Controversial Grice

Kramer in comment on "Implicature in the courtroom":

"My evolving view is that the "conversation" is
between the jury (which is understood to have
asked "Why should we decide in your favor?" and the
lawyer, whose answer is a combination of factual
claims (delivered by extraction from the witness)
and inferences delivered by the lawyer in
opening and summation."

Very good. I take the liberty of take the next comments 'out of context', -- i.e. outside the courtroom cross-examination, and comment. Or something.

Kramer:

"I think that there is a difference between an answer
that is evasive and one that is false in English, even
if the latter is true in Logiclandian."

Good. I would perhaps distinguish:

Legalese --- Logiclandian -- English
(I do this to provoke Kramer. If he thinks he can distinguish between logiclandian and English like that, perhaps someone may argue that Legalese is also a variety that sprang from English, but is not quite quite exactly LIKE English?)

English:
answer (a) evasive
------ (b) false

Logiclandian
answer:(a) evasive
----- (b) false/true.

Legalese
answer: (a) evasive
---- (b) 'technically true' but misleading. 'false'.

Etc.

Kramer:

"Q. Do you like Good and Plenty?
A. I like M&Ms."

where the implicature seems to be: "No". i.e. "I like M&Ms instead".

-----

I would distinguish that from yet another sort of reply:

Q: Do you like Good and Plenty and M&Ms?
B: I like M&Ms.

--- where, again, the implicature seems to be: "I do NOT like M&Ms". Again, implicated.

In the Kramer scenario:

Q: Do you like Good and Plenty?
A: I like M&Ms

How is the implicature, "I do not like Good and Plenty" retrieved? The thing is so easy I will NOT attempt an answer. But note that Grice DOES allow for change of topic in conversation. And for him, 'relevance' is mainly "topical relevance". So, while "No, I do not like M&Ms" is a plausible and possible implicature -- I know that given the right intonation (by a nephew to an uncle) it may mean: "Look: I won't care to answer your rhetorical question, and will instead focus on something which I find of topical relevance -- but in no way infer from what I say that I do NOT like Good and Plenty -- Plus, I know you are asking because I know you know (and you know that I know that) that Good and Plenty are NOT available -- so why admit to me liking Good and Plenty?", etc.

Kramer's second example:

"Q. Did you visit the museum 25 times"
A. I visited it 5 times."

--- this in connection with the District Court, which was admittedly vague: "If U KNOWS that he's been to the store 50 times, to say he's been to the store 5 times is, while technically true, not yet 'clean'".

Q: Did you visit the museum 25 times?
A: I visited it 5 times.

----

Kramer:

"I would draw a distinction between implicatures in the question and implicatures in the answer."

Yes. That's a very good thing to do.

"The question "how many?" can, in some contexts, be
universally understood to mean "exctly" or "at least" or
"not more than." If "no more than" is imported into the question,
then "5" is not a true answer where 25 is the case, whether or not the statement "I did it 5 times" would be true as an answer to any other understanding of the question. Justice is not served by lawyers and witnesses doing the Moore/Russell "Beyond the Fringe" apple routine."

----

So, here I would bring back a trichotomy:

a. English
b. Logiclandian
c. Legalese (or courtroom examination routine).

I agree that there are various understandings, etc. ---

Notice that there is a slight difference between Kramer's example:

VERSION A:

A "Did you go to the store fifty times?
B. "I went five times."


(adapted -- giving a conversational completion to the District Court fictitious case).

Version B:

"B: I went to the store five times"

---- (which is the type of thing the District court is referring to as it being 'technically true' even if the U thinks that he's been to the store 50 times. Or at least they are considering the scenario where that would be 'technically true')

And there's the erotetic version C:

A: How many times did you go to the store?
----- an x-question rather than a yes/no question.
B: Five times.


So I agree with Kramer that we have to distinguish between implicatures (and indeed entailments) in questions from implicatures and entailments in answers. It seems a yes/no question is NOT exactly what academics call an "Essay question". An x-questions DOES give you more of an opportunity to ... expand. But this may be restricted by "nothing but the truth". And it is not uncommon to see the lawyer (in films in any case) to go, "Just answer: yes or no". Recall the survey in the Club: "Is Grice the greatest philosopher that ever lived?". I know, ridiculous. But the person who created that query added, a "Truth-Value Gap" that already collected a few. ---

----

Kramer:

"In contrast, there is nothing comparable that can be imported into "Do you like Good and Plenty?" that would make "I like M&M's" false rather than merely evasive."

---- Well, but again, the implicatum in question is:

"U does not like Good and Plenty".

And there's the case of 'and' reduction:

A: Do you like Good and Plenty and M&Ms?
B: I like M&Ms

where the implicatum seems to be carried -- by default. It IS cancellable in that the addition, "... But of course I'm in no way suggesting that I do not find Good & Plenty very liked by me, too" (I wouldn't say "likeable" because that's modal).

A: Do you like Good & Plenty?
B: I like M&Ms (+> "No, I do not like Good & Plenty")

------

In what ways does it breach the CP?

1. Quantitas: B's utterance is NOT an answer.

2. Qualitas: U (Utterer, i.e. B) is possibly in a position
-- to know if he does like Good & Plenty. Of course he may
have never have heard of them, and not wanting to lose
face, just sticks to what he knows: M&M. He's ignoring a point
rather than evading it?
---------- Cfr. "Do you like Gilbert & Sullivan?"
----------------B: "I like Sullivan".
----Cfr. "Do you like Puccini?
------ B: I like Verdi.

-- Do you like Boston?
-- B: I like New York.

A bit of the previous sequence would help. If the topic is East Coast seaboard towns one enjoys, B's answer is not really evasive, is it. It means, "I like New York better." -- Out of the blue, the question SEEMS to show an interest in the questioner as to whether U likes Boston or not. But I know a lot of people who drop names just to 'keep the conversation' going, and one, aware of that fact, ignores some of the names dropped and sticks to what one likes the conversation to proceed, or something.

---

Kramer:

"In such a case, the implicature is wholly the witness's, and the lawyer is obliged to cancel it."

This conclusion pertains to cross-examination, and legalese, I would submit. So, I was wondering if one can expand to Logiclandian and English simpliciter.

What is that is wholly the witness? The implicatum:

"U does not like Good and Plenty".

---
What about the tag:

"The lawyer is obliged to cancel it".

What about English speakers simpliciter, or logicians?

Logicians are ALWAYS obliged to cancel implicatures. They are just NOT allowed in a proof derivation. You cannot proceed from a premise involving (Ex) "some" -- to a premise involving "not all" (~(x)) for example. -- And 'numerical quantifiers' ARE a bother -- in that the 'at least' and 'at most' require clauses that are best represented in the logical form. As in Grice's expansion of "The king of France is bald" -- there is at least a king of France; there is at most a King of France, and whatever is king of France is bald."

(( --- This holds well for 'king of France' but it is still different for:

"The three stooges are fun" -- for there is at least three stooges and there is at most three stooges, and whatever is a stooge is fun -- does not quite respond to the 'uniqueness' of the three stooges as it responds to the uniqueness of 'the' entailed by 'THE' king of France.

--))

---

So I would submit that the only ones who are NOT obliged to cancel an implicature, by elimination, is the English speaker!

---

Here, Grice speaks of the very interesting notion of 'non-controversial'. And I would add that in cross-examination, EVERYTHING should in principle BE controversial.

--- How does Grice introduce 'non-controversial'. It's easy enough to formalise: it's square-bracketed:

[There is at least one king of France & there is at most one king of France] and whatever is king of France is bald.

--- The first two clauses are common ground or uncontroversial. His example is his aunt's cousin:

Grice is referring to some bit of the semantic component as

"hav[ing] what I might call common-ground status, and
therefore, [as] not something that is likely to be
challenged."

--- and this may interest J, since it's Grice at his Toulmin best.

"And", Grice adds, "one way in which this might happen"

--- i.e. an element in an utterance to have 'common-ground status'

"would be if [U(tterer)] were to think or assume that it
is common knowledge, and that people would regard it
as common knowledge" (that p).

"But that would be only ONE way in which it could arise."

---- Comes his aunt.

"For instance, it is quite natural to say to
somebody, when we are discussing some concert,

---- "My aunt's cousin went to that concert."

"when we know perfectly well that the person we
are talking to is very likely not even to know
that we have an aunt -- let alone know that our
aunt has a cousin."


"So: the supposition must NOT be that it is common
KNOWLEDGE<, but rather that it is noncontroversial."

"And this in the sense that is something
we would expect [A(ddressee)] to take
from us (if he does not already know."

---- This has an impact on cross-examination, which gives the King in "Alice in Wonderland" a forehead ache, naturally --, for as Kramer stated, Rule Number 1 is
that the question already HAS a known answer -- but on the other hand, that's not the
natural use of 'take [something] from us', as a witness is being cross-examined.

---

Grice goes on:

"That is to say, I do NOT expect,
when I tell someone that my aunt's cousin
went to a concert, to be QUESTIONED
whether I have an aunt, and, IF SO,
whether my aunt has a cousin."

---- I suppose an unexperienced lawyer would perhaps go with such questions -- and an unexperienced logician too. I would be particularly confused, though, if U starts to add negative particles, "It is not the case that my aunt's cousin did not really enjoy the concert, though". I may like to work variants on where the ~ is supposed to have scope over -- i.e. provide a 'slate' with the actual commitments of the utterance.

Grice goes on -- this is all WoW:274ff --

"This is the sort of thing"

--- that I have an aunt and that my aunt has a cousin --

"that I would expect HIM to take from me,
that is, to take MY WORD for."

---- But again, in cross examination, they take an oath from you, but may still inquire it to SOME limits!

Etc.