Gricean (im)politeness

« previous post | next post »

Does Paul Grice's "cooperative principle" enjoin politeness? Jessica Wildfire sees it that way ("Maybe you’re not rude after all", Splattered 6/29/2018):

A teacher sent me home for showing my underwear in fifth grade. The same year, I also got in trouble for asking a classmate about their gender identity. Stuff like that was always happening. I always managed to break some invisible rule out of social blindness. […]

I was a real trouble maker. So rude. Why couldn’t I just be polite, like everyone else? […]

And then linguistics happened. Halfway through college, I started taking courses in language theory.

That’s when I started to learn something important. Something that changed my life forever.

Most of our politeness rules are bullshit.

What was her crucial "language theory" insight?

You see, sociolinguists study how people talk. The more I read into language theory, the more these elusive, hidden, and unspoken rules and expectations became clear. My problems came into focus.

For example, you can make pretty decent conversation if you just follow four principles laid out by Paul Grice in the 70s — quantity, quality, relevance, and manner. That was revolutionary.

Grice taught me that when someone you barely know says, “How are you?” they actually don’t give a shit. They’re just being polite. You’re supposed to observe the maxim of quantity and say, “I’m fine.”

If you tell most of your friends or coworkers how you actually feel, they’ll think you’re unhinged. All because you offered more information than they wanted. I could go on, but you get the point.

See, quantity.

Paul Grice was a philosopher, not a sociolinguist. And the goal of his work on conversational maxims was to rescue the truth-conditional meanings of words such as not, and, or, if, all, some, the. He proposed to do this by explaining the apparent divergence of their ordinary-language use from simple truth-table meanings as cases of "conversational implicature", a "mistake [that] arises from an inadequate attention to the nature and importance of the conditions governing conversation".

But it's true that his 1967 analysis of "Logic and Conversation" is founded on what he calls the "Cooperative Principle":

Our talk exchanges do not normally consist of a succession of disconnected remarks, and would not be rational if they did. They are characteristically, to some degree at least, cooperative efforts; and each participant recognizes in them, to some extent, a common purpose or set of purposes, or at least a mutually accepted direction. This purpose or direction may be fixed from the start (e.g., by an initial proposal of a question for discussion), or it may evolve during the exchange; it may be fairly definite, or it may be so indefinite as to leave very considerable latitude to the participants (as in a casual conversation). But at each stage, SOME possible conversational moves would be excluded as conversationally unsuitable. We might then formulate a rough general principle which participants will be expected (ceteris paribus) to observe, namely: Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged. One might label this the COOPERATIVE PRINCIPLE.

So can we summarize the cooperative principle as "Be Polite"?

Problem: many of Grice's own examples of conversational implicatures are downright nasty. For example, he invites us to

Compare the remarks:
(a) Miss X sang 'Home sweet home.'
(b) Miss X produced a series of sounds that corresponded closely with the score of 'Home sweet home'.

It's the maxim of quantity that allows version (b) to generate negative implicatures about Miss X's vocal stylings.

But maybe that example should actually be seen as a more polite way of saying "Her singing was horrible"?

In the end it seems that Jessica Wildfire is taking "polite" to mean "consistent with the norms of conversational interaction", from the perspective of someone who finds them opaque:

On the flip side, telling people exactly what you want is considered rude. Some of us fall into this trap all the time. If you visit my home, you’re supposed to keep saying, “I’m thirsty” and wait for me to offer you water. You can’t just ask me. And you definitely can’t help yourself to my faucet. Don’t be rude. Just pass out from dehydration.

This rule also sounds like bullshit. If you ever visit my house, just ask me for water. Or bourbon. I have both. I’m willing to share.

Most people learn these rules naturally. People like me, on the spectrum, have to learn them explicitly.

And maybe there really is a sort of positivity bias in conversational implicatures, such that "speaker's meaning" is generally in some sense more "polite" than a literal expression of the same content would be. This reminds me that in "Reverse sarcasm?" (10/27/2003) I discussed a perhaps-analogous sort of positivity bias in the case of sarcasm.



  1. David Cameron Staples said,

    July 31, 2019 @ 7:45 am

    "In the end it seems that Jessica Wildfire is taking "polite" to mean "consistent with the norms of conversational interaction", from the perspective of someone who finds them opaque"

    As another one of those people who finds them opaque, I fail to see what other definition "polite" could have.

    What is "polite" is exquisitely sensitive to time, place, context, social class, thousands of years of accumulated historical accident and the whims of changing fashion.

    It's all basically arbitrary, and as far as I've made historical study of it, it would seem that that's the point. You soak up what is polite in your own culture, but because it's so random, it's really hard to learn somebody else's to any degree of fluency. (So to that end it's like language, only without the redeeming logical sense of grammar.)

    Do you take your shoes off before entering a house? Do you offer someone a drink? Food? A meal? Do they ritually refuse your hospitality before accepting it or is it an insult to even suggest such a thing? Do you finish everything on the plate, or leave a morsel?

    This is why diplomats need so much training.

    [(myl) The trouble is that what is "impolite" is subject to the same kinds of variation. You need a combination of cultural and personal knowledge and communicative reasoning to insult or offend someone effectively. So is it "polite" to be effectively offensive?]

  2. Jerry Friedman said,

    July 31, 2019 @ 8:53 am

    Diplomats need a lot of training in foreign languages too, and maybe there are similarities in arbitrariness and complication between the rules of politeness and those of grammar. Jessica Wildfire has obviously mastered the grammar of her native language, probably in the usual way, but says she couldn't learn the rules of politeness of her native society in the same way. Could being "on the spectrum" be involved here, and if so, how? Maybe by her not noticing people's reactions when they considered her and other people's behavior rude? If neurotypical people can understand why some impoliteness is offensive but Wildfire couldn't, the rules might not be so arbitrary after all.

  3. Circeus said,

    July 31, 2019 @ 11:11 am

    Jerry Friedman, as someone "on the spectrum", myself (AKA Autism spectrum/asperger's), the sentence " Could being "on the spectrum" be involved here, and if so, how?" is about the most condescending and uninformed thing I've read on the subject in a while coming from a neurotypical.

  4. J.W. Brewer said,

    July 31, 2019 @ 12:02 pm

    For many people who study linguistics it is enjoyable to learn how to explicitly see, analyze, and discuss the tacit rules (both of grammar and more broadly of conversational interaction) of our native language that we have in many instances been faithfully yet unconsciously following since childhood without ever having been explicitly taught. (If you don't find that experience actively enjoyable, there's nothing wrong with you but you are probably going to major in something that isn't linguistics!) That someone who for whatever reasons (could be neurological, could be having grown up in a subculture marginal to the larger speech community) failed to pick up some of those tacit rules could derive a whole different set of benefits from that — i.e. being able to say aha that's a tacit rule I apparently didn't acquire the way most people did and now I understand why my failure to follow that tacit rule I hadn't acquired has caused people to react to me oddly from time to time — is an extra benefit I had maybe not considered very much before, and really rather moving. Obviously it would ideal to help people in those circumstances learn explicitly what others have already picked up less explicitly at an earlier point in their lives than a college sociolinguistics class. (Some people may be able to internalize and act on the tacit rules once it's explained to them explicitly; others may find that more difficult. That's a separate set of issues.) And it would also be good for more people in a varied and pluralist society to be aware that sometimes the tacit rules that govern these interactions are not shared quite as universally as a simple model would assume, and that when people talk in ways that seem "off" (could be seemingly-impolite, could be just weird in some other dimension) it may be helpful to construe things charitably, consider the possibility that there is a failure to hold some relevant tacit rule in common, and try to adjust.

    Characterizing the norms as "bullshit" however, just seems a category error. They are arbitrary, in the sense that they could be otherwise and quite possibly in some other society are otherwise. But you can't have a grammar, much less broader sociolinguistic patterns of communication, without a bunch of shared conventions that are always going to be arbitrary in the sense that they could be otherwise. Language is just like that. And indeed sometimes early exposure to one set of conventions blocks your ability to adjust to others, as when it becomes for many people practically impossible post-childhood to ever really acquire a "native-speaker" accent in a language with phonology very different from their L1.

    If a particular set of conventions is easier for some people to follow than others, there's unlikely to be some magical objective solution that overcomes that obstacle, other than, as I suggested above, maximum charity by all concerned and not viewing genuine inability to comply as a moral failing.

  5. Gregory Kusnick said,

    July 31, 2019 @ 12:59 pm

    It doesn't seem quite right to me to say that "when someone you barely know says, 'How are you?' they actually don't give a shit." If, in that situation, you were to answer honestly that "I'm frankly kind of down today because my dog died yesterday" or something similar, what you'd elicit would most likely be genuine sympathy, along with an awkward sense that perhaps you're sharing a bit more than is entirely appropriate to that social situation. That's not quite the same as not giving a shit.

    [(myl) There are many other cues, in a situation of that kind, about how much the interlocutor really wants to know, or how they're reacting to one kind of answer or another as the encounter develops. So the problem is not so much knowing what "How are you?" means in the abstract, but rather figuring out what's going on in a particular interaction.

    Someone who gets that wrong — and I think all of us do, some of the time — is much more likely to get negative feedback for responding with too much information than with too little. Which might be where the idea comes from that real answers to such questions are impolite.]

  6. Bloix said,

    July 31, 2019 @ 1:54 pm

    I use Gricean concepts when I prep witnesses for depositions. I explain that the exercise feels like a conversation, but its not one. It's the creation of a record. It will never be read as a whole – it will exist as an archive. Statements will be extracted without reference to what came before or what might come after.
    And then I work with the witness on how to answer questions specifically and completely, without reference to a cooperative principle.

  7. Roscoe said,

    July 31, 2019 @ 2:32 pm

    The "ask me for water" example brings to mind an old Jewish joke:

  8. Jerry Friedman said,

    July 31, 2019 @ 3:22 pm

    Circaeus: I apologize. Since the question was uninformed, can you suggest some place where I can get reliable information?

  9. David Cameron Staples said,

    July 31, 2019 @ 5:48 pm

    "The trouble is that what is "impolite" is subject to the same kinds of variation."

    If "polite" means "consistent with the norms of conversational interaction", then surely "impolite" means inconsistent with …", and, yes, is just as arbitrary. Not taking your shoes off is impolite. Leaving your chopsticks sticking into the rice is impolite.

    And it's context dependent and subject to interpretation. Give an OK sign to this person, it's a sign of approval. To that person, it's a strong insult. To that person, it's a signal that you're (secretly wink-wink) a white supremacist.

    And it's not just in conversation in the strict sense, but in the wider sense of any interaction between people.

    Yes, linguists learn languages, and people can learn politeness systems. But just like someone in a monolingual culture doesn't tend to think about their language much, it seems that most people don't analyse their own politeness systems. It isn't until you're immersed in an alien one that you look at your own. And sometimes not even then.

    At one level, English speakers don't think about man/men vs book/books vs house vs houses, or what the hell happened to "-ough", or why is/was/were/be/are. When you come in from the outside, or study it a bit, then the answer is "it doesn't have to make sense, that's just how it is, memorise all the contradictory rules, and everyone will judge you for getting any of them wrong. When you've studied it a lot, then the etymology of "to be" is explained, the various plural forms makes some sense, and -oȝ is obvious behind -ough. But that knowing why doesn't always help in fluency in the current language. Knowing why one shakes hands in a historical and social sense doesn't help you know how tightly or how long to grip. Knowing that it's polite to offer a drink and polite for them to refuse even if they want one, so it's polite to offer a second time, doesn't help you remember when someone's actually in your house.

    People on the Autism spectrum do have our own native politeness systems… it's just that it's orthogonal to that of the neurotypical folk around us. We tend to value honesty, accuracy, and directness. Most politeness systems seem to value lying, vague wooliness, and interminable abstruse handshaking protocols which don't seem to have any function other than to be performed.

    So when an autist is asked "Does this dress make me look fat", our first instinct is to say "actually, yes it does." In our native politeness system, honesty and truth is the polite thing: I wouldn't want you to waste your time on something that doesn't flatter you. This is, apparently, the wrong answer. When someone asks me "Hi, how are you?" at the coffee shop, then, yes, if I say "not very well, actually, my father is ill and my cat just died", then the response will be sympathy, but it will be strained and confused. That's not how it's supposed to go. You're supposed to say "fine, thanks, you?" and order your coffee. So, basically, the polite thing to do is to lie in response to what is in outward form a direct question. I get around the cognitive dissonance by half-joking "undercaffeinated". If you want someone to do something, an Autist would find it polite to make a direct clear request: "Could you please do the dishes?" whereas for most people the polite thing is to hint and waffle "the dishes need to be done" (I'm sure they do. And?) or "did you want to clean up" (not really, no. Why, did you want me to?), to the extent that we may not even realise that we've been asked or told to do something. To that extent, the politeness system we grow up in is an alien one. And we get that it has a purpose, it's just hard to get past the first-order opaque arbitrarity of it all. This lie is polite. That lie is impolite. Learn the difference or get punished. (And, because of our inherent difficulty figuring it all out, we tend to spend our childhoods getting punished for things we don't understand, and never adequately get explained.)

    And it doesn't make sense, no matter how natural it feels to you to do it. A question on QI once made this point where the wrong answer to "How do you do?" was "Very well, thank you". The correct answer was, in its time, "How do *you* do?" It's a formula, and the meaning is in the performance of it, not its literal semantic content. Autists see the literal form, and have difficulty with the ritualised aspect of it, and this difficulty is usually read as impoliteness, or outright rudeness (where, if we're going to play with definitions, you might say that impoliteness is "I didn't know", where rudeness is "I know, but don't care", passive vs active inconformity with conventions).

    And as for the "everyone gets that wrong" … Everyone has parsing errors in speech, no matter how fluent they are. That's a qualitatively different thing from a comprehension disorder, or hearing difficulty. Everyone makes mistakes in speech (not just different forms for written or spoken, or different levels, or a variant but correct grammar, but things which are just plain wrong). That's a different thing from aphasia or simple lack of fluency. Most people can't run a mile in ten minutes, but that's not the same as "everyone's a little bit paraplegic".

  10. Gregory Kusnick said,

    July 31, 2019 @ 8:33 pm

    David: I think it's an oversimplification to say "the polite thing to do is to lie". I value honesty just as you do, and I don't like lying to people if I don't have to.

    But in many cases, to answer with unrestrained candor ("my cat just died") risks dragging someone willy-nilly into an emotional space where they may not be prepared to go. So in that sense the rule against oversharing isn't arbitrary; it emerges from a desire to avoid doing needless harm by burdening strangers with your troubles unnecessarily.

    In that sort of situation my inclination is to answer "How are you?" with something along the lines of "OK, all things considered". This feels honest to my own emotional state (I'm not great, but I'm together enough to go out in public) and leaves my interlocutor an opening to follow up with further questions if they care to. If not, no harm done.

  11. peterv said,

    August 1, 2019 @ 12:45 am

    David Cameron Staples said:

    “People on the Autism spectrum do have our own native politeness systems… it's just that it's orthogonal to that of the neurotypical folk around us. We tend to value honesty, accuracy, and directness. Most politeness systems seem to value lying, vague wooliness, and interminable abstruse handshaking protocols which don't seem to have any function other than to be performed.”

    These descriptions are tendentious. It would be just as accurate to say that most politeness systems seem to value indirection, ambiguity, nuance and subtlety. All of these properties can be desirable from both an individual and a group perspective, which is no doubt why politeness systems are widespread and why they persist. Politeness systems value these features because directness and precision in conversations with people we do not know well may cause unintended offence. Handshaking protocols, like much of speech, serve a phatic function, helping to establish or to maintain good relations between strangers or near strangers.

  12. AntC said,

    August 1, 2019 @ 1:25 am

    @Circeus get over yourself. If you are "on the spectrum" (and I work in an industry which seems to pride itself on employing many who are), then it's likely you're causing all sorts of conversational/contextual disfluencies. I'm not attributing that to malice or lack of caring on your part. Shall I be condescending enough to say 'you can't help yourself'?

    Then how dare you be affronted when somebody else causes such disfluency for you. Or are you claiming not to be uninformed on @Jerry F's personal situation? This being the largely-anonymous internet, we can't all go tiptoeing around the gamut of possible sensitivities posters/readers might bring. (I daresay there's some implicature to the effect if we all kept to saying things that couldn't possibly upset anybody, we'd be unable to say anything. Even "How are you?" for fear somebody's cat has just died, and they burst into tears — as happened to me.)

    If you expect others to make allowances for you, then you can darn well make allowances for others. That's a behaviour you can learn if you can't intuit it.

  13. loonquawl said,

    August 1, 2019 @ 1:36 am

    @Circeus: honestly asking: Wildfire herself hinted (in the last part of the last quote) that the reason behind her social-learning disablity could be her 'being on the spectrum' – why does the post by Jerry Friedman irk you?

  14. peterv said,

    August 1, 2019 @ 3:02 am

    “ interminable abstruse handshaking protocols which don't seem to have any function other than to be performed."

    Of course, in addition to the phatic function, another function of handshake protocols, both in human and in machine systems, is to prevent interactions with entities who do not know the protocol. The fact that the rules of these protocols and other conversational practices may be hard to discern or learn is, from a societal point of view, a feature, not a bug.

  15. rcalmy said,

    August 1, 2019 @ 7:32 am

    Maybe I don't get the rules of politeness myself, but the last quoted passage just sounded strange to me. I would find a guest's unprompted statement of "I'm thirsty" more than a bit off-putting, while an unprompted "Could I have a glass of water?" sounds entirely reasonable.

  16. Jerry Friedman said,

    August 1, 2019 @ 7:38 am

    Circeus: I owe you another apology for getting your name wrong. I'm sorry about that.

    David Cameron Staples: I read your comments with great interest, thanks.

  17. Polyspaston said,

    August 1, 2019 @ 4:33 pm

    "It would be just as accurate to say that most politeness systems seem to value indirection, ambiguity, nuance and subtlety. "

    I don't think it would. In the Netherlands, for instance, "indirection, ambiguity, nuance and subtlety" are often taken to be dishonest and time-wasting, and therefore very rude, and excessive displays of courtesy (e.g., saying 'thank you' and 'sorry' a lot) to be marks of insincerity and sucking up. Obviously, these aren't absolutes – there is still such a thing as politeness, not oversharing, etc., but the contrast with, for instance, British notions of what is polite, is noticeable.

  18. DWalker07 said,

    August 2, 2019 @ 1:49 pm

    Jessica Wildfire didn't just "hint" that she was "on the spectrum". She said:

    "Most people learn these rules naturally. People like me, on the spectrum, have to learn them explicitly."

  19. Martha said,

    August 2, 2019 @ 5:37 pm

    rcalmy, I had the same reaction. I was going to say it was a poor example, but now that I think about it, it's a good example of her lack of understanding of "the rules."

  20. peterv said,

    August 4, 2019 @ 6:14 am

    Of course, “interminable abstruse handshaking protocols”, being interminable, would be of little value for handshakes. Infinite protocols for other purposes are studied and applied in computer systems, for example for operations that must remain ever-on, such as control systems for power and telecoms networks.

  21. Trogluddite said,

    August 8, 2019 @ 11:41 am

    peterv: "[David Cameron Staples'] descriptions are tendentious."

    While I don't disagree at all with your analysis, as an autistic person myself, I would not label the original remarks as tendentious. They are a pretty accurate rendition of how politeness rules _seem_ to many autistic people, and I imagine that David Cameron Staples included those "seem"s quite deliberately. As stated in the post in question; "…we tend to spend our childhoods getting punished for things we don't understand, and never adequately get explained." In the absence of such explanations, our personal impression of the rules and their social intent is all that we have to work with. Attempting to infer them purely by trial and error, without cognition of the multitude of subtle variations in social context which modify them, is very likely to make them _seem_ much as D C Staples described them. If our impressions of them are thus misguided, it is largely due to the absence of guidance.

    Your explanation is of precisely the kind which many autistic people would wish were easier to come by in our everyday interactions, but most often we are simply berated and/or ignored, and expected to go on our way with no clear explanation of exactly what we did wrong – and thus liable to repeat our faux pas. Throughout my life, it has been implied on many, many occasions that if I'm "too stupid to have worked it out for myself by now" then I am somehow undeserving of the knowledge which might help me to improve my social skills. And, rather ironically, I suspect that among more sympathetic people, their own politeness is a barrier to such clarification – people's desire not to seem condescending or patronising preventing them from offering the simple advice which might be so beneficial. Of course, it doesn't help that, when asked for friendly advice, most people seem to suddenly realise that they apply the rules and conventions so sub-consciously that explaining them explicitly is not so easy as they may have initially thought!

RSS feed for comments on this post