Born to be Grice

In "Born to be Good: The Science of a Meaningful Life" (Norton, 2009, pp. 153-6) Dacher Keltner shows a love for Griceian maxims. (Thanks to R. O. Doyle for sharing this with the club -- and to his wife for reading such interesting stuff!).

D. Keltner writes "THIS AND NOT THIS".

Keltner writes:

"In the mid-1960s, philosopher [H.] Paul Grice outlined four principles of communication that would profoundly shape the study of pragmatics - that is, how people speak."

Indeed. I would think, with R. O. Doyle, that 'principle' is overstated. Doyle and I share a love for all things Greek. He loves the old Greek philosophers who developed free-will, and I just love the Greek language. In Greek, 'principle' (principium, in Latin), is "arkhe". One may wonder if there could be more than ONE principle. I doubt it. I would think principles have to be ONE. For Grice, it was the co-operative principle (or principle of cooperation, as I prefer, since a principle, in principle, cannot be itself cooperative).

In his 1964 earlier lectures on Logic and Conversation (Oxford, rather than Harvard) Grice had talked of two different (two!, note!) principles: of self-love and benevolence (strictly, 'principle of conversational beneveolence' and 'principle of conversational self-love'. While it would be odd to think that Grice is echoing Kant in those earlier lectures, history did show that it was Kant who Grice was seriously echoing in the Harvard lectures. As he notes in "Retrospective Epilogue", the cooperative principle was meant to echo Kant's categorical imperative. Not so much in substance, but in the fact that from it one can derive, rationally, some _maxims_.

----

Keltner goes on:

"Sincere communication, according to Grice, involves utterances that are to be taken literally."

Indeed. I follow J. Saul in thinking that people are being, essentially, uncooperative when, e.g. they utter a metaphor. ("The moon is made of cheese"). Truly, Grice says that in metaphor (or any flout, really, to the cooperative principle) the SPIRIT, if not the letter, of the cooperative principle, is still being respected. But surely, there is a lot of misunderstanding that "The moon is made of cheese" can yield. For Davidson, who adhered to a truth-conditional semantics, "The moon is made of cheese" is true if and only if the moon is made of cheese. From which one yields that "The moon is made of cheese" means that the moon is made of cheese. Grice's example,

You're the cream in my coffee,

again, qua compliment, can yield a few misunderstandings. ("I didn't know I was a liquid").

---

Keltner goes on:

"These statements should adhere as closely as possible to four maxims (see table below)."

The use of "Table" is excellent and perfect in that it's _so_, oh, _so_ Kantian. (Tabel).

Keltner goes on:

"Statements should follow the rule of quality - they should be truthful, honest, and based in evidence."

While Grice does use 'conversational rule' on occasion, in the Harvard lectures, he is clear, from the "Prolegomena" that he is a campaign against Searle. As R. O. Doyle knows, Searle had and has the gift for the good, effective, phrase ("scandal of free will"). In this, Searle, following, I think, Rawls, had said that rules could be constitutive or regulative. Grice found that otiose, and his campaign was one where the idea of 'rule' was being minimised. Doyle, who has, like me, criticised Chomsky, understands the issue here: to think of a rule -- as in Witters on "following a rule" -- is a deep difficult question. Conversational 'rules', since they can be flouted so easily, are not cognitive like that. Grice also says, 'rules of the game', even 'conversational game' -- my PhD dissertation maintained that terminology --. But it is controversial that no conversation (qua game) is played if 'rules' are 'broken'. In my PhD dissertation, I argued, typically pedantically, that what is involved here is a _weak_ transcendental justification, alla Kant. For, not the 'existence' of 'conversational moves' ("it is raining", when utterer believes it is not -- a plain lie) but for the 'appropriateness' of 'conversational moves' (hence the paradox of Moore: "It is raining, but I don't believe it").

----

Keltner goes on:

"Statements should be appropriately informative - the rule of quantity - and avoid the Strunk and White catastrophes of being too wordy or opaquely succinct."

---

Again, Grice is _playing_ with Kant. And in Kant's table of categories (ontological categories, or epistemico-ontological), it's qualitas, quantitas, relatio, and modus. Kant is building on Aristotle. Indeed, simplifying Aristotle (who said categories were 10, rather than 4). Grice is 'echoing Kant' as he puts it, in sticking with those four categories, which Grice now dubs, within the context of the cooperative principle for the generation of conversational implicature, 'conversational categories'.

---

Keltner goes on:

"Statements should be relevant and on topic and avoid meandering into digressions, irrelevances, or stream-of-consciousness flights of fancy."

Again, this is Kant's relatio (Relation). A category in Aristotle, too. Along with qualitas and quantitas. Note that Relatio marks a divergence. Both qualitas and quantitas, in Latin, beging similarly, since the quale and the quantum are formations out of the root, qu-, which is common to both. This is maintained in Greek, too. Aristotle is indeed credited with having _coined_ 'qualitas' and 'quantity' in Greek.

Keltner goes on:

"Finally, in honoring the rule of manner, statements should be direct, clear, and to the point sorry if I've violated this one)."

---- The category of Modus (also in Aristotle and Kant). Note that this all applies to categories OTHER than 'substance' which is the primary category. Only when we have a substance (of conversation -- cfr. topic?) can we then go to apply these four categories to it: quality --. E.g. "Socrates is white". Quantity: "Socrates is one" -- "I have two eyes" -- Relatio: "Socrates is to the right of Plato" --. Modus: "Socrates is asleep". And so on. In Kant, these are categories of JUDGEMENTS. And so, there is a rather different treatment, e.g. of 'relatio', which relates (sic) to the three types of judgement (molecular judgement). In Kant, interestingly, and somewhat arbitrarily, there is systematically a three-fold division of each category. None of this limitation in Grice (or indeed Aristotle). Note that e.g. for Kant, in terms of qualitas, it is the affirmatio or negatio of a judgement that is quoted; and it is the distribution of Predicate P onto subject S that is assessed when it comes to Quantitas. The modus, which is rather secondary in Grice, relates to central issues for Kant, such as necessity and analyticity (apodeicticness).

----

Keltner goes on:

"Utterances that follow these four simple rules are on-record, end are to be taken literally. When an MD provides a prognosis about a life-threatening condition, she should follow these four piles of on-record communication. So too should the financial advisor announcing the unexpected loss of a family fortune -- these are not the best moments for exaggeration, intentional falsehoods, fantastical description, obvious repetition, digression, meandering, or catchy metaphors or poetic obliqueness. Much of our social life, in fact -- romantic declarations, sealing business deals, critiques at work, teaching young toddlers reaching to touch red-hot burners or rabid dogs -- transpires in this realm of literal on-record communication."

I'm not so sure about romantic declarations! But then blame Strawson on that. He said that to think that conversation follows rules is as abstract as to think that all romance should proceed along the guidelines provided by "Romance of the Rose"!

----

Keltner goes on to systematise:

"GRICE'S MAXIMS OF COMMUNICATION


LINGUISTIC PRINCIPLE CRITERION VIOLATIONS

QUALITY TRUTH EXAGGERATION, FANTASTICAL DESCRIPTION

QUANTITY INFORMATIVENESS REDUNDANCY, REPETITION, EXCESSIVE BREVITY

RELATION RELEVANCE DIGRESSION

MANNER CLARITY VAGUENESS, OBLIQUENESS, METAPHOR

----

A few nitpicks: I would think 'metaphor', for Grice, is a basic "Quality" flout: "The moon is made of cheese". "You are the cream in my coffee". Yes, manner is involved. But it's more like a flout to "be truthful". Note that Bach and Harnish, following Grice, think that everything has to be generalised. I did so in my PhD dissertation -- I was happy Habermas read it! -- So I had to be careful to AVOID 'truth' -- but 'trustworthiness' was used, rather. "Close the door!" can hardly be said to be _true_ (or false, for that matter). But a trustworthy utterance of "close the door!" is one meant to have the door closed. Grice considers countersuggestion. If I know (as in fact I did know, on occasion) that B will do exactly as I DON'T tell him to do, by uttering "Do x!" I may mean, in a sneaky fashion, that B is NOT to do x. But I'm not being trustworthy. E.g. "Please don't tell Dad about this!". And so on.

Keltner goes on:

"When we intentionally violate Grice's maxims, we signal that alternative interpretations of the utterance are possible. We say "this" with our words, and "not this" with violations of Grice's maxims, pointing to other possible meanings of our utterance."

Indeed. Good way of putting it. I used to call this "reading between the lines" -- but then I found that intralinear reading was a British custom which should be forbidden. To read between the lines was, originally, to read the English text of, say, Caesar's Gallic Wars. In Loeb's Classical Library -- that Doyle and I adore -- one cannot do that. One can read, 'across the page', but that's a different story.

---

Keltner goes on:

"We signal "not this" by resorting to obvious falsehoods or exaggerations of the truth (which violate the rule of quantity). We can provide too much information, for example in systematic repetition, or too little information, thus violating the rule of quantity. We can dwell in the irrelevant to violate the rule of relation. And we can resort to various linguistic acts - idiomatic expressions, metaphors, oblique references - that violate the rule of manner and its requirements of clarity and directness."

Indeed. Grice does not provide, for the record, too many examples of violations to "relatio"

---- A: Mrs. X is a windbag.
------- B: The weather has been deligthful this summer, so far, no?

---- Grice's gloss: B means that A has committed a _gaffe_. It may well be the ONLY example Grice ever gave of this. Susan Mura has written on violations to Grice and she must have gathered other examples:

This from "Conversation":

A: How much are they paying you at XEROX?
B: I think I _will_ order the chocolate mousse.

----

Note that Grice couldn't have been thinking that "Relatio" (of this type) is of grand philosophical importance. In my PhD disseration, I focused on truth-functors for molecular utterances ('and', 'or', 'if') and so, did discuss how Relatio does fit in for each of these. But these are not really examples that Grice elaborates explicitly, even if the ideas are, of course, all there.

---

Keltner goes on:

"As important as sincere speech is to our social life, so too is this realm of nonliteral communication."

Perhaps 'literal' is too literal. It comes from Greek, 'graphos', littera. But surely oral communication (as evolutionary theory) shows is primary! So, we need perhaps a better monicker. I prefer, oddly, 'explicature' (explicatura, in Latin). To contrast it with 'implicatura' (as used by Sidonius). So, the explicatura is

"You ARE the cream in my coffee"

The implicatura is "I love you", almost.

---

So, the literal is "you are the cream in my coffee". The nonliteral, the implicature, the unsaid, is "You are my pride and joy".

Grice notes that implicatures can build one over another. "You are the cream in my coffee", NOT in a declaration of love, can be used, ironically, to mean, "You are my bane" (Grice's gloss, again).

----

Keltner goes on:

"Our brief utterances can take on the opposite meaning of what the words denote (irony, satire). We can connect disparate concepts in communicative acts that leap beyond narrow literal denotation (metaphor). We can endow our utterances with multiple layers of unbounded, aesthetically pleasing meaning (poetry). The relevance of Grice's maxims to teasing, ironically enough, is revealed in linguists Brown and Levinson's 1987 classic, "Politeness."

----- now a separate book.

"Brown and Levinson carefully document how in the world's languages speakers add a layer of politeness to their utterances when what they say risks embarrassing the listener or themselves."

Actually, it was Grice who formulated the rule,

"Be polite!"

He wrote. Lakoff (Robin Talmach) heard that. Penny Brown and Levinson, too. But since Grice (unlike, perhaps R. Lakoff, or P. Brown, or S. C. Levinson) was a philosopher, and giving the William James lectures at that, did not elaborate. Rather, he did.

He wanted to say that

"Be polite!"

is a DIFFERENT sort of maxim. It cannot be so central as to generate the type of implicature he was into -- the 'conversational' implicature -- a bit like Nowell-Smith, that Doyle and I adore -- and his 'contextual implications'. Grice's conversational implicature is a basic phenomenon for philosophical elucidation. Cf. the implicatures of "of his own free will".

"Be polite!", rather, works at a different level. It is not about the maximally efficient exchange of info. It works at a different level. Grice did recognise this level, and allowed it to exist, of course. Only that, as coiner of expressions, he asked please to be distinguished from generators of 'conversational implicature' proper.

Thus, when it comes to "Logic" -- recall that the title of the James lectures, and indeed the Oxford ones is "Logic and Conversation" -- there is little (if anything at all) that "be polite!" contributes to 'and', 'or' or 'if'.

By the same token, Grice formulates that 'be polite' may fall within other maxims which he dubs 'moral' (sic) or 'aesthetic'. This has Kantian relevance, in that the grounding of the cooperative principle on maxims which are _not_ moral makes the whole campaign NOT circular.

("moral", not 'utilitarian').

---

Keltner goes on:

"Politeness is achieved through systematic violations of Grice's four maxims. Consider the simple act of making a request. If someone asks you for the time, or directions, or to pass the rutabagas, or not to talk so loudly during the previews, that act is fraught with potential conflict."

Indeed, in an excellent development of this metaphor, R. Lakoff speaks of 'minding your ps and qs' ('p', for Grice and logicians, in logic, is short for 'proposition'; 'q' is an arbitrary symbol for 'proposition' other than given 'p')

Keltner goes on with the pragmatics of requests:

"The recipient of the request is imposed upon and risks being revealed as incompetent, boorish, or disinterested in social conventions. The requester risks being perceived as intrusive and impolite."

In a genial move that connected with Goffman, Levinson and Brown spoke of face, which, as we know -- or rather the Japanese, apparenty, knew, since this is an "Oriental" notion -- can be 'positive' or 'negative'.

Keltner goes on:

"To soften the impact of requests and other potentially impolite acts such as recommendations, or criticism, people violate Grice's maxims to communicate in more polite fashion. Say your best friend is being a bit boisterous, with elbows flying, at your Friday evening line dancing group that you've generously invited him to. To encourage a bit more restraint, you might politely resort to indirect questions ("Have you ever seen yourself dance?"), rhetorical questions ("Have you done line dancing before?"), metaphors ("Wow, you holler like a howler monkey")"

--- oddly, this is a comparison. "You holler, you howler monkey you" seems to _presuppose_ a metaphor, even. "You are a howler monkey" is the underlying pure metaphor).

Keltner continues:

"and obliqueness ("I bet you'd be a terrific clown"). We break the rules of sincere communication to be polite. Equipped with this analysis of nonliteral communication, a careful examination of the tease reveals that teasing and politeness are surprisingly close relatives. THE ART OF THE TEASE. What gives the tease the playful genius of the jester's satire are systematic violations of Grice's maxims."

What an excellent summary, analysis, and application Keltner has engaged us in.

Keltner concludes the Griceian reference:

"A first principle is exaggeration, which marks the playfulness of the tease by deviating from Grice's rule of quality. Teasing can involve copious detail, excessive profanity, or an exaggerated characterization."

and so on.

So, born to be Grice indeed, or the art (and science, why not) of a -- Grice's, say -- meaningful life.

(Thanks again to R. O. Doyle for sharing this Keltner reference with the club -- and indirectly, to his wife for reading the material in the first place!)