Charitable speech

Today’s xkcd:

The late philosopher H. Paul Grice tried to make some of these rules explicit; his most famous attempt to do this can be found in his essay “Logic and Conversation,” published in his book Studies in the Way of Words (Harvard U.P., 1989 and 1991, pages 22-40.)  Grice there lays down a set of rules in the form of a series of maxims.  Grice begins with an overarching maxim that he calls the “cooperative principle”: “Make your contribution such as it is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged.”  He breaks the requirements of the cooperative principle down into several maxims.  So, one ought to be truthful, one ought to provide the listener with enough information to make it easy for the listener gather one’s meaning, one ought not to provide the listener with so much information that it is difficult to gather one’s meaning, one ought to provide information that is relevant to the conversation, one ought to express oneself clearly.

As Carole J Lee points out in her article “Gricean Charity: The Gricean Turn in Psychology” (Philosophy of the Social Sciences, 2006; 36(2), 193-218,) what Grice has given us in these maxims are not “universal norms of conversation” (203.)  Still less, despite their phrasing, are Grice’s maxims commands that he would have insisted we follow.  Rather, they are a sketch of the expectations that listeners tend to bring to a conversation.  One tends to expect that a speaker will behave according to the maxims.  When a listener finds that a speaker is systematically violating one or more of the maxims, that listener might react with laughter.  That laughter shows that the listener, relying on the maxims, had constructed a different meaning than the one the speaker intended.  Or, the listener might react with distrust or frustration, if reliance on the maxims has led him or her to a dead end.  A charitable listener will bring these expectations to a conversation, assuming that each speaker is displaying the competence the maxims outline.  To refuse to give a speaker credit for following Grice’s maxims is to fail to show that speaker the charity that makes cooperative communication possible.

In the strip, the hairless stick figure is frustrated by the hairy figure’s decision to “interpret an obviously sympathetic ‘I’m sorry’ as an apology.”  Hairless seems to be frustrated that Hairy is not giving him* credit for following Grice’s maxims.  The maxim of quantity requires that each speaker provide just that information the listener needs; the “Why?” in “Why? It’s not your fault” suggests that he is leaving out something essential, something which his sarcastic reply might supply.  The maxim of relevance requires that each contribution to the conversation bear on the topic at hand; that Hairy would react to a conventional expression of sympathy as she does would suggest that she does not regard Hairless’ sympathies as relevant to a the topic of her mother’s misfortune.  This rejection of Hairless’ expression of sympathy might well strike anyone as rather harsh.

Grice acknowledges that all of the properties that his maxims enjoin are complex.  For example, he says of the maxim of relevance: “Though the maxim itself is terse, its formulation conceals a number of problems that exercise me a good deal: questions about what different kinds and focuses of relevance there may be, how these shift in the course of a talk exchange, how to allow for the fact that subjects of conversations are legitimately changed, and so on.”  Of course, different cultures conceive of relevance in vastly different ways, a fact that serves as the starting point for many studies which question the usefulness of Grice’s maxims as a guide to the analysis of conversational behavior around the world.**

To the extent that Grice’s maxims are an attempt to outline the expectations listeners have of speakers, it is unsurprising that they do not lay injunctions upon listeners.  If we were to complement them with a set of maxims that outline what speakers might reasonably expect from listeners, perhaps the first maxim would call for charity.  That charity would be as multidimensional, and no doubt as culturally specific, as the properties Grice’s listeners expect from speakers.  We might break our maxim of charity into several sub-maxims to give an idea of what the dimensions of this charity might be.  First, assume that the speaker is performing his or her communicative project competently unless s/he provides evidence to the contrary.  One might refer to Grice’s maxims in defining competent communicative performance.  Second, assume that if the speaker has failed to perform his or her communicative project competently, any failures are as small as possible.  So, if a speaker provides a logically invalid argument, a charitable listener might look for the simplest available premise that can be added to make the argument valid.  Third, assume that the speaker is a person of goodwill.  Under this assumption, one should try to find the least obnoxious possible interpretation for any unclear passages.

In the strip, Hairy has been uncharitable to Hairless, albeit in a rather subtle way.  Hairless’ sarcastic response is of course a grossly uncharitable one.  Grice was interested in sarcasm, which he saw as something that happens when when “the maxims are flouted.”  What Hairy flouts in a subtle way is what Hairless responds by flouting in a massive way, the requirements of charity as I have tried to formulate them in the paragraph above.  When speakers think that their listeners are refusing to follow these requirements, they often do respond with anger.

Two examples come to mind.  Recently, I took part in a discussion thread on one of my favorite websites, the mighty Language Log.  One of the mighty Log’s most distinguished authors, linguist Geoffrey Nunberg, posted a little piece about a newspaper columnby a writer named Kathleen Parker.  Parker opens with a story about a brief encounter among a group of about ten strangers sharing an elevator at Presbyterian Hospital in New York City.  Parker was among this group; apparently she had not seen any of the other riders before.  Two riders did know each other, a woman and a man.  The woman spent the moments of the elevator ride shouting at the man.  In the course of her diatribe, the woman called the man a “motherfucker” several times.  By the time this pair left the elevator, the woman had made it clear that the man was her son.  Parker describes the discomfort that she and the other riders displayed while in the sharing the elevator with the shouting woman.  She goes on to allow that coarse words are not in themselves particularly dangerous to the moral health of society, giving as an example of her willingness to tolerate them her relaxed response to a fellow guest at a very proper tea party who called sometime golf champion Tiger Woods an “asshole” in an incongruously “refined accent.”  Still, Parker claims, “Lack of civility in words bleeds into a lack of decency in behavior… An “MF” here or an “FU” there might not constitute the unraveling of society, but each one uttered in another’s involuntary presence is a tiny act of violence against kindness, of which we surely could use more.”

There are any number of things one might say about this column.  It’s interesting that the word “motherfucker” has come to be so frequently used as a symbol of obscenity without any evocation of the meaning of its root words that a mother could unselfconsciously  apply it to her son.  It’s puzzling what Parker means by a “refined accent”; she does say that the tea party guest was British, and evidently she was using what we used to called Received Pronunciation.  But obscenity is now so common a feature of the speech of Britons of all classes that I cannot imagine any vulgar word being incongruous when spoken in any British accent.

What Professor Nunberg in fact chooses to do with the column is to accuse Parker of veiled bigotry.  Noting that Parker’s response to the woman at the tea party had been to say that it would be all right for the woman to curse so long as she used her plummy accent while doing it, Nunberg says: “Now I can come up with at least five contexutal parameters that explain why the public harangues of the elevator expletiviste were obnoxious and offensive in a way that the English (one assumes) woman’s wasn’t. But the posteriority of the initial vowel of the epithet isn’t one of them.”   This is a failure of charity on Professor Nunberg’s part; Parker’s use of the story to show her own relaxed attitude towards a non-threatening use of taboo words makes it clear that she was joking with the woman at the tea party.  Lest we think that Professor Nunberg is himself joking, and merely pretending that he doesn’t see Parker’s joke, he goes on:

Why should it be? Would the incivility of the first woman’s rants have been tempered if she had called her son an asshole in an accent like Emma Thompson’s? Would the tea party lady’s reference to Tiger Woods be more offensive if the woman had sounded like Wanda Sykes? While we’re on the subject of vulgarity — and insolence — can we linger for a moment on the smug suburban gentility of that word refined? You’re left with the unsettling implication that the acceptability of allowing a naughty word to cross one’s lips depends, in part, on how thick they are.

Where does Professor Nunberg find this remarkably ugly implication?  He seems to find it in Parker’s telling of the story of the elevator ride.  Quoting key bits of the story, he comments:

The race and class of the woman and her companion weren’t specified, but readers might have been able to divine those attributes from the particular word Parker chose to report (or was that the only vulgarity the woman used?), helped along by the setting at Broadway and 168th Street and the mentions of the separated father and in particular of the young man’s “baggy drawers,” which presumably were intended to convey some relevant information. (If it had been an upper-middle-class white woman screaming “motherfucker” at a phat-pantsed white preppie, communicative cooperativeness would have obliged Parker to mention that fact lest the reader draw the wrong conclusions.)

None of the markers Professor Nunberg mentions in this paragraph is at all strongly correlated with race and class in New York City in 2011, as commenters on the post pointed out.  Your humble correspondent mentioned a couple of items showing that, while “Motherfucker” may have originated among African Americans in the early twentieth century, it has been in general use among whites and others for decades.  Professor Nunberg himself added to this list, strengthening my point (perhaps inadvertently.)  Another commenter pointed out that Broadway and 168th is in the middle of Spanish Harlem, so that if that location tells us anything about the woman’s ethnicity it would suggest either Puerto Rican or Dominican heritage.  Still others have pointed out that New York Presbyterian has more than one Manhattan location.  One might also mention that Manhattan is rather a compact place, so a wide variety of ethnicities may be found in any of its hospitals.  In the column, it is quite clear that “the “[mention] of the separated father” was “intended convey some relevant information,” and it is also clear what that information is.  It is when the woman in the elevator mentions the man’s father that it becomes clear that she is his mother, and so it is when Parker quotes that line that she provides the punchline about the woman’s use of “motherfucker.”  “Baggy drawers” or “phat pants,” like the use of “motherfucker” as an all-purpose obscenity, likely originated among African Americans and doubtless once was seen almost exclusively in African American communities, but one needn’t spend much time among young American whites to see that those days are long gone.

I appended many comments to this thread.  Here are links to all of them, in the admittedly unlikely case that you are interested: 1, 2, 3, 4, 5, 6, 7, 8.  I’ll copy the last one, because in it I finally got around to saying what I’d been trying to say all along:

@Daniel: 1. “This post doesn’t really need any more comments, but”- Yes, I sympathize. It’s as hard to stop as it is to stop eating a salty snack. An angry, angry salty snack.

2. “”drawers,” which I definitely associate with AAVE.” The structure of this debate seems to be, S1: Parker reports (item.) I associate (item) with African Americans. Therefore, Parker wants her readers to associate the people she is denouncing with African Americans.” In reply, S2: “I don’t associate (item) with African Americans.”

The least involved premises we can add to S1’s assertions to make them into a logically valid argument would surely be “I am the sort of reader Parker had in mind when she wrote her column, and she should have known that I would associate (item) with African Americans.” In the absence of any data about what Parker’s readers think of when they think of these various items or about what she has reason to expect from her readers, we have no reason to suppose that these premises are true.

3. “Rhetorically, that is clearly implying that the British lady did not have context or delivery issues.” Let’s remember the context and delivery Parker describes in her opening vignette. She was on an elevator, which is to say, in a confined space. A person entirely unknown to her joined her and several others in that confined space. For a few moments, this unknown person shouted obscenities. Anyone might feel uncomfortable in that situation.

By contrast, Parker met the British lady (if British and lady she in fact was***) in a situation where she was free to move about. Parker and this second person were introduced and participated in a conversation. In the course of that conversation, the second person used a word that often classified as objectionable. Parker did not object to it. In fact, she made a little joke to forestall objections.

What, then, is the simplest explanation of the difference between Parker’s response to the two situations? Is it that person 1 was of a different ethnicity than person 2, and that Parker is hostile to the ethnic group person 1 represents and friendly towards the ethnic group person 2 represents? Or is it that she would rather engage in a conversation at a social event with someone who clearly poses no threat to her than be trapped in an elevator with an angry stranger?

Granted, Professor Nunberg does make a nod in the original post to  “at least five contexutal parameters that explain why the public harangues of the elevator expletiviste were obnoxious and offensive in a way that the English (one assumes) woman’s wasn’t.”  His remarks before and after that, accusing Parker of “nudge-nudge allusions to race and ethnicity… the way people intimate someone’s Jewishness by saying they’re ‘very New York'” and of implying “that the acceptability of allowing a naughty word to cross one’s lips depends, in part, on how thick they are” does not leave much opportunity to put serious weight on these contextual parameters.

I’ll quote a bit from my second-to-last comment as well, since in it I brought up the points I’m making here about charity.  In particular, I mention that a listener who is charitable to a speaker in one way may have to think ill of that speaker in another way:

@Peggy: “Keith M Ellis: “The whole point of dog-whistle racism is its deniability.” The whole point of accusations of dog-whistle racism is their unfalsifiability.” That’s going a bit far, surely. One ought to be charitable to people with whom one disagrees; one form of charity is to assume that when a logically defective argument can be made valid by the addition of an unspoken premise, the speaker has omitted the simplest possible premise that can achieve that result. So, suppose I ask you “Are the sidewalks wet?” and you reply, “They must be- it’s raining.” You’ve then made an argument that could be presented thus: (Premise) It’s raining. (Conclusion) The sidewalks are wet. By itself, this argument is invalid. The simplest way to make it valid is to add, as a second premise, “If it’s raining, the sidewalks are wet.” It would be an uncharitable listener who refused to make so small a cognitive leap in the course of a conversation.

Likewise, if one were to encounter an invalid argument that could most readily be made logically valid by the addition of a racist idea as an additional premise, it would be charitable to consider the possibility that the person making the argument has simply omitted that premise. I could mention some newspaper columnists who in fact do precisely that on a regular basis.

I’d extend that maxim of charity beyond the logical structure of arguments to the emotional response people exhibit to various stimuli. If, for example, Parker were usually happy when she was required to share a confined space with white people repeatedly shouting obscenities, but unhappy when an African American did the same thing, and she spoke as if no explanation was necessary for this difference in reaction, then charity to Parker would warrant the assumption that she was right, that the reason for the difference was in fact so simple that anyone could find it without an explanation. Charity to such a speaker might lead us to suspect the speaker of race prejudice.

Again, the column under consideration obviously does not meet this description. However, it is far from rare for people to exhibit emotional reactions to stories about misbehavior among African Americans that are not only grossly disproportional to the reactions the same people exhibit when they have heard similar stories about people of other ethnic backgrounds, but which also kick in long before any evidence is presented showing that the stories are even true. When that happens, it is neither irresponsible nor unfalsifiable to claim that racism is at work.

The more I thought about this exchange, the stranger it seemed to me that someone as learned as Professor Nunberg could take Parker to task for seeing a threat in a situation where she was confined in an elevator with a stranger.  Granted, the idea that this represented a threat to civility in general is a bit underargued, but once one describes the situation it is hard to see what need there is to attribute Parker’s anger to racial prejudice.  Yet shortly after, I noticed a news item describing an equally distinguished academic replicating Professor Nunberg’s behavior with some precision.

Evidently there is a woman named Rebeca Watson who writes a blog that is popular among atheists and other irreligious folk.  In June, Watson attended a conference at which her fans were well represented.  Very late on one of the nights of this conference, Watson found herself alone in an elevator with one of these fans, a man who made an awkward pass at her.  Watson declined the man’s offer, and said on her blog that it made her uncomfortable.  She gave it as an example of the wrong way for a man to approach a woman.

That, one would think, would be that.  Were one to think so, one would be reckoning without Richard Dawkins.  Professor Dawkins took it upon himself to post the following, as comment #75 in a thread on P. Z. Myers’ blog:

Dear Muslima

Stop whining, will you. Yes, yes, I know you had your genitals mutilated with a razor blade, and . . . yawn . . . don’t tell me yet again, I know you aren’t allowed to drive a car, and you can’t leave the house without a male relative, and your husband is allowed to beat you, and you’ll be stoned to death if you commit adultery. But stop whining, will you. Think of the suffering your poor American sisters have to put up with.

Only this week I heard of one, she calls herself Skep”chick”, and do you know what happened to her? A man in a hotel elevator invited her back to his room for coffee. I am not exaggerating. He really did. He invited her back to his room for coffee. Of course she said no, and of course he didn’t lay a finger on her, but even so . . .

And you, Muslima, think you have misogyny to complain about! For goodness sake grow up, or at least grow a thicker skin.


First, I would point out that Professor Dawkins has far more demands on his time than I do, so the fact that he would enter such a discussion shows that he uses his time even more foolishly than I do. In fact, even before the rise of the internet, Dawkins used to send unsolicited mail to people containing precisely these sorts of gratuitous insults.  Some years ago, I read a magazine profile of a Harvard graduate who became a spokesman for Creationism.  One day he opened his mail to find a letter from Richard Dawkins, whom he had never met, telling him that he was either an idiot or a villain, more likely a villain.  I can’t recall the man’s name, since I’d never heard of him before or since.  But for some reason Richard Dawkins wanted to send him hate mail.

Second, if Dawkins were as obscure a figure as am I, his remark would have been forgotten the moment comment #83 appeared in the same thread, from “Forbidden Snowflake”:

Dear Richard,

What right have you to bemoan the teaching of creationism in your country while people are dying of malaria in West Africa?
Seriously, this “you have no right to complain about your problems as long as there are bigger problems somewhere in the Universe” is nothing but a silencing tactic.

But of course Professor Dawkins is world-famous.  And he tried to defend his remark, making it much worse, and again, making it still worse.  So, there’s been a great deal of controversy.

Professor Dawkins’ invocation of Muslim women who live in countries where genital mutilation is practiced and severe form of Islamic law obtain might be expected to trigger anger from readers who identify with Rebecca Watson and assume that the professor is sharing information which he assumes to be relevant.  If one were charitable to him, assuming that he was so competent a communicator that the relevance of the information to Watson’s post is likely easy to find, the premise one might supply would be that Watson is self-centered and unconcerned with the sufferings of such women.

Unsurprisingly, Watson’s defenders have responded in kind.  Gawker quotes blogger Jen McCreight:

[It] makes me want to cry a little when you live up to the stereotype of a well-off, 70 year old, white, British, ivory tower academic. But let me spell it out for you instead of just getting mad (though I’ll do that too):

Words matter. You don’t get that because you’ve never been called a cunt, a faggot, a nigger, a kike. You don’t have people constantly explaining that you’re subhuman, or have the intellect of an animal. You don’t have people saying you shouldn’t have rights. You don’t have people constantly sexually harassing you. You don’t live in fear of rape, knowing that one wrong misinterpretation of a couple words could lead down that road.

Gawker quotes Watson’s expansion on McCreight’s remarks, then asking: “Can it really be that Dawkins has never been exposed to insults as odious as the ones mentioned by Ms. McCreight? As a jump-starter of the modern atheist revival, doesn’t Dawkins probably get a lot more threatening hate mail than all of his critics combined?” Considering that he’s a man who lives in the UK, I’d be willing to stake any amount of money on the proposition that Professor Dawkins has been called a “cunt” quite a few times  in his life, and considering the amount of hate mail he gets I suspect that he’s been on the receiving end of the other words and claims frequently as well.  Of course, words that have an association with a specific group wouldn’t have the same force when applied to someone outside that group, so Professor Dawkins is no contender for the title of Most Aggrieved.  McCreight’s choice of words is still unfortunate, however.

At any rate, surely this whole matter calls for charitable reading.  Only by dint of a most uncharitable reading of Watson’s original remarks could Professor Dawkins have thought it was appropriate to come at her in the way he did.  A charitable reading of Dawkins’ words would not leave him looking at all good, but might have kept his critics from responding in anger when pity would have been not only more suited to the quality of his thought in this matter, but also more likely to inspire in him some feeling of embarrassment.

*Hairless stick figures in xkcd often turn out to be male, hairless ones female.  Also, author Randall Munroe uses the first person singular to refer to Hairless in this strip, so I’m assuming it represents him.

**Some early examples of these studies are mentioned in Mary Louise Pratt, “Ideology and Speech Act Theory,” Poetics Today, vol. 7 no 1 (1986,) pages 64-65.

***She in fact was.  I’d forgotten the relevant bit of Parker’s original column and was distracted by a side discussion in the thread about other properties of the tea party guest.

Previous Post


  1. Like everyone on Rebecca Watson’s side you are missing the point.

    For about the previous two month Rebecca had been on a role complaining about sexism in the atheist community. She went to conferences and steered the conversation all the time to how horrible the atheist men were and never gave any real evidence (she cited a few emails, but they were mostly from religious trolls). Time and time again she asserted that atheist conferences were hell for women, like some kind of sexist gulag, never really giving other evidence, just claiming that other unnamed women agreed with her (even when other women disagreed with her about how terrible the atheist dudes were, she simple dismissed them out of hand).

    So people were very hungry for an actual example of the actual “misogyny” Rebecca kept ranting about so they could consider and debate the situation, but when if finally came it was utterly pathetic. Of course she got jumped all over, rightly so.

  2. acilius

     /  September 2, 2011

    Thanks for reading and commenting!

    Thanks also for explaining why, as you put it, many “people were very hungry for an actual example of the actual ‘misogyny’ Rebecca kept ranting about,” and when she told this story saw it as such an example. Certainly, people who felt that she was accusing them of something shameful would be ready to pounce when she gave them a target to pounce upon. But where in the video is there any reason to suppose that she wanted us to see the elevator story in the context of her previous complaints? By putting it at the end of a generally pleasant report about the conference, it would seem that she was going out of her way to make it clear that it was not part of that discussion. And of course there is her relaxed body language and measured choice of words, which P. Z. pointed out in his rebuttal to Dawkins on Pharyngula.* If she’d wanted us to take it as the long-awaited concrete example that was to be the test of her general case against the male contingent of the skeptic movement, surely she would have made some rhetorical move to emphasize it, as opposed to the many rhetorical moves she made to de-emphasize it.

    Moreover, there is the sheer fact that Richard Dawkins posted a nasty comment, directed at a blogger unknown to the general public, in position #75 in a comments thread on Pharyngula. Dawkins is a major public figure and an eminent academic; is that really the most effective means available to him of making his views known to the general public?

    That’s why I disagree with Jen McCreight’s choice of title. “Richard Dawkins, your privilege is showing“. The problem seems to be that a media celebrity and academic eminence like Dawkins is behaving as if he were the stereotypical internet user, typing furiously in his mother’s basement, desperately trying to get someone, anyone, to pay attention to him.

    As for Rebecca Watson providing evidence that there is misogyny among male skeptics, well, look at how this debate has been going. If she’d been devising a Sokal hoax to trick misogynistic skeptics into showing themselves, she couldn’t possibly have hoped to succeed so spectacularly.

    Thanks again for the comment! I hope we’ll see more of you on Los Thunderlads.

    *Obviously you remember this, but there’s a nonzero probability that someone new to the issue might be reading. Not far above zero, perhaps, but not exactly zero.

%d bloggers like this: