Grice and Carnap on "central" meaning

By Roger Bishop Jones for The City of Eternal Truth
.
This is a follow up to a post by J.L.Speranza at The Grice
Club, entitled "Salva Veritate"
http://griceclub.blogspot.com/2010/04/salva-veritate.html
.
This is mainly concerned with strand 5 of the "Retrospective
Epilogue" in Grice's "Studies in the Way of Words".
Here Grice is looking for some "central" aspect of
significance (meaning) of expressions.
Grice is talking about ordinary language, and Carnap has
conceded in the analytic/synthetic debate that a precise
account of the semantics of ordinary language may not be
possible. Furthermore, Carnap is concerned mainly with the
advancement of scientific method in philosophy and science
through the adoption of formal language.
.
On the road to "The City of Eternal Truth" which we are
navigating through a dialogue between Grice* and Carnap* as
interpreted by Speranza and Jones, this topic confronts the
greatest danger of disruptive discord.
Surely Carnap and Grice would here either be unable to
reconcile their attitudes toward ordinary language, or would
simply find their areas of interest disjoint and fall into
silence?
.
Well, on behalf of Carnap* I'm going to suggest that we need
not despair.
Insofar as I am able to discern the purpose of Grice's
search for a central concept of significance, I think it
plausible that this is the very same purpose which is
addressed by the adoption of formal languages.
.
What is that purpose?
That purpose is to make more rigorous the conduct of
deductive reasoning.
.
What does that involve (centrally!)?
It involves devising languages with deductive systems which
are SOUND.
A sound inference is one in which "whenever" the premises
are true so are the conclusions.
The "whenever" here expresses universal quantification over
all possible "conditions" (possible worlds, situations,
states-of-affairs).
.
What we see from this is, that if we are concerned about
rational discourse, and want to be able to reason
deductively, then for that purpose the key element in the
significance of sentences is their truth conditions.
In default of definite truth conditions, the notion of sound
deduction fails.
.
Now Carnap* I suggest, will not have much to say about
"centrality" itself unless Grice can make its meaning more
definite, but he has here a candidate for what he might call
an explication of "central" which has some resonance with
what Grice is saying.
.
There is one thing which Grice mentions, almost as an aside,
as necessary to both his logically independent candidates
for centrality, and that is "truth conditions".
Since he also requires that the central should be simple, it
follows that the central notion of significance if it does
include truth conditions, should contain nothing else, all
else should, in his words "cluster around" this central
core.
.
Thus we may interpret Grice's search for a central core of
significance in ordinary language as being addressed by
Carnap's formal techniques, which are intended to give
languages in which the significance is just that central core
of truth conditions.
.
It is arguable that however significant implicature and other
non-central significance might be in conversational contexts
it is out of place in published scientific results which
should rely exclusively on direct explicit statement, but
that the key requirements on signification which support
sound deduction (viz truth conditions) are equally desirable
in conversational contexts, and thus form a part, and hence
for the sake of the required simplicity, the whole of the
centre.
.
In my first response to "Salva Veritate" there was a lot of
detailed discussion of the points mentioned by JLS, I shall
have to look back and consider whether to follow up with
some of that detail.
.
RBJ