Communication fails when the participants in a conversation aren't talking about the same thing. This can be something as subtle as having slightly differing mappings of verbal space to conceptual space, or it can be a question of being on entirely different levels of conversation. There are at least four such levels: the level of facts, the level of status, the level of values, and the level of socialization. I suspect that many people with rationalist tendencies tend to operate primarily on the fact level and assume others to be doing so as well, which might lead to plenty of frustration.

The level of facts. This is the most straightforward one. When everyone is operating on the level of facts, they are detachedly trying to discover the truth about a certain subject. Pretty much nothing else than the facts matter.

The level of status. Probably the best way of explaining what happens when everyone is operating on the level of status is the following passage, originally found in Keith Johnstone's Impro

MRS X: I had a nasty turn last week. I was standing in a queue waiting for my turn to go into the cinema when I felt ever so queer. Really, I thought I should faint or something.

[Mrs X is attempting to raise her status by having an interesting medical problem. Mrs Y immediately outdoes her.]

MRS Y: You're lucky to have been going to a cinema. If I thought I could go to a cinema I should think I had nothing to complain of at all.

[Mrs Z now blocks Mrs Y.]

MRS Z: I know what Mrs X means. I feel just like that myself, only I should have had to leave the queue.

[Mrs Z is very talented in that she supports Mrs X against Mrs Y while at the same time claiming to be more worthy of interest, her condition more severe. Mr A now intervenes to lower them all by making their condition seem very ordinary.]

MR A: Have you tried stooping down? That makes the blood come back to your head. I expect you were feeling faint.

[Mrs X defends herself.]

MRS X: It's not really faint.

MRS Y: I always find it does a lot of good to try exercises. I don't know if that's what Mr A means.

[She seems to be joining forces with Mr A, but implies that he was unable to say what he meant. She doesn't say 'Is that what you mean?' but protects herself by her typically high-status circumlocution. Mrs Z now lowers everybody, and immediately lowers herself to avoid counterattack.]

MRS Z: I think you have to use your will-power. That's what worries me--I haven't got any.

[Mr B then intervenes, I suspect in a low-status way, or rather trying to be high-status but failing. It's impossible to be sure from just the words.]

MR B: I had something similar happen to me last week, only I wasn't standing in a queue. I was sitting at home quietly when...

[Mr C demolishes him.]

MR C: You were lucky to be sitting at home quietly. If I was able to do that I shouldn't think I had anything to grumble about. If you can't sit at home why don't you go to the cinema or something?

The level of values. Here the participants of a discussion are primarily attempting to signal their values. Any statements that on the surface refer to facts actually refer to values. For instance, "men and women are equally intelligent" might actually mean "men and women should be given equal treatment" while "there are differences in the intelligence of men and women" is taken to mean "it's justified to treat men and women unequally".

The level of socialization, also known as small talk. You aren't really talking about anything, but instead just enjoying the other's company. If the group is seeking to mainly operate on this level, someone trying to operate on the level of facts might get slapped down for perceived aggression if they insist on getting things factually correct.

For rationalists to succeed in spreading our ideas, we need to learn to recognize which level of conversation the discussion is operating on. One person acting on the level of facts and another on the level of values is a conversation that's certain to go nowhere. Also, it took me a while to realize that there have been occasions on which I was consciously trying to act on the level of facts, but my subconscious was operating on the level of status and got very defensive whenever my facts were challenged.

Usually what rationalists would want to do is to move the conversation to the level of facts. Unfortunately, if a person is operating on the level of values, they might perceive this as an underhanded attempt to undermine their values. I'm uncertain of what, exactly, would be the right approach in this kind of a situation. Defusing the level of status seems easier, as people will frequently find their unconscious jockeying for status silly once it's been brought to their conscious attention.

 

New Comment
73 comments, sorted by Click to highlight new comments since:
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

This post is a decent first approximation. But it is important to remember that even successful communication is almost always occurring on more than just one of these levels at once.

Personally I find it useful to think of communication as having spontaneous layers of information which may include things like asserting social context, acquiring knowledge, reinforcing beliefs, practicing skills, indicating and detecting levels of sexual interest, and even play. And by spontaneous layers, I mean that we each contribute to the scope of a conversation, and then those contributions become discerned as patterns (whether intended or not).

Then iterate this process a few times, with my attempting to perceive and affect your patterns and you attempting to perceive and affect mine. Add some habitual or built-in (it's extremely hard to tell the difference) models in the mind to start from and it seems simple (to me) how something as complex and variable as human communication can arise.

7[anonymous]
Voted up. I think a better metaphor is that of dimensions (since a conversation can easily take on several simultaneously) rather than levels.
1Kaj_Sotala
I think you're probably right.
1SilasBarta
My peeve (well, one of many) is when I can't distinguish between flirty non-rejection rejections, vs. real, "you're inches from being a stalker" rejections, and there are severe penalties for erring in either directions. And I suspect that if it were possible to distinguish them, that would be a bad thing (for women). ETA: Downmod justification requested for this and my follow-up comment.
3Strange7
My inclination in cases like that is to walk away, and go play with someone who respects me enough to say what they mean.
6jhuffman
One of the more important functions of the social level interaction is signaling mating interest (not saying this doesn't happen on other levels too, but generally playful flirting is the first movement of the mating dance). Refusing to play is choosing to lose some opportunities. A lot of people think that because they are not yet good at a particular game (or even understand that it is a game) that the game has no value for them. Other people are inclined to try and figure out how to play it better.
5Strange7
I'm not categorically refusing to play; I'm just saying that after a particular move (flirty mock-rejection) on the part of the other player, there are no moves I can make with best-case outcomes good enough and reliable enough to outweigh the risks. Under those circumstances, it seems like the rational decision is to stop playing. As I see it, a game you only need to win once for decades of marital bliss (plus offspring, financial stability, etc.), or lose once to land on the sex-offender registry, is not a game which should be played casually or with excessive frequency.
2jhuffman
Well its not like you go straight from flirty mock-rejection to stalking or non-consensual sex. The worst you can do is spend more time with her after she has rejected you; and if her gentle (flirty) rejection doesn't work she's going to signal it more clearly.
7Strange7
One way or another, it sets a bad precedent. If she's just flirting, and I stay, I'm implicitly accepting the burden of deciphering any further ambiguous statements she might choose to make, for the duration of the relationship. If she honestly wants nothing to do with me, but is speaking gently because she also doesn't want to give offense, and I stay, that persistence might intimidate her into silence, which can lead to some very unpleasant outcomes indeed.
1jhuffman
This is where I insert a joke about married life. But I don't want to be down voted anymore. So just think up the joke yourself.
6thomblake
Good jokes get voted up, not down.
1jhuffman
Yes, I actually knew this. The sad part is I thought what I wrote was a good joke. But I was wrong.
0Strange7
Humor, grain of truth, etc. Whoever told that joke the first time, I'm guessing, settled for a pretty girl who'd talk to them at all instead of thinking it through.
2SilasBarta
That's fine, if you don't mind stalker-erring genes to dominate future gene pools. (Individually, it's probably better for you than risking legal consequences, but it's a Hobson's choice.)
2Strange7
I'll burn that bridge when I come to it.
0[anonymous]
That's fine, if you don't mind stalker-erring genes to dominate future gene pools. (Individually, it's probably better for you than risking legal consequences, but it's a Hobson's choice.)
0[anonymous]
One of the more important functions of the social level interaction is signaling mating interest (not saying this doesn't happen on other levels too, but generally playful flirting is the first movement of the mating dance). Refusing to play is choosing to lose some opportunities. A lot of people think that because they are not yet good at a particular game (or even understand that it is a game) that the game has no value for them. Other people are inclined to try and figure out how to play it better. Edit: I actually meant this as a response to Strange7.
-1[anonymous]
.
[-]Kevin110

I think there's a higher level than the level of facts: the meta-conversation. You know, the stuff we do here. I can now basically win almost all philosophical or political debates (at least with people in their early 20's, I could use a bigger sample set of people to debate) by taking the conversation meta. I get us to settle on definitions, find out what we disagree on, analyze the thought patterns that lead us to that disagreement, dissolve the question as necessary, and usually find out that we don't disagree as much as we thought we did. For example, I find that most political debates can be reunderstood in terms of pragmatism vs. idealism. If you do it right, it doesn't even feel like an argument -- you shouldn't be triggering an emotional reaction from the other side.

I haven't won that style of conversation yet with a real live anti-reductionist though. My one encounter with an anti-reductionist in the last several months ended with him contradicting himself and then changing the subject.

7nerzhin
What do you mean "win"? If you have a conversation with someone, during which you discover a previously held belief you had was wrong, and revise your belief, have you "lost"?
2Kevin
I was being overly dramatic with the win/lose terminology. In those I have conversed with , I have caused a change in belief of greater magnitude than the typical belief change resulting from another philosophical or political debate, but not a true "win", where I cause someone's beliefs to update in agreement with my own.

I suspect that many people with rationalist tendencies tend to operate primarily on the fact level and assume others to be doing so as well, which might lead to plenty of frustration.

Also, it took me a while to realize that there have been occasions on which I was consciously trying to act on the level of facts, but my subconscious was operating on the level of status and got very defensive whenever my facts were challenged.

Usually what rationalists would want to do is to move the conversation to the level of facts.

Oh boy am I guilty of this. I've been ... (read more)

3reaver121
I have another annoying habit. I tend to get rather ... enthusiastic in discussions thanks to applying The mind projection fallacy to my discussion partner. Sometimes if find a certain fact X so glaringly obvious that I tend to assume that other people also find it obvious. So, when the discussion starts I think that we are both on the same page when we are not. This leads to me misunderstanding their arguments. From my point of view I seems like they are doing it on purpose which makes me rather flustered. I usually takes me a while in such cases to realize that they not know about X.
[-]Jack80

I suspect that many people with rationalist tendencies tend to operate primarily on the fact level and assume others to be doing so as well, which might lead to plenty of frustration.

Even when rationalists realize they aren't talking about facts they often try to apply their rules and methods for dealing with facts to these other levels (for about the 12th time in the past week I reiterate my objection to calling any value system more rational than another).

In status level, there is a classic trap rationalists (in status level communication is the right... (read more)

Also, we should probably avoid using "rationalist" when we really mean something like "on the autism spectrum"

I think we should avoid using "on the autism spectrum" when we mean something like "about one sigma removed from stereotypical social norms".

I think we should avoid talking about people as being either on or off the autism spectrum. For the term to be useful, "autism spectrum" should refer to a spectrum that includes everybody. There's a cluster that we can call neurotypical, and off in one direction we have some people who share a combination of traits called autism, and if you draw an imaginary n-dimensional line between them, you can meaningfully talk about where any particular person is on the autism spectrum.

"Autism spectrum" should refer to the spectrum of variation in the traits that autism affects. The current usage is confusing and doesn't do much to help us talk about the people who are (as you put it) only about a standard deviation or so from the norm.

Conversation differs from argument in more ways than just being about "socialization" rather than being about facts.

For instance, you can think of conversation as a process of sending out "probes" that are echoed in modified form by your interlocutors; the distortions reveal something about the mind that did the echoing.

Okay, so can you operationalize the distinction between the levels? Or offer an idea as to why the other levels would masquerade as fact-level exchanges? Plausible deniability?

And is 'rationalist' a good term for the behavioral pattern you're writing about, seeing how it tends to get associated with the contrasting notion of empiricism (at least in my mind)?

2Kaj_Sotala
Nice to see you registered here. Do stick around. :) A decent operationalization is rather tricky, especially since the levels mix and intertwine with each other, as others have pointed out. As for why the others look like fact-level exchanges... good question. Plausible deniability doesn't sound like a bad guess, especially in case of status games. Since a lot of the jockeying for status seems to happen without us even being consciously aware of it, there might also be a degree of self-deception involved. Though for what purpose, I am unsure. Also, it might simply be the fact that it's easier to layer on new content on existing systems of communication than come up with entirely new ones. If most people are capable of keeping the different levels apart without conscious thought, then there's no particular reason to separate them more. The definition of "rationalism" used on this site is somewhat different from the traditional rationalism vs. empiricism division - see this post. Of course, even going by our definition, it's not that an inability to differentiate between the levels would be part of the rationalist definition in any way (quite to the contrary). It's only that people drawn to the kind of analytical reasoning characteristic of rationalists tend to often have weaker social skills. (In other words, nerds are overrepresented among rationalists.)

I think this post underestimates the possibility for mis-communication on the "fact" level.

First, facts are hard to figure out. This is a fact (heh) worth repeating again and again in a community like this one. When one claims to search for objective truth, one must be held to a very high standard.

Second, two people operating at the 'fact' level might misunderstand each other if they are making observations on different conceptual levels. Russel never accepted Godel's incompleteness work because he was unwilling to engage with Godel's meta-leve... (read more)

0jhuffman
Where does the OP say we are trying to discuss values objectively? If you are signaling your own values, obviously you are attached to them. If there are objective facts about values, they can be discussed objectively on the fact level. Whether or not in practice anyone can objectively discuss facts about their own values is another question.

The level of socialization is centrally about more than just enjoying the other's company. Usually it's about forming and strengthening a friendship i.e. alliance. Normal people don't care much about the interests of random people, but they do care about their friends i.e. allies. The level of socialization is about acquiring and keeping allies.

(There are, however, probably large differences between different cultures how strongly smalltalk is about what I described above. In some cultures one smalltalks with random people all the time, without expecting to meet them again, whereas in others it's mostly done with just friends and potential friends.)

2jhuffman
I think elevator conversation patterns are interesting. Typically in an office in the US we don't discuss anything at all on an elevator if the people are all strangers - it can be stone silence for 70 floors. However, if there is anything at all discussed (beyond "four please") - even something as trivial as the weather or how bad the traffic was getting off the loop - then usually we feel compelled to impart some sort of leave-taking expression when one of the parties leaves the elevator.
[-]Rain30

Usually what rationalists would want to do is to move the conversation to the level of facts. Unfortunately, if a person is operating on the level of values, they might perceive this as an underhanded attempt to undermine their values.

Why not switch to talk on the level of values, if that's what they want to talk about? After all, there are facts about their values which might be very interesting once you've taken time to examine them, their sources, etc.

I can see now that's what I should have done yesterday when I tried arguing politics with someone from work. Politics is the mind killer because it rarely operates on the level of facts.

[-]Jack140

Politics is the mind killer because it rarely operates on the level of facts.

Or rather, because it almost always operates on both the level of facts and the level of values without a procedure for disentangling the two.

4BenAlbahari
I agree. I run TakeOnIt, a database of opinions of experts and influencers, where the opinions (in the form of quotes) can be labeled with "pitches". Pitches are persuasion patterns, kinda like the tv-tropes of argumentation. One of my goals is to help people become aware of these persuasion patterns. Or to put in another way, give people a procedure for disentangling pitches from facts. You can see the pitches here. I introduced the concept on Less Wrong a little while ago, not with the word "pitch", but with a terrible name that I shall never speak of again. On a separate note, David Foster Wallace has a great piece of journalism that covers this concept in an analysis of the political talk radio industry, here.
0Kevin
Upvoted for DFW's Host!
3Rain
Yes, thank you, that's a useful distinction. Funny, how I hadn't thought of mixing the levels once they'd been neatly labeled and described. Something to watch out for.
3Jack
Come to think of it, most political discussions are entangled with status issues as well.
6Rain
On further reflection, I think the word "level" is misleading. It seems more of a focus, purpose, or goal of the speech, with varying percentages of each goal possible within a single encounter. This also makes me wonder what other topics might be added to the list. I'm reminded of the purposes of writing: to entertain (social), to inform (fact), and to persuade (status/values), but I don't think these map very well, and the categories provided by Kaj may be more useful.
0Jack
Trust building, flirting, group building (think people who just met agreeing or talking about something they all have in common), order giving, and catching up. To start with.
5Nanani
Those all sound like they fit primairily into socialization, with varying doses of status thrown in. Maybe a Venn diagram would work better than strict levels.
0magfrump
Voted up for suggesting a Venn diagram.

blinks

So thats why I always end up in arguments over trivial things with my girlfriend.

So, uh, if you didn't know already, don't worry about 100% accurately identifying what happened that made your girlfriend angry. Even if you did absolutely nothing wrong. She doesn't care about the facts, she's operating on another level...

9Rain
Note: does not apply to all girl/boyfriends.
5AdeleneDawner
Agreed, emphatically. Few things annoy me more than people assuming that I'm like the stereotypical female in that way.
5Blueberry
I think it's fair to say that in general, human beings, male or female, frequently get in arguments about little things when the argument is actually about something bigger. If I'm upset about something major, I'll put that emotional state onto minor details that otherwise wouldn't bother me. I don't know that it's "stereotypical female" as much as "human with emotions".
2AdeleneDawner
People vary. In my case, I'm usually very good at noticing when my emotional reaction to one situation is affecting my reaction to another situation, and I'll admit that that's the case. That's generally not a problem. What is a problem is when someone refuses to believe that something that they've done has actually crossed a line, because 'I must be having a bad day' or 'I'm just oversensitive'. Note that my point isn't that one gender is less prone to that kind of confusion than the other. My point is that women are stereotyped as being more prone to it, which means that people are more likely to use that reasoning to ignore womens' communications than they are to use it to ignore mens', and I really don't like having my communications ignored or discounted.
0Blueberry
I agree completely with your last paragraph. As far as someone refusing to believe they've crossed a line, that's very dependent on their expectations about people's boundaries in general, and yours in particular. They may be used to people who have completely different boundaries, and who genuinely only react to some line-crossing behavior when they are having a bad day. And a lot of the time, when people have a bad day, they aren't good at realizing that: in my experience, most people are horrible at knowing what they're really upset about, and at telling when their boundaries shift due to their mood.
1Jack
Is it that she doesn't care about the facts or that she cares about a different set of facts than the one she is explicitly stating?
1Kevin
You ever try dissolving the question?

Interesting to say that having an unusual medical problem raises someone's status. It sounds intuitively right, but I don't think any theory of status discussed here so far quite covers it.

9AndyWood
I think it shows that status can be contextual. If a small group begins competing over who has the worst illness, then illness becomes a de facto status marker in that gathering. It doesn't mean that illness is a global status marker among humans in general. In context, it may be no more than saying "I am the most superlative!"
0Divide
In Poland there's a whole genre of jokes based on one-upping such ad hoc status markers. It could well be "my husband is so stupid that he...". "I'm so ill that..." fits the genre perfectly.
4Nanani
Might it be as simple as being Special? Someone with an unusual medical problem is deserving of extra attention by virtue of having it.
2Scott Alexander
That goes well with common sense, but how would you go about giving it predictive value? I mean, if I were to announce to a bunch of people one day that I had urinary incontinence, I imagine that would lower my status quite a bit. Certainly from a evo psych theoretical standpoint, it's odd to gain status by talking about how unhealthy and unfit you are. I like what Andy Wood said about it being contextual, and hopefully if Kaj writes a summary of that chapter that'll help explain how the context works.
0TheOtherDave
Well, the first approach that comes to mind would be looking for some actual metric of status... preferably one that doesn't interrupt the conversation. My guess would be that how often a person is interrupted vs. allowed to finish a statement, patterns of eye contact, and participants' physical orientation relative to one another, and other indicators of attention are all correlates of status of the kind we're talking about here. (This is, of course, only useful for determining relative status of individuals within a conversation, and possibly within a culture; different cultures reflect status differently.) Then measure it, and look for speech patterns that correlate with it. Anthropologists (and in a sense, ethologists) do this sort of thing all the time. Well, if low-status players all start talking themselves up as a way of gaining status, they create a context where someone can ostentatiously refuse to talk themselves up as a way of signaling that they are too high-status to care, which in turn creates a context where someone can ostentatiously talk themselves down as a way of one-upping that. If something like this is going on, then it's possible that talking about how unfit I am becomes a way of signaling how fit I am. OTOH, talking about how unhealthy I am can also be a form of malingering, which has certain immediate practical benefits. That is, it might not be an evolved trait at all, and its prevalence in a particular culture might just be historically contingent. One way of teasing this apart would be to see, in communities that do this sort of status-jockeying, whether the people who most claim illness are in fact perceived as ill by their peers.
1Kaj_Sotala
Incidentally, I recommend that people interested in the topic of status read the entire linked chapter. It contains rather interesting insights.
0Morendil
It's actually an excerpt - 1/4 of the chapter entire. (Just got the book from Amazon yesterday, going through it to research the topic.)
0Kaj_Sotala
Yeah, that's true. I have the entire book, and started thinking about writing a summary of that whole chapter to be posted here. Maybe in a few days.
0Morendil
I've been working on something along those lines as well. Let me know if you'd like to coordinate, or trade notes, or just each do our own thing.
0Kaj_Sotala
I haven't really gotten started, so if you've already began working on it, I'll probably just let you handle it - I have some other things I should be posting on, too. :)

For rationalists to succeed in spreading our ideas, we need to learn to recognize which level of conversation the discussion is operating on.

For rationalists to improve their skill at persuading others of their complex opinions, perhaps we should study the experts: politicians. I would find it surprising if the average politician was not disturbingly good at switching between the levels of communication outlined in this post.

Nice post. I don't think people ever really step out of the status level. Maybe when thinking alone or among trusted friends...

The level of facts only works right when the topic is status neutral. This is my guess from numerous anecdotal evidence.

I guess I question the accuracy of breaking up communication into separate levels, and/or these levels in particular. This isn't a taxonomy we're talking about (or is it??) Also, I don't like the example of "status conversation" given here. What if I disagree with your analysis? Well, you can't say much, because it's not an objective subject, which is exactly why you divorce that category from the category of facts. But if you're divorcing it from facts, don't inject it full of meaning that may not be correct. Instead, let's say it's non-factual oriented, and then figure out an assessment that's guided by that definition.

3[anonymous]
It took me a long time to figure out that you were indeed saying something meaningful, rather than random babbling along these lines: "You might be wrong. I don't like it. What if I disagree? You can't say much about this subject because it's not objective. You're doing something wrong; you might be wrong. Let's use this term instead." I still don't understand your comment fully. I wish you had written something like this: "I guess I question the accuracy of breaking up communication into these levels. It's difficult to talk about the idea of levels of communication, because it's so subjective. or: In a status-level conversation, a person can't say much, because they're not really talking about anything objective. This is why you consider the status level and the fact level to be separate. But if you're considering these to be separate, don't make the assumption that the so-called status level is actually about status; call it the non-fact level, and make inferences based on that definition." Notice that there was one sentence that I managed to come up with two entirely different interpretations of, and there were a few sentences I omitted because I couldn't figure out how they were relevant to anything.
-1zero_call
I always aspire to speak as clearly and plainly as possible. Unfortunately that may not come across as accurately as I always wish. Anyways, you basically decoded what my meaning was here, and I appreciate your attention to clarification. In lieu of your two interpretive sentences, I would restate "What if I disagree with your analysis? Well, you can't make a strong argument against my interpretation, because the subject of both our interpretations is a subjective one. The interpretation of people's comments (and social interaction in general) does not follow a logical truth table. Etc etc" It looks like you didn't understand my point about the taxonomy. To restate myself, what I'm saying is that it's not clear that communication is something that can be chopped into bits and exist as the same thing. It's not like we're talking about an stew with a ham bone, peas, chicken broth, etc. inside of it. We're talking about a complex entity that doesn't necessarily amend itself to this procedure. On the other hand, maybe it does -- but that requires justification. All of this (in my mind) should come before the principal part, the justification of the particular sub-levels that have been offered. And when those sub-levels have been offered, they ought to fit naturally into this scheme of a greater understanding of what communication constitutes. Overall, I would be looking for a more thorough analytic process than the one that starts the beginning of this post.
0Kaj_Sotala
These are levels (or dimensions) that seem right to me, based on my experiences. If you disagree, you're welcome to offer alternative ones.
0RobinZ
Remember the analysis of the utterances as status communication occurs on the level of facts - you certainly can disagree with the analysis factually.
0zero_call
Of course, but that doesn't make it convincing.
2RobinZ
I wasn't trying to make it so. What is your interpretation of the example?
0[anonymous]
(No, the analysis is what I am saying is subjective.)