Kaarlo Jaakko Juhani Hintikka and Herbert Paul Grice: Implicature as a Game

Speranza

"Logic of Conversation as a Logic of Dialogue" is K. Jaakko J. Hintikka's 
contribution to the Grice festschrift, in P.G.R.I.C.E., Philosophhical
Grounds  of Rationality: Intentions, Categories, Ends.

In this celebratory essay, Hintikka -- whose affiliations to the Oxonian 
school of philosophy to which Grice belonged were minimal -- compliments
Grice  on a body of work.

However, Hintikka singles out “Logic and Conversation” to criticize --
i.e.  the second lecture at Harvard, where Hintikka was a fellow. It should be 
reminded that this was just ONE out of a series of lectures that Grice gave
at  Harvard, as he tried to combine his interests in implicature
(introduced as a  coinage -- but cfr. Sidonius, "implicatura", Lewis/Short, Latin
Dictionary) and  meaning.

Hintikka believes that the Gricean maxims "are not, and cannot be, the rock
bottom of a satisfactory analysis of the logic of conversation" (Hintikka
1986,  273). 

Hintikka is using 'rock bottom' figuratively. Figures are tricky. Grice's 
example is

i. You are the cream in my coffee.

which is literally false as it involves a categorial mistake, unless we 
suppose that one's addressee -- cream -- has ears to understand what Utterer
is  saying.

Similarly, a maxim is not a rock bottom.

Kant knew this, and by using "maxim", Grice is merely, and jocularly (such 
was his obscure sense of humour) referring to Kant (as does Joaquin Phoenix
in  the recent "Irrational man" when he discusses Kant's categorical
imperative and  Kant's universal prohibition on lying).

One of the reasons Hintikka thinks this -- that a maxim is not a rock 
bottom -- is his belief that, “when the time comes to conceptualize the results 
of […] discourse-theoretical observations Grice often seems to retreat back
to  formulations that pertain to utterances taken one by one rather than to
the  interplay of different utterances in discourse".

Which is of course an OVER-SIMPLIFICATION.

In the original OXFORD (*not* Harvard) lectures on implicature, Grice 
tortured his students ALWAYS with dialogues, between whom Grice called "A" and 
"B". Perhaps his most famous example is:

A: I'm out of petrol.
B: There's a garage round the corner. (Implicature: Which is open and with 
petrol to sell).

Grice is considering co-operativeness (or helpfulness) and you cannot have 
THAT with a single utterance, unless you are God ("Let there be light!").

Hintikka is interested in a "different, more flexible framework in which 
the dynamics of discourse are spelt out more explicitly."

The use of 'discourse' is vague. Grice prefers the more conversational 
adjective, 'conversational' and its corresponding noun, 'conversation'.

First-order predicate logic is clearly not the logic of dialogue. 

This point, as Hintikka wants to explore, leads to a fundamental difference
between propositions and the utterances of dialogue. 

The new strategy for understanding conversation Hintikka wants to employ is
explained as follows.

Grice says that one of his ‘avowed aims is to see talking as a special 
case or variety of purposive, indeed rational, behaviour’.

If so, the bag of conceptual tools one can profitably use in studying 
conversational "logic" (although Grice never used this phrase) should be a 
special case, or variety, of the conceptual tools one uses in studying the 
rationality of human behaviour in general.  One such tool is game  theory.

This had been tried before Hintikka, and at Harvard, too, by D. K. Lewis in
his dissertation under Quine. Lewis was thinking that co-operation is 
conventional and that conventions invite a game-theoretical approach.

Thus, Hintikka argues that the framework for studying dialogue needs to  be
shifted from formal logic to game theory. 

Game theory is geared toward better understanding which appropriate 
strategies one ought to use in given situations, or games. 

The use of 'strategy' is vague. Strictly, for Grice, there are no 
strategies, since he knew Greek, and 'strategy' comes from a Greek word meaning 
'general' -- as in military. And a military person KNOWS that his strategies 
have to be SECRETIVE. But Grice wants all ABOVE-BOARD. So we need another
term  to categorise what is going one when, to use S. Yablo's phrase,
'implicature  happens'.

Hintikka sketches a simple schema in which conversations can be viewed 
game theoretically.

He writes, “Utterer 1 and Utterer 2 make ‘moves’  alternately. 

There are four different kinds of moves for Hintikka.

-- The phrase 'conversational move' and 'conversational game' and 
'conversational rule' DO APPEAR in Grice's "Logic and Conversation", so it's not 
like Hintikka is being 100% original --

(a) Assertoric moves. 
----- as in "We've been having some delightful weather this summer, no?" --
This implicating: "You've just committed a social gaffe: what about
changing the  topic?")
(b) Interrogative moves. 
----- Humpty Dumpty: How old did you say you were?
----- Alice: Seven years and six months.
----- Humpty Dumpty: Wrong! You never said a word like it!

(c) Deductive moves.
---- as in what Grice calls trivial reasoning:
-------------------- A: You have two hands, right?
-------------------- B: Yes.
---------------------A: Well, if I had THREE more hands, you  would have 5
hands.
---------------------B: And if I were permitted to double that, I  would
have 10 hands.
---------------------A: Exactly: But If four of your hands were  removed,
you would still remain with six hands.
---------------------B: Exactly, and I would still have four more  hands
than I have now.
---------------------A: Brilliant.

(d) Definitory moves, as in
-------------------A: All swans are black.
-------------------B: You never ventured outside Australia, right?

Hintikka then explains how each of these steps work. 

First, a "player" must make an assertoric move -- as in "The weather has 
been delightful this summer, no?" --  in which he or she “puts forward a  new
proposition (a new ‘thesis’)".

An interrogative move is a questioning move, the answer (if one can be 
given) to which "is then added to the list of the answerer’s theses".

The deductive moves are pretty straightforward, it is comprised of "a 
logical conclusion from the totality of his/her opponent’s theses," and previous
conclusions obtained by the same means.

Finally, definitory moves are when one "introduces a new non-logical symbol
by and appropriate explicit definition".

These four conversational moves in the conversational game are used to 
prove all the players’ theses, but according to Hintikka the goals can be 
varied. 

Hintikka believes that the Gricean maxims can be incorporated  into this
model. Had he attended Grice's original implicature lectures at  Oxford, he
would have known Grice had already done that!

In referring to maxims enjoining 'strength' or 'informativeness' of 
conversational moves, Hintikka notes, while Grice remarks that one ought not 
violate these maxims for fear of confusing one's co-conversationalist, for 
Hintikka there "is nevertheless operative,… in ordinary discourse, a different 
pressure against extra information.  Everything a player of my dialogical 
games says can be used against him (or her) by the opponent".

Here, the player will want his discourse to be as weak as possible.

Thus, requiring him to prove less by the rules of the game. 

This is a fundamentally different reason to act in accordance with the 
maxims enjoining strength or informativeness. However the "end result" is  the
same. 

The same result will be found regarding the maxims of  quality, enjoining
trustworthiness.

Recall that Grice uses 'quality/quantity/relation/mode' just to tease  Kant!

Because one only gets a payoff by proving the maximum amount of statements 
in the dialogue, one will only want to propose things that he or she may be
able  to show to be true. 

Surprisingly, the maxims enjoining trustworthiness are also satisfied  by
this game. 

Hintikka writes: "iif my opponent gives true answers to my question, if the
opponent is fairly well-informed, and if the effects of my own answers can
be  discounted, then it is ceteris paribus in my own best interest to put
forward  true theses".

Modus, unlike quantity, quality, and relevance, is not of interest to 
Hintikka. 

He states that it "is different in kind from the first three" -- which 
shows that Hintikka does NOT share Grice's obscure sense of humour. What is the
point of poking fun of Kant if you are not going to accept the FOUR 
'categories'. Note that in an outburst of humour, Grice calls these the 
'conversational categories' -- in his critique of conversational reason, of  course!

This should become apparent, as arguments to this affect will be made 
later. 

Relation, however, must be addressed and is actually reworded to state that
it is a move within the rules to increase one’s pay-off. 

This Hintikka must explain.

He states: "For instead of the relevance of the several utterances in a 
dialogue I could collectively speak of the coherence of the dialogue".

Hintikka refers to a Sherlock Holmes story (now played by Ian  McKellen) in
which Holmes solves a mystery about a prize race horse by  asking a
shepherd an apparently irrelevant question about the recent status of  his sheep.

However, this question, as is often the case with the solutions to 
intricate puzzles, was the crucial link between a series of facts that  ultimately
achieved the goal of solving this mystery. 

This shows that Hintikka read the novels of Sherlock Holmes, as did Grice, 
and Grice's mother.

From all this, one can see that Hintikka has crafted a formal game that 
models the Griceian maxims. 

This game does not however require a Cooperative Principle, which was 
another outburst of humour on Grice's part. In his Oxford lectures on 
implicature, he had spoken of desiderata of candour, benevolence, clarity, and 
self-interest, to account for the same phenomena! He wasn't literally wedded to 
calling his desiderata, principles, maxims or stuff THIS or THAT!

For Hintikka, in fact, this becomes a competition between the players of 
the game. 

Still, there are clearly some problematic results of this account. 

For example, intuitions of conversation stray far from this schema. 

Conversations are certainly not games in which one must prove, or at least 
hope to prove, all the propositions that one puts forward. 

Still, the idea of conversation as a goal oriented game, with pay-offs and 
costs, is certainly an idea which has not been explored, and may have some 
benefits. 

The primary significance of Hintikka's Griceian exegesis, however, may  be
his alternative approach to the theory of conversation, which is based  on
rationality theory in a way that explores Grice's ideas under a different 
light.

Oxonians prefer Grice's Oxonian light, though -- any night!