Implicature. Pragmatics. So far in class we’ve been concerned with literal meaning . But people mean more things when they use words than just what those words literally mean. Presupposition. “Who stole the money?” Presupposes someone stole the money. “Michael’s brother won the race.”
So far in class we’ve been concerned with literal meaning.
But people mean more things when they use words than just what those words literally mean.
“Who stole the money?”
“Michael’s brother won the race.”
“Michael stopped using heroin.”
Presupposes Michael used to use heroin.
Grice introduces a new word ‘implicate’ to describe a certain phenomenon.
S1: “How is X doing in his new job at the bank?” S2: “He’s doing well, he likes his colleagues, and he hasn’t been to prison yet.”
S2 implicates that S1 is troublesome and liable to steal from the bank he works for (or something like that).
[I write on your application to graduate school]: “She has very good handwriting.”
This is a phenomenon often called “damning by faint praise.” I implicate that you’re not a good philosopher, because although I praise you in the letter, I don’t praise you high enough, or on your relevant abilities.
Implicature is something that a speaker does, not something that a sentence does.
What a speaker implicates is different from what s/he says.
Implicatures are also not what the hearer learns, beyond the literal meaning, from what the speaker says.
Suppose I say: “Stop walking so slowly! Get out of my way!”
You may learn that I am a very disagreeable person. But I am not implicating that, because I am not attempting to get you to believe that I am disagreeable.
Grice’s investigation is going to be to find out how speakers implicate what they do.
That’s what we’re going to do too.
The literal meaning of a sentence is along the lines of its normal, dictionary-definition meaning.
“I said she had good handwriting. I didn’t literally say that she was a bad philosopher.”
Grice says that what a speaker says is closely related to what the literal meaning of the sentences the speaker utters.
However, what a speaker says is not equivalent to the literal meaning of her utterance. We also must take into account the contributions of:
(a) Resolutions of anaphora
(b) The context of the utterance
(c) Resolutions of ambiguity
(a) We must resolve the anaphoric reference: who does the speaker mean by ‘he’?
(b) We must determine the context of the utterance: if he “is” in the grip of a vice, what time was the present time when the speaker spoke?
(c) We must resolve the ambiguity: does ‘in the grip of a vice’ mean here that he is caught in a certain kind of tool, or that he can’t rid himself of a bad character trait?
Grice argues that not all conventional meaning is literal meaning, and part of what is said.
For example, if I say the sentence: “He is an Englishman; he is, therefore, brave,” I have implicated that all Englishmen are brave by using the word ‘therefore.’
An implicature is said to be cancelable if you can deny the implicature right after saying something that seems to implicate it.
[Suppose again you’re applying to be a professor of philosophy and I write on your recommendation:] “She has good handwriting—and in addition, she’s a great philosopher.”
An implicature is detachable if you can rephrase what you just said in such a way that the new sentence has the same literal meaning, but doesn’t have the implicature. For instance, in the handwriting case, the implicature is NOT detachable:
“She has good handwriting”
“Her handwriting is good”
“I’m impressed by her handwriting” etc.
We can identify conventional implicatures with ones that are detachable and non-cancelable.
“Even Ken knows that’s stupid”
Implicates: Ken is the least likely person (among some relevant group of people) to know that the action in question is stupid.
You can’t cancel the implicature:
??“Even Ken knows that’s stupid, but it’s not unusual or surprising that he does.”
But you can detach it:
“Ken knows that’s stupid too.”
Although Grice believed in conventional implicatures, many linguists and philosophers don’t.
Locus classicus: Kent Bach, “The Myth of Conventional Implicature”.
Conventional implicatures have been recently revived by Chris Potts, but most people think Potts is talking about something else.
“Those fucking kids won’t stay off my lawn.”
Implicature: I have a negative attitude toward those kids.
Non-restrictive relative clauses:
“The pizza delivery boy, who was wearing a necktie, thanked me for the tip.”Implicature: he wore a necktie.
In conversation, we don’t merely make random or disconnected remarks. Conversations have purposes: we engage in them for reasons.
The purpose of a conversation can be introduced by a question or set of questions; it can also evolve as the conversation progresses.
At each point in the conversation, certain “moves” (assertions, questions, etc.) will be “unsuitable”—that is, at odds with the purpose of the conversation.
“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.”
The Cooperative Principle, according to Grice, gives rise to four categories of maxims (rules), that must be obeyed if conversation is to proceed cooperatively: the categories
Maxim 1: “Make your contribution as informative as is required (for the current purposes of the exchange).”
Maxim 2: “Do not make your contribution more than is required.”
Supermaxim (includes the others): “Try to make your contribution one that is true.”
Maxim 1: “Do not say what you believe to be false.”
Maxim 2: “Do not say that for which you lack adequate evidence.”
Maxim: “Be relevant.”
Difficulties with elaborating on the maxim:
Supermaxim: “Be perspicuous.”
Maxim 1: “Avoid ambiguity.”
Maxim 2: “Avoid obscurity of expression.”
Maxim 3: “Be brief.”
Maxim 4: “Be orderly.”
Grice admits there are other maxims, such as “be polite,” that guide us in conversations.
But he believes these maxims are not intimately related to the purposes of rational communication in the way that the maxims in the categories of quantity, quality, relation, and manner are.
The maxims are not moral recommendations. Grice is not telling you that you have to be truthful, or that you ought not to be obscure.
The maxims are a description of how we assume other people are behaving in cooperative speech. Why that assumption is warranted is a question Grice will try to answer.
“The maxims presuppose an almost Utopian level of gentlemanly conduct on the part of a speaker and an old-fashioned standard of truthfulness that George Washington might have found irksome. They remind one of the early Puritanism of the Royal Society…”
“A speaker should give not too much but just enough information, hold his tongue about what he believes to be false, or for which he has insufficient evidence, be relevant, be brief and orderly, avoid obscurity of expressions and ambiguity.”
“. . . Would we want to have dinner with such a person, such an impeccably polite maxim observer?”
Grice believes that the Cooperative Principle and the categories of maxims he outlines, extend to other cooperative human endeavors.
Imagine that you and I are making a cake. There are some rules we should obey…
(Quantity) Make your contribution neither more nor less than is required. If I need a cup of sugar, don’t hand me half a cup; and don’t hand me 10 cups either.
(Quality) Make your contribution “genuine and not spurious.” If I need a cup of sugar, don’t hand me a cup of salt.
Make your contribution appropriate to the immediate needs at each stage of the cooperative endeavor.
If I need a cup of sugar, don’t hand me an oven mitt (though do hand me it later when I need it); and don’t hand me an interesting book (though do when we’re preparing for a long plane flight).
Make your contribution perspicuous.
If I need you to grease a baking sheet before I can proceed, don’t do it in secret so that I don’t know whether I can proceed; nor should you take 2 hours to do it.
What is the basis on which we can expect other speakers to be cooperative?
This question is important to Grice, because most implicatures depend on the assumption that the other participants in the conversation are cooperative.
People are simply in the habit of being cooperative, and it is a difficult habit to break.
For instance, Grice points out that it’s easier to tell the truth than to lie.
Grice isn’t satisfied with this solution because he thinks people must have a reason to be cooperative that transcends the fact that it’s habitual.
He doesn’t want why we in fact are cooperative. Instead he wants to explain why it is reasonable for us to be cooperative-- why we should not be uncooperative
Entering into a conversation is like entering into a quasi-contractual agreement, where all parties agree to aid one another for a common goal over a short period of time.
First, in a quarrel, for instance, neither party seems to have some ‘contractual obligation’ to continue the quarrel until its aims are met satisfactorily.
Second, if one participant in the conversation is obscure or ambiguous, it does not seem that he has let down others (not fulfilled his half of the contract) by not cooperating; he has rather let down himself.
Grice suggests that being cooperative (in the way he outlines) may be a precondition for the individuals engaged in conversation to achieve the goals of the conversation (e.g. “giving and receiving information, influencing and being influenced by others”).
He does not spell out this suggestion in further detail. This is because, he says, a large part of demonstrating that this is so will rest on determining what the nature of relevance is, and when and how it is required.
#1 Someone can simply violate a maxim. She can say something underinformative, something she believes to be false, something irrelevant, or something ambiguous.
“I’m sorry, I can’t give you that information” (opting out of Quantity)
“Here’s what I think, but I admit I’ve got no evidence for it.” (opting out of Quantity)
“On a completely unrelated matter…” (opting out of Relation)
When one filibusters, one has typically opted out of Manner.
Suppose you are asked “How many children does John have?” You know that he has more than one, but not exactly how many he has. You could say:
(ii) “More than one” (fulfilling Quality, but not Quantity)
(iii) “Exactly three” (fulfilling Quantity, but not Quality)
To flout a maxim is to blatantly fail to fulfill it.
(a) Lying is typically a case of violating Quality (it’s not obvious one is lying)
(b) Overstatement is typically a case of flouting Quality (it is obvious that what one is saying is false)
Flouting, Grice will argue, gives rise to conversational implicatures. This he calls ‘exploiting’ maxims.
Speaker S, in saying that p, implicates that q, if and only if:
Hearer H reasons: “S said that p; unless q were true, S would be uncooperative in saying that p; but S is cooperative; therefore he is implicating that q.”
S1: “I’m out of gas”
S2: “There’s a gas station around the corner.”
All S2 literally said was that there was a gas station around the corner. This is consistent with the gas station being closed.
But, if S2 believes the gas station to be closed, his utterance would violate the cooperative principle (it would be irrelevant, in the context, because the purpose of the conversation is to resolve S1’s gas worries).
S1 assumes that S2 is being cooperative (why else has he stopped to help?) and thus concludes that S2 must believe, and intend to get S1 to believe, that the gas station around the corner is open.
S1: “How many children does Sally have?”
What S2 literally said, in context, was that Sally had two children. This is consistent with Sally having exactly five children, because if you have five children then you have two. (Compare: “Do you have two dollars?” does not mean: “Do you have neither more nor less than two dollars?”)
But, if S2 believed Sally to have more than two children, his utterance would violate the cooperative principle (it would not give all of the information requested, thus violating the first maxim of quantity).
S1 assumes that S2 is being cooperative (why else has he bothered to answer?) and thus concludes that S2 must believe, and intend to get S1 to believe, that Sally has no more than two children.
S1: “Do you have 2 dollars?”
S2: “Yes, I have 100 dollars.”
S1: “Did many of the students pass the exam?”
S2: “yes, all of the students passed.”
S1: “Did you have an OK summer?”
S2: “Yes, I had an excellent summer.”
“John has 2 kids” → “John doesn’t have 3 kids or more”
“Some students failed the exam” → “Not all the students failed the exam”
“The mail usually comes on time” → “The mail doesn’t always come on time”
“It’s possible that 7/11 has champagne” → “It’s not likely that you can get champagne at 7/11”
“You’re permitted to leave” → “You’re not required to leave, you can stay”
“You can have tea or cake” → “You can’t have tea and cake”
These implicatures all follow from the Quantity maxims. Since speakers are required to say as much as they can (that’s true and relevant), not saying the high-up thing (choosing something low) implicates that the high-up thing is false.
In the standard cases, speakers don’t fail to fulfill the maxims. Implicatures are generated because the assumption that speakers are cooperative requires other things to be true, besides what the speaker says.
In cases of flouting maxims, speakers do fail to fulfill them, and implicatures are generated in a different way.
Suppose you’re selecting candidates for a philosophy job. For candidate X, you have received the following letter of recommendation:
“Dear Sir, Mr. X’s command of English is excellent, and his attendance in classes has been regular. Yours truly, Professor so-and-so.”
You reason first that the letter writer is flouting the first maxim of quantity. He has taken the time to write the letter, so he cannot be opting out of the conversation; additionally, he must have more information, because the person for whom he is writing the letter is his student.
Thus, the writer must be attempting to convey more information than he has literally said. But why would he not come out and say it? Perhaps because what he has to say about the student is negative.
The form of a flouting inference is something like this:
“The speaker is openly failing to fulfill such-and-such maxim, though I have strong reason to think the speaker wants the purpose of the conversation to succeed. The only good reason to be uncooperative, given that, would be if q. So the speaker must want me to believe q.”
One typically flouts the maxims of quality when one engages in irony, metaphor, understatement, and hyperbole.
For example: The speaker said “Dick Cheney is a monster”; but he can’t really believe that Cheney is a monster, because monsters don’t exist; he must really mean that Cheney is a terrible person.
A case of flouting the maxim of relation:
[Location: a prim and proper tea party]
S1: “Ms. X is an old bag”
S2: “My, it’s such lovely weather we’ve been having of late!”
S2’s utterance is clearly not relevant to the topic of conversation introduced by S1.
The most plausible explanation is that S2 believes that it is improper to be discussing Ms. X in the way that S1 wants to.
So S2 has implicated, by exploiting relation, that this is an improper topic of conversation.
A case of flouting: be brief!: a reviewer for a stage performance says “Ms. X produced a series of sounds resembling the lyrics to ‘Home Sweet Home.’”
The reviewer could have said something much briefer, namely: “Ms. X sang ‘Home Sweet Home.’” There must be some explanation for his lack of brevity; most likely, he believes that Ms. X did not do a good job of singing the song; thus, this is what he implicates.
An implicature is particularized if it only arises in a special context. For example:
S1: Is Sue pretty?
S2: She has a wonderful personality.
S1: Oh, so she’s not pretty.
Here, B implicates that Sue is not pretty, because he’s not observing Quantity and Relation: he was required to give information about Sue’s attractiveness, but he instead gave information about her personality.
S1: I’m looking for someone to go on a date with.
S2: Take Sue, she has a wonderful personality.
Here, S2 doesn’t implicate that Sue is not pretty, because his utterance doesn’t violate/ flout or otherwise exploit any of the maxims. It’s true, relevant, orderly, and a sufficient reason to go out on a date with someone (that is, it satisfies Quantity).
An implicature is generalized if only in special circumstances does it not occur.
S1: I’m looking for a woman to go on a date with.
S2: Take Sue, she’s pretty.
S1: Oh, so she’s not beautiful.
Since S2 is contributing information on Sue’s attractiveness, he should state the greatest degree of attractiveness that she has. Since he didn’t say ‘beautiful’ which is a greater degree than merely ‘pretty’, she must be pretty but not beautiful.
In special circumstances, however, this implicature is not present:
A: No one here is pretty.
B: Sue is pretty!
Here, the point of B’s utterance is to give evidence against A’s claim. Thus, he satisfies the quantity requirement by saying that Sue is pretty, and this is compatible with him also thinking that she’s more than just pretty, she’s beautiful.