Why Grice Found Kant Boring -- and why Grice preferred to quote "Logic and Conversation" from Davidson/Harman

From an online review by Harman of Grice 2001:

"Grice's "Aspects of Reason" offers a sketch of parts of an ambitious project to derive important philosophical consequences from the idea of a rational being, with
hopes of improving on Aristotle and Kant."

"Jonathan Dancy offers a good account
and critique of this project."

"In this review, I will mention and discuss
some important things that Grice says along the way about what reasoning
is, what reasons are, the relation between theoretical, or

“alethic”,

reasons
and practical reasons, and the analysis of practical and

alethic

modality."


"As with all of Grice’s work, the book is rich with interesting, often competing,
ideas."

"As usual, there is great power and compression of argument."

"It is not light reading and requires concentrated study."

----- with a frequent visit to the Club helps.

"The value of the
work lies in not its highly tentative conclusion but in its consideration of
possibilities, suggestions, objections, and replies."

-- and reprints!

"The book is a significant
resource for serious students of the issues it considers."

"What is reasoning?"

"Grice begins with a simple model based on logical
argument or proof."

"“Let us, then, take as a first approximation to an account
of reasoning the following: reasoning consists in the entertainment (and often
acceptance) in thought or in speech of a set of initial ideas (propositions),
together with a sequence of ideas each of which is derivable by an acceptable
principle of inference from its predecessors in the set” (5)."

"Grice then notes a
number of problems with the model."

"First, not all reasoning is good reasoning, but the model does not apply to
bad reasoning."

"What makes some thinking an instance of bad reasoning
rather than not reasoning at all?"

"Here Grice suggests modifying the model
to allow that

“steps in actual reasoning . . . are either validly made or are
thought to be validly made” (6)."


"A second problem with the model is that good reasoning often does not fit
the model because of being inexplicit about what premises are appealed to
and/or incomplete in specifying intermediate steps."

----- this may relate to Jones's point about entailments. If one is committed to the entailments of what one says, how explicit should the treatment of what one says be?

"We cannot just attribute
reasoning to someone by adding premises and intermediate steps that would
make the argument fit the original model."

"For one thing there are different
ways to do so."

"For another, it would not allow us to distinguish good
reasoning from bad reasoning."

----- as I prefer to say: No book can be so bad that does not contain something good about it. Ditto with reasoning. And with EVERYTHING in general:

---- "No ... can be so bad that there is not something good about it."

---- or as my mother, quoting Bambi's mother, prefers: "if you don't have anything nice to say, don't!"

---


"Grice observes that we cannot determine what reasoning people had in mind
by asking them, because when asked people typically construct a further
argument rather than report on what they had in mind (12-13)."

"One idea is to count as good reasoning any reasoning which “may be converted
into reasoning which conforms to canonical standards of respectability
by the addition of further premises which the reasoner has in mind, either
(i) explicitly or (ii) subliminally.”"

"We might also include reasoning for which
“the reasoner thinks that such premises exists . . . ” (8)."

"This does not yet handle cases in which the explicit reasoning does not
include needed intermediate steps."

"Grice compares filling in steps of some
fairly bizarre reasoning by a fellow student he once knew with a logician who
offers a proof (or sketch) that is six pages long whereas the filled in proof is
eighty-four pages long (11)."

"We might want to attribute the filled in proof
to the logician but not the filled in reasoning to Grice’s fellow student."

"Perhaps here we must add that the person in question must reasonably believe
in the existence of the needed premises and intermediate steps."

"Must
such a reasonable belief itself be the result of reasoning?"

"If so, a possible
regress looms."

"Anyway, one might reasonably believe in the existence of intermediate
steps needed to get to Andrew Wiles’ proof of Fermat’s theorem,
but that is not enough to attribute to one the reasoning that is spelled out
in his proof."

"Of course, even Wiles’ reasoning is not fully explicit."

"Perhaps what’s needed is that the person in question must have a good idea
of how the proof would go and must know that he or she could spell out the
argument in greater detail if necessary."

"Grice’s third problem with the initial model of reasoning as a kind of valid
argument or proof is that it is too strong."

"One can produce a string of trivial
consequences without it being appropriate to say one is reasoning."

E.g.,

“inferring”

P

from

P, or from

P&Q.

---

"Such performances “offend against
something . . . very central to our conception of reasoning . . .Mechanical applications
of ground rules of inference . . . are reluctantly (if at all) called
reasoning” (15)."

"“Some examples are deficient because they are aimless or
pointless. Reasoning is characteristically addressed to problems . . . ” (16)."

---- But Grice allows he is punning on his earlier self obsessed with implicata. Surely trivial reasoning IS reasoning, as a red pillar box which seems red may BE red.

---

"In the extreme, if one supposes that corresponding to basic principles of
logical implication are rules to be followed if one is to be rational, then once
one was given some premises, one would forever be making inferences, with
no way to stop (36)."

"Fourth, there are many cases of reasoning that do not seem to fit the logical
argument model."

"Grice gives an example of someone who has promised to
give a series of talks, which he hasn’t yet started to think about, who is
asked for the titles of the talks so they can be announced."

"The speaker will
think of various possibilities, will wonder whether to cancel, will recall how
things have worked out in the past, and will do other things that do not
seem to fit the model at all."

"But in doing those things the speaker will be
reasoning."

"So, the original model has various flaws."

"Nevertheless, Grice proposes in
what follows to assume that some form of the model is acceptable."

"Since one
thing that the model leaves out is that reasoning is an “activity” connected
with “the will,” he proposes that

x reasons (informally) from A to B just in
case x thinks that A and intends that, in thinking B, he should be thinking
something which would be the conclusion of a formally valid argument the
premises of which are a supplementation of A.

----


"With Dancy, I think this is problematic."

"Reasoning may sometimes involve
constructing an argument, but not always because one is reasoning from the
premises of that argument."

"The argument is often an explanatory argument
and one is reasoning from the conclusion of that explanatory argument to a
conclusion that is a premise of the argument."

"Furthermore, it is not obvious
that reasoning always involves construction of an argument or argument
sketch."

"Perhaps we can understand Grice as concerned especially or only
with those cases in which it does, but that can be very misleading, especially
(as indicated below) with respect to practical reasoning."

"In any event, given this model of reasoning, Grice is concerned with whether
the basic principles of reasoning and rationality apply equally to

alethic

and

to

[non-alethic] practical reasoning."

"Grice suggests that modal terms like “must”, “ought”,
and “necessary,” might be seen as applying univocally both to things to
be believed and things to be done.2

"Although a sentence like

Jack should leave soon.

has at least two interpretations,

ALETHIC:
It is likely that Jack will leave soon.

NON-ALETHIC:
It would be good if Jack were to leave soon.

"Grice argues
that this does not show any ambiguity in “should”."

"The ambiguity lies in
whether the underlying clause to which “should” applies, has an alethic or a
practical interpretation."

"According to Grice, this clause combines a sentence
“radical” together with a mood (or “mode”) operator applied to that radical
and indicating the relevant mode, practical, alethic, or interrogative."

"He is
mostly concerned with the first two only."

"As Dancy explains, Grice introduces the following formalism."

--- if this is formalism, my aunt understand Einstein.

"He uses “!”
for the imperative form and “/-” for the propositional or

“alethic”

form."

"So,
“/-Jack go” means

Jack goes.

and

“! Jack go” means

Jack is to go.

"Then the
two interpretations of

Jack should leave soon.

are represented as

Should ` Jack leave soon”

Should ! Jack leave soon.”

"Grice explains how the meanings of these mode operators might be explained
within his general theory of meaning."

"He also suggests that it may be
possible to explain various modal notions in terms of derivability from some
system of principles, where the operators ! and /- might help to indicate (or
put constraints on) the system to be used."

3

"Grice proposes further that variation in these principles might account for
a kind of ambiguity in practical uses of modals that does not appear to
occur in

alethic

uses."

There seem to be at least two interpretations of

Jack should leave soon.

One interpretation assigns a certain responsibility on
Jack for leaving; the other does not.

"The one interpretation appears to
treat “should” as specifying a relation between Jack and a certain possible
action."

"The other interpretation appears to treat “should” as a sentential
operator indicating the

DESIRABILITY

of a certain state of affairs, not implying
any responsibility on the subject of the sentence."

"Suppose that Jack and Bob are to play tennis and consider the remark,

Jack should beat Bob.

said not as a prediction but with one or the other
of the practical interpretations."

"In the relational interpretation, this remark
is not equivalent to

Bob should lose to Jack.

"In the sentential operator
interpretation, the two sentences are equivalent.

Analogously,

It would be good if Jack beats Bob.
It would be good if Bob loses to Jack.

are
equivalent, but

Jack has duty to beat Bob.

is not equivalent to

Bob has a duty to lose to Jack.

"Grice’s idea is that this ambiguity is due not to an ambiguity in “should” but
to variations in the system of principles in relation to which the sentence is
to be evaluated."

---- enough of an ambiguity for my aunt!

---

"In the relational interpretation the relevant system specifies
principles concerning what that particular subject is to do."

"Just how the
relevant principles are determined needs to be worked out, however, and it
is not clear that this can be done."

"As an example of a principle of reasoning that applies both to practical and
to alethic reasoning, Grice offers a “Principle of Total Evidence.”"

"The idea
is, roughly, that one can infer from P supports Q and P to Q only if one
does not accept some further assumption R and believes that P&R does not
support Q."

"This holds both for the case where

accepting P

is accepting /- P
and for the case where it is

accepting ! P.

"This particular example seems right, although I worry that the real reasoning
in a case like this is the thinking that is used to get the support statements,
which are therefore derivative of that reasoning and not clearly appealed to
by the reasoning."

"But other principles that Grice states seem to yield bizarre results."

"Consider
the principle that an

alethic

disjunction is true just in case one of its disjuncts
is true."

"Grice proposes to extend that to the principle that a disjunction of
any sort is

SATISFACTORY

[apres Kenny, rather than Hare or Hofstadt] just in case one of its disjuncts is satisfactory.

"This
implies that if

Mail these letters!

is satisfactory, so is

Mail these letters or burn them!

------ exactly Ross's paradox, as saved in a Griceian way by Hare.

"Much depends on how practical reasoning is to be understood."

"If practical
reasoning modifies intentions, then it is plausible that there is a common
requirement of consistency."

"Oe’s beliefs and intentions should be consistent
with each other."

"On the other hand, if practical reasoning is reasoning
that modifies desires, then consistency seems less relevant."

"There is nothing
rationally troublesome about having desires that one knows cannot all be
fulfilled in the way that it is rationally troublesome to have intentions that
one knows cannot all be fulfilled."

"Dancy notes a significant difference between

alethic

and practical reason

when there is support for a disjunction of two incompatible possibilities and
no sufficient reason for choosing one rather than the other."

"In the alethic
case, one clearly ought to suspend judgment."

"In the practical case, failure
to decide in such a case can be deeply irrational."

------ I especially love Grice's treatemtn of alethic akrasia, or weakness of the judgment: "you fail to believe that it is raining, when it is".

Harman:

"Grice does not address these last points, but he does emphasize a related
point, namely, that beliefs play an important role in practical reasoning and
that goals play an important role in theoretical reasoning."

"Practical and

alethic

reasoning are deeply intertwined in ways yet to be fully understood."

"In short, Grice’s John Locke Lectures does make several significant contributions to our understanding of reasons and reasoning."

"Some of these contributions have to
do with understanding when we can attribute a particular argument or proof
to someone."

"His discussion of mood or embedded “mode” operators may
have some importance for philosophy of language and linguistics, especially if
these operators can be shown to be connected with more usual classifications
of embedded clauses into tensed that-clauses and tenseless infinitive clauses."

----- and also because it's fun.

"Grice’s proposal to account for relational interpretations of modals in terms
of contextual selection of certain sorts of systems of principles deserives
further study too."

---- and entertainment. "All work and no play makes Jack a dull boy"

"In fact, the book is teeming with interesting ideas." -- and some!

9
Notes

1This issue.

2Compare RogerWertheimer, The Significance of Sense; Meaning, Modality,
and Morality. Ithaca, NY: Cornell University Press, 1972.

3Wertheimer, op. cit. makes a similar proposal