a lot until it is automatic and cheap System 1 behavior!)
You would be assuming that the other person will interpret contradiction as an invitation for further discussion rather than as a dismissal, insult, threat, or other sort of speech act.
Yes. In this context, and most contexts, that’s my null hypothesis. Isn’t it yours? People are here to discuss, and not dismiss, insult, or threaten.
Do you think I’m here to dismiss, insult, or threaten people? Do you think a large percentage of people here are? Do you think that anyone who says “you’re wrong” is? That strikes me as a bizarre and thoroughly inaccurate prior. Or I certainly believe and hope it is.
Am I wrong? Is it just foolish innocence on my part to think that people are here to discuss, and not stomp on other people to social climb or satisfy sadistic impulses? It wouldn’t be the first time. In other contexts, yeah, there’s a lot of that going on. And it admittedly took me a long time to figure that out. But I don’t see it here. The trouble is, if it were, most of the people who know aren’t going to tell you.
I was referring more to the comment thread, which is filled with detailed writing in support of sending blunt communication while ignoring that such behaviour ends up losing in practice. If you haven’t actually read that article and its comment thread, you really should.
Losing, in what game? Are you sure EY knows the game everyone is playing? I think he is making implicit assumptions about motivations that are incorrect.
I disagree with his strategic analysis. In some contexts I would consider it correct. Yes, I knuckle under and be what “normal people” want me to be, to avoid the costs of being myself, just as all those normal people are busy being what they think other people want them to be.
But where I can, I seek to escape that mutual cage. Internet forums are a place where escape is possible, because the normals no longer have an overwhelming majority, or might not even have a majority at all, and the cost of anyone’s disapproval online is less.
Dale Carnegie teaches you to be the person other people want you to be; I’d rather find the people who like who I want to be, and want to be who I like.
An anecdote from my dissertation adviser. He was having much the same discussion with me, telling me how professors in Asia were allowed less direct intellectual confrontation. Perhaps EY would be proud.
But the discussion went on to the joy of moving to the US, exemplified by another professor he knew, who responded to someone else in a discussion by gleefully retorting “I Disagree! I Disagree! I Disagree!” Free at last, free at last, thank God almighty, free at last! Free to be honest, free to be open, free to be who you are.
I want to sit at the table where they’re dealing that game. It seems like there are enough people of my ilk at this party for us to have a few tables. If the cool kids don’t want to sit at the nerd tables, that’s fine, and hardly anything new.
Yes. In this context, and most contexts, that’s my null hypothesis. Isn’t it yours?
Not really, no. People use language for a hell of a lot of other things besides making statements of fact at each other. I expect that in any given speech act, a speaker may be doing a lot of things: stating facts, affirming or challenging a social relation with the listener, causing the listener to have expectations about the speaker’s future actions (promises, threats, plans, etc.), and so on. And that a lot of these things may be going on unconsciously.
If someone tells you that the way you speak gives the impression that you are arrogantly dismissing them, you could respond by merely instructing them (in the very same tone that they were talking about) that you do not intend to arrogantly dismiss them. However, doing that is not likely to be very convincing!
In your prior for Less Wrong discussions, when someone responds to a statement of yours by saying that you’re wrong, and cites evidence for his claim, what are the probabilities you place on the following potential motivations for his reply—he wants to discuss the point, he is threatening you, he is dismissing you, he is insulting you, other?
Sorry, I should have been more specific — I can tell because you’re asking a question that would only make sense in a different context. My probabilities about whether you intend to be threatening are are not at issue here.
At issue in this thread is that some portion of the audience are not sticking around — and are forming negative conclusions about LW — because the words here come across as hostile, unfriendly, cold, and so on. This is a danger to LW’s goals.
This is a matter of instrumental rationality, not only epistemic rationality. We want to accomplish something with words, not merely possess accurate beliefs in our own solipsistic internal monologues. So we have to ask, are our uses of words accomplishing the goals that we care about?
If you emit sentences that are consistently misinterpreted, and you are informed of this, you have a few options of what to do. You could conclude ① that your audience is listening wrong, and needs to correct their assumptions about you before they will be able to understand you; or ② that you are speaking wrong, and you need to correct your assumptions about your audience before they will be able to understand you.
If you care about getting your meaning across, which of these conclusions is more likely to give you the ability to accomplish that goal? Either one is consistent with the evidence; but which conclusion strengthens you, and which weakens you?
You can’t reach into your audience’s minds and force them to interpret your words differently.
You can’t force them to stick around and listen to you correct their assumptions, either.
You can change the way you speak.
Concluding that you are misinterpreted because your audience is listening wrong, or is coming into the conversation with crazy priors, weakens you. Thinking that way would make you incapable of fixing the situation; less able to accomplish goals by speaking. Concluding that you are misinterpreted because you have misspoken, or failed to understand where your audience is coming from, gives you the power to fix the situation and accomplish goals. This strengthens you.
At issue in this thread is that some portion of the audience are not sticking around — and are forming negative conclusions about LW — because the words here come across as hostile, unfriendly, cold, and so on. This is a danger to LW’s goals.
Well, failing at epistemic rationality because we prioritized PR over truth-seeking is an even bigger danger to LW’s goals.
Maybe I missed it; did you give the same speech about how empowering it is to focus on what you can change about yourself to those who are taking offense at the speech of others? You understand you could have, right?
Concluding that you are misinterpreted because your audience is listening wrong, or is coming into the conversation with crazy priors, weakens you.
Not if it’s true.
If it’s true, knowing the truth strengthens me. Just because I think it’s true, doesn’t mean I can’t choose to adjust my speech to them.
And yes, when someone has a false impression, and wants to have an accurate one, you can often change their minds by offering evidence for them to update on.
Concluding that you are misinterpreted because you have misspoken, or failed to understand where your audience is coming from, gives you the power to fix the situation and accomplish goals. This strengthens you.
And assuming that I’ve mispoken does weaken me. It assumes I can “fix” the situation by speaking differently. Ok, compared to what have I misspoken? Compared to preemptively changing my speech patterns so that those with extremely high priors of hostile intent from me are less likely to take offense? Do I have a better alternative?
As a first cut, I’m better off talking to people who don’t assume hostility from my style of speech, who can talk to me ‘as is’ in a productive manner. Seems to be a number of such people. To the extent that they’re similar to me, they will be annoyed and possibly offended by speech acts which seem aimed at managing their potential hurt feelings over my disagreements with their opinions. But even removing these emotional factors from the equation, my attempts to manage their feelings take time and effort from me, and wastes time and effort for them on issues extraneous to the topic at hand. At best, altering our styles will waste our time, and at worst, annoy the hell out of each other. That’s a cost.
I’m to bear that cost, for what?
To talk to others who find my manner hostile? Should I unilaterally cave to every demand that I change my manner when they say they feel hurt or offended? On a game theoretic basis alone, that seems like a bad idea. I am to be malleable to their preferences. Ok, I willing to look at that.
I have been having discussions on adjusting my speech patterns to avoid impressions of hostility in others. Even came up with an idea that someone thought was a good step forward—say “I disagree” instead of “You’re wrong.”
And them? Are they to be malleable to my preferences? Not that I’ve seen.
Where are the discussions of the “offended or hurt” adjusting their priors to better reflect the reality that “meanies” like me really aren’t here to insult, offend, or demean them? Or even where are the discussions that assume the priors are correct, but look for ways to suck it up and develop a thicker skin to better deal the hateful bastards trying to hurt them?
I don’t see those. What I’ve seen are offended rejections of any suggestion they might work on changing their reactions, by them, and often by those defending them.
I was the person who said going from “You’re wrong” to “I disagree” was an important step. I’m glad it registered.
Becoming less thin-skinned takes time and sometimes a good bit of work. You don’t know where any particular person is in that process.
You might be in a Pareto’s Law situation—it’s not that you need to avoid offending the most fragile people, a small amount of effort might lead to not offending 85% of thin-skinned people.
Yes, but I didn’t want to finger you as the culprit.
Becoming less thin-skinned takes time and sometimes a good bit of work.
Indeed.
My concern is that some seem to consider it a crime to suggest that this would be a desirable thing, or to suggest that people adjust their inaccurate priors for hostility downwards when interpreting the actions of others.
The nicies want the meanies to try harder to be understood.OK, fine. But if the meanies suggest that the nicies try harder to understand, that’s just one more thing for the nicies to get offended about.
I would note to all the nicies—if LW feels hostile to you, you’ve led a very sheltered existence. I’m from the HItchens party of debate, that prefers a sharp point be embellished with a barb. That’s part of the fun, in the same way that a decleating hit in football is part of the fun. And that’s not even hostility, that’s just style. And that’s a common style.
By my estimate, LW has a very “Just the Facts Ma’am” culture. Going beyond the facts and putting any relish into a debate is rarely done, and frowned on when it happens. Maybe there were nastier times in the long long ago that led to this culture. LW does seem relatively unique in the equal mix of LIbertarians and Progressives.
FWIW, one of the things that caught my attention about this community, and encouraged me to stick around, was the emphasis on valuing accuracy and precision (which I value) without the “barbs are part of the fun/putting relish into a debate” style you describe here (which I dislike intensely).
There are lots of “nice is more important than true” spaces on the net, and lots of “being unpleasant to people is part of the fun” spaces; and an astonishing number of spaces that are both. A space that manages to even approximate being neither is rare.
Barbs are largely the part of the classical rhetoric that play on biases in the listener. That’s probably true of niceness as well.
And I agree that the Lesswrong tone seems relatively unique, particularly given the broad and general nature of discussion topics, and the variety in political opinions.
One of the good things which contributes to the tone here is people reliably getting credit for saying they’ve changed their mind for some good reason. I can’t think of any other site where that’s in play.
Way back when, I remember discussing exactly that point on the Extropians list. In many ways a similar group to here. But some very smart guys were arguing that it was a huge loss of face to admit you were wrong, and better to deny or evade (I’m sure they put it more convincingly than that).
When someone is wrong, graciously admitting and accepting it scores major points with me.
Thinking about it, maybe I can make a better argument for denial. There are two issues, being wrong, and whether one admits being wrong. If admitting being wrong is what largely determines whether you are perceived as being wrong, then denying the error maintains status.
For people driven by social truth, which is likely the majority, truth is scored on attitude, power, authority, popularity, solidarity, fealty, etc. The validity of the arguments don’t matter much. For people driven by epistemic truth, the arguments are what matters, so denying the plain truth of them is seen as a personality defect, while admitting it a virtue.
The thing is, it’s not that the deniers are aliens. I am. I and my kind. For us, in an argument, it’s the facts that matter, and letting other considerations intrude on that is intruding the rules of the normals into the game. That’s largely what this whole thread is about.
One side says we’ll be more effective playing the normals game. It’s a game my kind strongly prefers not to play. Having to behave as normals is ineffective for us, and the opportunity to play by our rules is extremely valuable to us.
Josh Waitzkin’s book The Art of Learning describes his various encounters with unsporting conduct and cheating in chess tournaments and competitive tai chi. He wrote that he’d developed the approach to just work so hard at developing his own skill at the game that he was able to ignore the distractions the opponent was trying to pull and proceed to win anyway. He claimed that the opponents would generally become agitated and careless once they noticed that they couldn’t get any sort of upset out of him.
Discussions aren’t games with rules, but you might still get something out of the idea that social gamesmanship is basically just compensating poor skill with cheating, and you need to work hard enough on your epistemic skills that it won’t stop you even when it does get thrown in your way.
Well, I’d say that social gamesmanship isn’t cheating, it’s playing a different game.
Being very good with your epistemic skills has mileage socially too, and importantly, mileage with people with personal properties you’re more likely concerned about. And refraining from the usual types of social gamesmanship earns you points with those people as well.
Yes, but I didn’t want to finger you as the culprit.
I’m actually very fond of being told I was right. (I only figured that out when a friend mentioned that he’s very fond of other people admitting they were wrong.)
It’s true that there’s currently a belief that it’s very bad to tell people they should be less thin-skinned. People generally want a social environment which suits their preferences, and while it’s not likely that anyone will get a total victory, it’s certainly possible to push the balance towards your preferences.
Thin-skinned people are apt to hear a demand that they be thicker-skinned as “You shouldn’t care about the way I keep hurting you.” The more aggressive among them have started shoving back. Interesting times.
IMO when you write, you should be asking yourself: “What’s the worst way someone could interpret this?”, because surely, someone will interpret it that way. And when you read, you should ask yourself: “What’s the nicest way I could interpret this?”, because that’s probably the way they meant it.
when you write, you should be asking yourself: “What’s the worst way someone could interpret this?”
When dealing with people, habitually searching for only the worst that can happen is a very bad habit, in my experience. It’s a habit I’ve been trying to break. Through availability bias, your world becomes a horrible place. Your priors are distorted toward the bad, and you miss opportunities. Too careful, too risk averse, too distrusting.
And when you read, you should ask yourself: “What’s the nicest way I could interpret this?”, because that’s probably the way they meant it.
I think that’s the right policy, even if it’s not true. It will generally be the more productive assumption—particularly for online forums.
Just work out the cases. Search for everything that can happen. Either a person has basic good will towards you, or they don’t.
If they do, the nice interpretation is likely right, and you understand someone with good will toward you. That makes for a good discussion. Further, if the guy meant it in a nasty way, your response as if he were nice might soften his mood, or not. If it softens, things have at least improved. If not, most observers will likely think him a schmuck, and he is just very unlikely to be a good discussion partner anyway.
If they do have good will, but you assume that it is bad, you’re likely limiting the positive outcomes available with them. If they don’t have good will and you assume they don’t, you have maybe avoided some aggravation and saved yourself some time.
Having worked out the general case, you don’t have to do a de novo analysis each time. Commit to the policy, and blithely move on. Sometimes someone won’t like you. Ok, you knew that was going to happen.
This is what I’ve tried to do in general with my own defensiveness with people. Don’t focus on the worst that a person might do. Try to have an accurate prior on intentions (most people are not con men or mass murderers, and they’re not really out to get me—I’m not that important to them.) Pick a decision based on an analysis of of what their intent and attitudes might be, and the differing outcomes based on your actions.
Most of the analysis applies, except real world encounters carry more serious risks. I live in the Seattle are, which is pretty safe and so real world risks are limited, though I realize not everyone lives in such a safe place, so YMMV.
In general, the best strategy is to act assuming approval and good will, because those situations present the best opportunities.
when you write, you should be asking yourself: “What’s the worst way someone could interpret this?”
When dealing with people, habitually searching for only the worst that can happen is a very bad habit, in my experience.
Ah, but that wasn’t what I meant. I just meant to say that you should be careful when writing, because even when 99%+ of people won’t have any problems with what you write, someone is sure to misinterpret it, if it possibly can be. Communication is hard, and written communication even more so.
I’d say more briefly “someone is sure to misinterpret it”, because it is always possible to do so. There’s going to be a level of misinterpretation no mater how you agonize over what you write.
I agree with you that the underlying good will or lack of it is a crucial factor. I’m still trying to figure out what tends to build good will or damage it.
“Offended or hurt” doesn’t enter into it. This isn’t about hazy feelings; it’s about hard practical effects of actions: do we accomplish what we want to accomplish?
Let’s say you and your interlocutor disagreed about your intention in saying that they were wrong (about whatever). Your interlocutor believes that your intention was for them to shut up and go away, but actually that wasn’t what you meant at all; you meant to invite more discussion.
They are wrong about you.
And you want them to have a correct belief about you.
But … how can you cause your interlocutor to possess a correct belief about your intention? You could lecture them about how wrong they are to have misinterpreted you. But that won’t work if they will take your lecturing as meaning “shut up and go away” … and may very well do so.
That’s all I’m saying. You can’t force people to understand you, or to want to understand you. If you really want to get your ideas across (because you care about those ideas — not because you’re trying to find people who will easily like you) then you use the try harder which probably involves restating them in a way that doesn’t repel people.
Or … well, you could say that you never really cared about that kind of person’s understanding, and really you never wanted a discussion with that kind of person.
But in that case … they weren’t wrong about you, were they?
But in that case … they weren’t wrong about you, were they?
There are plenty of people who would be correct in concluding that I would bear them hostility if I knew what they were like.
They would be incorrect to conclude that the priors I assign to that type of person among LW is very high, and incorrect to assume that my asserting that someone is wrong indicates I have concluded the person is that type of person, so that my comment indicates hostile intent.
Perhaps I’ve given you an incorrect impression.
If you really want to get your ideas across
While I have proselytizing tendencies, that’s not my fundamental goal, particularly in a forum disagreement. Given my limited resources of me, my proselytizing attitude is to sing to those with the ears to hear. People who are assuming that I am hostile are not the low hanging fruit in that regard.
But people who assume I am hostile can be perfectly fine partners in a disagreement. In a disagreement, I am primarily hoping to change my own mind, whether in correcting an error, or clarifying hazy positions of my own. They might even be better, in that they won’t cut me slack when I am sloppy. People who dislike you can be perfectly useful in a discussion. The enemy of my enemy (our ignorance) is my friend.
But I find it strange that you think I should find it hopelessly futile to try to change a person’s assumptions about my intent, but a productive use of my time to try to change their minds about some other fact of reality.
Indeed I agree that it is possible, and probably desirable, to phrase the argument less bluntly than I did.
However, it seems to me that submitter B is arguing against making such arguments at all, not arguing to make them in a more polite fashion.
Furthermore, here of all places, “If you (think you) posses evidence that I do not, show it and update me!” should be a background assumption, not something that needs to be put as a disclaimer on any potentially-controversial statement.
If they said they didn’t understand, or even that they didn’t believe me, that would be workable.
Which I read to mean that she is not opposed to them expressing confusion or saying something like “Huh, you always seemed more like a pure thinker to me.” (as opposed to “No way. You’re totally a thinker.”) It seems precisely how the statement is phrased and how the discussion is conducted that is at issue here.
After the discussion, I think I’ve got a more concise option that achieves this end.
Option 4: I disagree, because blah blah blah.
Concise, and makes it about my differing perceptions and evaluations. Better than my original “you’re wrong, because blah blah.” I doubt that this entirely satisfies the nice camp, but I think it’s a baby step in their direction.
How about a third option:
Person A : Person B, my model predicted Y because of evidence X. But your experience sounds like ~Y, so I was surprised and want to update. Tell me more about your ~Y experiences!
In other words, consider that the other person possesses evidence that you do not, and invite them to update you instead of trying to update them.
A non-gender example:
Atheist: Pentecostal, my model predicted that people would go home from church feeling bored, guilty, or self-righteous, because former church people I know talk about those experiences, and church people who are active in politics seem to be big on guilt and self-righteousness. But your experience sounds like church is a fun party, that you go home from feeling giddy and high. I was surprised and want to update. Tell me more about your religious experiences!
My communicating my differing perception to the other person in Option 1 is my invitation to have them update me.
Going through the song and dance of your third option is not required with some people, making them more efficient partners at finding the truth. I find people who require constant ego stroking in this manner, or who give it, literally tiresome in an intellectual endeavor.
It seems to me that flat contradiction without any communication of being open to being convinced is a strongly suboptimal invitation to update the speaker. This is especially so in cases of strongly asymmetric information (either direction).
‘Song and dance’ appears to me to be a dysphemism (perhaps unintentional) for ‘communicating what you mean’ as opposed to ‘indicating something in the general vein and hoping the receiver figures out what you meant’.
Edited to add: option A is much more reasonable than I credited it, so while I’ll stand by my first paragraph above, it’s not particularly relevant to the post above. And yes, option 3 could be streamlined.
It works just fine with a lot of people.
For me, you can take that I’m open to being convinced as the null hypothesis. Most civilized people are. Aren’t you?
Thank you! I’ve been looking for that word forever.
Not really, because ‘communicating what you mean’ was not what I meant. I was referring to kabuki dance of your ritualized formula for disagreement to stroke a person’s ego so that he doesn’t feel a threat to his status by my disagreeing with him.
I don’t think the fellow is really confused about whether I’m open to being convinced of the error of my ways. If I say “I think you’re wrong because of X”, does not the human impulse to reciprocity sanction and invite him to respond in kind?
Does that fellow really need it explained to him that if I disagree with him on when the bus is coming, that he is free and invited to disagree with me right back? I don’t think so.
He: The bus is coming at 3:00.
Me: No, it’s coming at 3:10; that’s when I caught it yesterday.
He: But yesterday was Friday. Saturday has a different schedule.
That seems like an everyday, ordinary human conversation to me, that no one should get all excited or offended about.
I strongly suspect that tone and body language are a key component in whether the statement “that’s not right” is interpreted as “I disagree, let’s talk about it” or “shut up and think what I think”.
I further suspect that a tendency to interpret ambiguous or missing subtext in a negative or overly critical way correlates strongly with being “thin-skinned”. This is partly based on having both of these characteristics myself. A potential counter-argument here is that it is not “rational” or useful to always assume the worst in personal interactions if you have evidence to follow instead (Have people generally meant the worst things possible when I have been unsure in the past?), but the important thing to remember here is that we are not dealing with people who have had time to be trained in that way. A martial arts master does not go all out against a beginner knowing that they will one day be able to handle it.
It would be unwise to alienate a group of potential rationalists if there is a relatively simple way to avoid it. If it would cripple the discourse or otherwise be quite detrimental to implement any sort of fix, then I would not advocate that course of action. However, I believe that to not be the case.
At this time, I would like to agree with RichardKennaway’s observation that Plasmon’s option A was quite different from the situation posited by Submitter B, and further agree with his hypothesis that even option A is some sort of improvement (largely due to the word “may”).
My conclusion is that a few changes of word choices would be a low-cost, medium-reward first step in the right direction. This would include using words such as “may”, particularly in the context of someone’s perceived domain of expertise or cherished belief. Also, explicitly starting an evidence based conversation while voicing your disagreement.
Example: I disagree with your statement that “Most civilized people are [open to being convinced]”. As (anecdotal) evidence, I submit the large number of Americans who are closemindedly religious.
If one considers sufficiently impersonal topics like bus schedules? Yes, for the most part.
Microcultures with strong elements of authority will have a much harder time with this assumption, even in horizontal interactions. I would not call all of these uncivilized, though I’m not a fan of them.
It’s not complicated to frame a conversation as a search for truth as opposed to a vs. argument. Many people go overboard in this. I agree that this is obnoxious. I maintain that a flat contradiction is in many cases insufficient, especially in those cases where the matter at hand is contentious or personal, or there is any degree of hostility or unease between the conversants.
Option A wasn’t a flat contradiction only. In fact, the original person wrote it up in a more pussy footing way than I would.
Flat contradiction would be “you’re wrong”. I agree that’s not an invitation to further discussion.
My usual comment would be of the form: “That’s wrong. Blah di blah isn’t blah di blee, it’s hooty hooty.”
It’s “you’re wrong” plus some evidence on which I based my disagreement. Would that be unclear to you personally, that you’re welcome to disagree and cite evidence for your disagreement in turn?
Maybe we could try an example so that we’re talking about something concrete. I just don’t think it’s a mystery. I think that a great many people get very touchy when it comes to being disagreed with. I’m of another species that likes to be disagreed with, because then we have a contradiction to resolve, and that’s fun and potentially productive.
I’m sorry: for reasons I do not understand, I misunderstood what you were referring to with ‘option A’. Your response made perfect sense and mine did not.
You thought I meant “flat contradiction”?
That didn’t seem like an accurate characterization of option A to me, so I gave a concrete example:
and a concrete example of the option A alternative:
It would have been better to be more concrete.
Was that the issue?
I feel that in these more personal discussions abstract terms gets used, and each side is picturing a very different part of the spectrum for their concretes.
I think it was that I kind of short-circuited ‘option 1’ into meaning ‘the first option mentioned’, and from there ‘what the guy said in the first place’. This is not what you were referring to by ‘option 1’, and even though it’s an understandable error, I still should have been able to pick up on it from the context of the parent and grandparent comment to yours.
Well, except that you would not be actually stating an invitation or request for more information. You would be assuming that the other person will interpret contradiction as an invitation for further discussion rather than as a dismissal, insult, threat, or other sort of speech act.
(Humans use language for a lot of other purposes besides the merely indicative, after all.)
If you say, “I’m having a party on Saturday,” some people in some situations will take this to mean that you are thereby inviting them to come to the party. Others will think that you are merely stating a fact about your own social life. Still others will think that you are excluding them, just as if you had added, ”… and you’re not invited, you disgusting worm!”
Some people hear an invitation. Some hear a statement of fact. Some hear an exclusionary insult.
If you want to make it clear that you are inviting them, you say, “I’m having a party on Saturday, would you like to come?” or ”… and you’re invited!”
This is not bullshit song-and-dance ego-stroking. It is clear communication, and in particular a way to address people’s differing priors about what your communication could mean. It probably depends on recognizing that people have different priors, and that they arrived at those priors legitimately.
(For that matter, if expressing curiosity about other people’s experiences is an effective way to get data from them, then rationalists should practice doing it a lot until it is automatic and cheap System 1 behavior!)
Yes. In this context, and most contexts, that’s my null hypothesis. Isn’t it yours? People are here to discuss, and not dismiss, insult, or threaten.
Do you think I’m here to dismiss, insult, or threaten people? Do you think a large percentage of people here are? Do you think that anyone who says “you’re wrong” is? That strikes me as a bizarre and thoroughly inaccurate prior. Or I certainly believe and hope it is.
Am I wrong? Is it just foolish innocence on my part to think that people are here to discuss, and not stomp on other people to social climb or satisfy sadistic impulses? It wouldn’t be the first time. In other contexts, yeah, there’s a lot of that going on. And it admittedly took me a long time to figure that out. But I don’t see it here. The trouble is, if it were, most of the people who know aren’t going to tell you.
You’re talking intentions, they’re talking effects. This leads to you defecting by accident.
Talking about intentions is to blurt out something stupidly? I’m not following your point.
I was referring more to the comment thread, which is filled with detailed writing in support of sending blunt communication while ignoring that such behaviour ends up losing in practice. If you haven’t actually read that article and its comment thread, you really should.
I read the article, but not the thread.
Losing, in what game? Are you sure EY knows the game everyone is playing? I think he is making implicit assumptions about motivations that are incorrect.
I disagree with his strategic analysis. In some contexts I would consider it correct. Yes, I knuckle under and be what “normal people” want me to be, to avoid the costs of being myself, just as all those normal people are busy being what they think other people want them to be.
But where I can, I seek to escape that mutual cage. Internet forums are a place where escape is possible, because the normals no longer have an overwhelming majority, or might not even have a majority at all, and the cost of anyone’s disapproval online is less.
Dale Carnegie teaches you to be the person other people want you to be; I’d rather find the people who like who I want to be, and want to be who I like.
An anecdote from my dissertation adviser. He was having much the same discussion with me, telling me how professors in Asia were allowed less direct intellectual confrontation. Perhaps EY would be proud.
But the discussion went on to the joy of moving to the US, exemplified by another professor he knew, who responded to someone else in a discussion by gleefully retorting “I Disagree! I Disagree! I Disagree!” Free at last, free at last, thank God almighty, free at last! Free to be honest, free to be open, free to be who you are.
I want to sit at the table where they’re dealing that game. It seems like there are enough people of my ilk at this party for us to have a few tables. If the cool kids don’t want to sit at the nerd tables, that’s fine, and hardly anything new.
Not really, no. People use language for a hell of a lot of other things besides making statements of fact at each other. I expect that in any given speech act, a speaker may be doing a lot of things: stating facts, affirming or challenging a social relation with the listener, causing the listener to have expectations about the speaker’s future actions (promises, threats, plans, etc.), and so on. And that a lot of these things may be going on unconsciously.
If someone tells you that the way you speak gives the impression that you are arrogantly dismissing them, you could respond by merely instructing them (in the very same tone that they were talking about) that you do not intend to arrogantly dismiss them. However, doing that is not likely to be very convincing!
Yes, people use language in many ways.
But I should have been more specific.
In your prior for Less Wrong discussions, when someone responds to a statement of yours by saying that you’re wrong, and cites evidence for his claim, what are the probabilities you place on the following potential motivations for his reply—he wants to discuss the point, he is threatening you, he is dismissing you, he is insulting you, other?
Sorry, I should have been more specific — I can tell because you’re asking a question that would only make sense in a different context. My probabilities about whether you intend to be threatening are are not at issue here.
At issue in this thread is that some portion of the audience are not sticking around — and are forming negative conclusions about LW — because the words here come across as hostile, unfriendly, cold, and so on. This is a danger to LW’s goals.
This is a matter of instrumental rationality, not only epistemic rationality. We want to accomplish something with words, not merely possess accurate beliefs in our own solipsistic internal monologues. So we have to ask, are our uses of words accomplishing the goals that we care about?
If you emit sentences that are consistently misinterpreted, and you are informed of this, you have a few options of what to do. You could conclude ① that your audience is listening wrong, and needs to correct their assumptions about you before they will be able to understand you; or ② that you are speaking wrong, and you need to correct your assumptions about your audience before they will be able to understand you.
If you care about getting your meaning across, which of these conclusions is more likely to give you the ability to accomplish that goal? Either one is consistent with the evidence; but which conclusion strengthens you, and which weakens you?
You can’t reach into your audience’s minds and force them to interpret your words differently.
You can’t force them to stick around and listen to you correct their assumptions, either.
You can change the way you speak.
Concluding that you are misinterpreted because your audience is listening wrong, or is coming into the conversation with crazy priors, weakens you. Thinking that way would make you incapable of fixing the situation; less able to accomplish goals by speaking. Concluding that you are misinterpreted because you have misspoken, or failed to understand where your audience is coming from, gives you the power to fix the situation and accomplish goals. This strengthens you.
Well, failing at epistemic rationality because we prioritized PR over truth-seeking is an even bigger danger to LW’s goals.
Maybe I missed it; did you give the same speech about how empowering it is to focus on what you can change about yourself to those who are taking offense at the speech of others? You understand you could have, right?
Not if it’s true.
If it’s true, knowing the truth strengthens me. Just because I think it’s true, doesn’t mean I can’t choose to adjust my speech to them.
And yes, when someone has a false impression, and wants to have an accurate one, you can often change their minds by offering evidence for them to update on.
And assuming that I’ve mispoken does weaken me. It assumes I can “fix” the situation by speaking differently. Ok, compared to what have I misspoken? Compared to preemptively changing my speech patterns so that those with extremely high priors of hostile intent from me are less likely to take offense? Do I have a better alternative?
As a first cut, I’m better off talking to people who don’t assume hostility from my style of speech, who can talk to me ‘as is’ in a productive manner. Seems to be a number of such people. To the extent that they’re similar to me, they will be annoyed and possibly offended by speech acts which seem aimed at managing their potential hurt feelings over my disagreements with their opinions. But even removing these emotional factors from the equation, my attempts to manage their feelings take time and effort from me, and wastes time and effort for them on issues extraneous to the topic at hand. At best, altering our styles will waste our time, and at worst, annoy the hell out of each other. That’s a cost.
I’m to bear that cost, for what?
To talk to others who find my manner hostile? Should I unilaterally cave to every demand that I change my manner when they say they feel hurt or offended? On a game theoretic basis alone, that seems like a bad idea. I am to be malleable to their preferences. Ok, I willing to look at that.
I have been having discussions on adjusting my speech patterns to avoid impressions of hostility in others. Even came up with an idea that someone thought was a good step forward—say “I disagree” instead of “You’re wrong.”
And them? Are they to be malleable to my preferences? Not that I’ve seen.
Where are the discussions of the “offended or hurt” adjusting their priors to better reflect the reality that “meanies” like me really aren’t here to insult, offend, or demean them? Or even where are the discussions that assume the priors are correct, but look for ways to suck it up and develop a thicker skin to better deal the hateful bastards trying to hurt them?
I don’t see those. What I’ve seen are offended rejections of any suggestion they might work on changing their reactions, by them, and often by those defending them.
Why is change a one way street?
I was the person who said going from “You’re wrong” to “I disagree” was an important step. I’m glad it registered.
Becoming less thin-skinned takes time and sometimes a good bit of work. You don’t know where any particular person is in that process.
You might be in a Pareto’s Law situation—it’s not that you need to avoid offending the most fragile people, a small amount of effort might lead to not offending 85% of thin-skinned people.
Yes, but I didn’t want to finger you as the culprit.
Indeed.
My concern is that some seem to consider it a crime to suggest that this would be a desirable thing, or to suggest that people adjust their inaccurate priors for hostility downwards when interpreting the actions of others.
The nicies want the meanies to try harder to be understood.OK, fine. But if the meanies suggest that the nicies try harder to understand, that’s just one more thing for the nicies to get offended about.
I would note to all the nicies—if LW feels hostile to you, you’ve led a very sheltered existence. I’m from the HItchens party of debate, that prefers a sharp point be embellished with a barb. That’s part of the fun, in the same way that a decleating hit in football is part of the fun. And that’s not even hostility, that’s just style. And that’s a common style.
By my estimate, LW has a very “Just the Facts Ma’am” culture. Going beyond the facts and putting any relish into a debate is rarely done, and frowned on when it happens. Maybe there were nastier times in the long long ago that led to this culture. LW does seem relatively unique in the equal mix of LIbertarians and Progressives.
And you’re right about Pareto’s law too.
FWIW, one of the things that caught my attention about this community, and encouraged me to stick around, was the emphasis on valuing accuracy and precision (which I value) without the “barbs are part of the fun/putting relish into a debate” style you describe here (which I dislike intensely).
There are lots of “nice is more important than true” spaces on the net, and lots of “being unpleasant to people is part of the fun” spaces; and an astonishing number of spaces that are both. A space that manages to even approximate being neither is rare.
I can understand someone disliking the barbs.
Barbs are largely the part of the classical rhetoric that play on biases in the listener. That’s probably true of niceness as well.
And I agree that the Lesswrong tone seems relatively unique, particularly given the broad and general nature of discussion topics, and the variety in political opinions.
One of the good things which contributes to the tone here is people reliably getting credit for saying they’ve changed their mind for some good reason. I can’t think of any other site where that’s in play.
Isn’t it peculiar that most people are otherwise?
Way back when, I remember discussing exactly that point on the Extropians list. In many ways a similar group to here. But some very smart guys were arguing that it was a huge loss of face to admit you were wrong, and better to deny or evade (I’m sure they put it more convincingly than that).
When someone is wrong, graciously admitting and accepting it scores major points with me.
Thinking about it, maybe I can make a better argument for denial. There are two issues, being wrong, and whether one admits being wrong. If admitting being wrong is what largely determines whether you are perceived as being wrong, then denying the error maintains status.
For people driven by social truth, which is likely the majority, truth is scored on attitude, power, authority, popularity, solidarity, fealty, etc. The validity of the arguments don’t matter much. For people driven by epistemic truth, the arguments are what matters, so denying the plain truth of them is seen as a personality defect, while admitting it a virtue.
The thing is, it’s not that the deniers are aliens. I am. I and my kind. For us, in an argument, it’s the facts that matter, and letting other considerations intrude on that is intruding the rules of the normals into the game. That’s largely what this whole thread is about.
One side says we’ll be more effective playing the normals game. It’s a game my kind strongly prefers not to play. Having to behave as normals is ineffective for us, and the opportunity to play by our rules is extremely valuable to us.
Josh Waitzkin’s book The Art of Learning describes his various encounters with unsporting conduct and cheating in chess tournaments and competitive tai chi. He wrote that he’d developed the approach to just work so hard at developing his own skill at the game that he was able to ignore the distractions the opponent was trying to pull and proceed to win anyway. He claimed that the opponents would generally become agitated and careless once they noticed that they couldn’t get any sort of upset out of him.
Discussions aren’t games with rules, but you might still get something out of the idea that social gamesmanship is basically just compensating poor skill with cheating, and you need to work hard enough on your epistemic skills that it won’t stop you even when it does get thrown in your way.
Well, I’d say that social gamesmanship isn’t cheating, it’s playing a different game.
Being very good with your epistemic skills has mileage socially too, and importantly, mileage with people with personal properties you’re more likely concerned about. And refraining from the usual types of social gamesmanship earns you points with those people as well.
I’m actually very fond of being told I was right. (I only figured that out when a friend mentioned that he’s very fond of other people admitting they were wrong.)
It’s true that there’s currently a belief that it’s very bad to tell people they should be less thin-skinned. People generally want a social environment which suits their preferences, and while it’s not likely that anyone will get a total victory, it’s certainly possible to push the balance towards your preferences.
Thin-skinned people are apt to hear a demand that they be thicker-skinned as “You shouldn’t care about the way I keep hurting you.” The more aggressive among them have started shoving back. Interesting times.
IMO when you write, you should be asking yourself: “What’s the worst way someone could interpret this?”, because surely, someone will interpret it that way. And when you read, you should ask yourself: “What’s the nicest way I could interpret this?”, because that’s probably the way they meant it.
Postel’s law FTW!
When dealing with people, habitually searching for only the worst that can happen is a very bad habit, in my experience. It’s a habit I’ve been trying to break. Through availability bias, your world becomes a horrible place. Your priors are distorted toward the bad, and you miss opportunities. Too careful, too risk averse, too distrusting.
I think that’s the right policy, even if it’s not true. It will generally be the more productive assumption—particularly for online forums.
Just work out the cases. Search for everything that can happen. Either a person has basic good will towards you, or they don’t.
If they do, the nice interpretation is likely right, and you understand someone with good will toward you. That makes for a good discussion. Further, if the guy meant it in a nasty way, your response as if he were nice might soften his mood, or not. If it softens, things have at least improved. If not, most observers will likely think him a schmuck, and he is just very unlikely to be a good discussion partner anyway.
If they do have good will, but you assume that it is bad, you’re likely limiting the positive outcomes available with them. If they don’t have good will and you assume they don’t, you have maybe avoided some aggravation and saved yourself some time.
Having worked out the general case, you don’t have to do a de novo analysis each time. Commit to the policy, and blithely move on. Sometimes someone won’t like you. Ok, you knew that was going to happen.
This is what I’ve tried to do in general with my own defensiveness with people. Don’t focus on the worst that a person might do. Try to have an accurate prior on intentions (most people are not con men or mass murderers, and they’re not really out to get me—I’m not that important to them.) Pick a decision based on an analysis of of what their intent and attitudes might be, and the differing outcomes based on your actions.
Most of the analysis applies, except real world encounters carry more serious risks. I live in the Seattle are, which is pretty safe and so real world risks are limited, though I realize not everyone lives in such a safe place, so YMMV.
In general, the best strategy is to act assuming approval and good will, because those situations present the best opportunities.
I previously relayed an anecdote from a book on this: http://lesswrong.com/lw/s0/where_recursive_justification_hits_bottom/4wsn
Ah, but that wasn’t what I meant. I just meant to say that you should be careful when writing, because even when 99%+ of people won’t have any problems with what you write, someone is sure to misinterpret it, if it possibly can be. Communication is hard, and written communication even more so.
I’d say more briefly “someone is sure to misinterpret it”, because it is always possible to do so. There’s going to be a level of misinterpretation no mater how you agonize over what you write.
I agree with you that the underlying good will or lack of it is a crucial factor. I’m still trying to figure out what tends to build good will or damage it.
One problem is, what builds good will with one may erode good will in another. Life is full of trade offs.
“Offended or hurt” doesn’t enter into it. This isn’t about hazy feelings; it’s about hard practical effects of actions: do we accomplish what we want to accomplish?
Let’s say you and your interlocutor disagreed about your intention in saying that they were wrong (about whatever). Your interlocutor believes that your intention was for them to shut up and go away, but actually that wasn’t what you meant at all; you meant to invite more discussion.
They are wrong about you.
And you want them to have a correct belief about you.
But … how can you cause your interlocutor to possess a correct belief about your intention? You could lecture them about how wrong they are to have misinterpreted you. But that won’t work if they will take your lecturing as meaning “shut up and go away” … and may very well do so.
That’s all I’m saying. You can’t force people to understand you, or to want to understand you. If you really want to get your ideas across (because you care about those ideas — not because you’re trying to find people who will easily like you) then you use the try harder which probably involves restating them in a way that doesn’t repel people.
Or … well, you could say that you never really cared about that kind of person’s understanding, and really you never wanted a discussion with that kind of person.
But in that case … they weren’t wrong about you, were they?
There are plenty of people who would be correct in concluding that I would bear them hostility if I knew what they were like.
They would be incorrect to conclude that the priors I assign to that type of person among LW is very high, and incorrect to assume that my asserting that someone is wrong indicates I have concluded the person is that type of person, so that my comment indicates hostile intent.
Perhaps I’ve given you an incorrect impression.
While I have proselytizing tendencies, that’s not my fundamental goal, particularly in a forum disagreement. Given my limited resources of me, my proselytizing attitude is to sing to those with the ears to hear. People who are assuming that I am hostile are not the low hanging fruit in that regard.
But people who assume I am hostile can be perfectly fine partners in a disagreement. In a disagreement, I am primarily hoping to change my own mind, whether in correcting an error, or clarifying hazy positions of my own. They might even be better, in that they won’t cut me slack when I am sloppy. People who dislike you can be perfectly useful in a discussion. The enemy of my enemy (our ignorance) is my friend.
But I find it strange that you think I should find it hopelessly futile to try to change a person’s assumptions about my intent, but a productive use of my time to try to change their minds about some other fact of reality.
Indeed I agree that it is possible, and probably desirable, to phrase the argument less bluntly than I did. However, it seems to me that submitter B is arguing against making such arguments at all, not arguing to make them in a more polite fashion.
Furthermore, here of all places, “If you (think you) posses evidence that I do not, show it and update me!” should be a background assumption, not something that needs to be put as a disclaimer on any potentially-controversial statement.
I rather doubt that submitter B would have had a problem with, “Really? Why? I ask because from out here it seems like you’re a thinker.”
Certainly the cited reasons for the actual statement being objectionable do not apply to this modified form.
She writes:
Which I read to mean that she is not opposed to them expressing confusion or saying something like “Huh, you always seemed more like a pure thinker to me.” (as opposed to “No way. You’re totally a thinker.”) It seems precisely how the statement is phrased and how the discussion is conducted that is at issue here.
After the discussion, I think I’ve got a more concise option that achieves this end.
Option 4:
I disagree, because blah blah blah.
Concise, and makes it about my differing perceptions and evaluations. Better than my original “you’re wrong, because blah blah.” I doubt that this entirely satisfies the nice camp, but I think it’s a baby step in their direction.
I think it’s actually a fairly large step, but I’m probably a moderate on niceness.