[hist-analytic] Grice's Frown

Jlsperanza at aol.com Jlsperanza at aol.com
Fri Feb 27 13:53:58 EST 2009

Baynes was quoting from Reichenbach: 

"A communicative function begins only when 
there are certain rules  established for the use of the terms."

Reichenbach, Experience and Prediction (1933, 1938 -- cited in  OED)

(this, after, as Bayne points out, 'disussing some general
uses  of language, emotional ("Ouch") and communicative
(we all get this I  guess)'). 

In Baynes's exegesis:
>>>communication requires public 
>>>rules in order to have language.

Danny Frederick comments:

>>If that were so, how could  communication 
>>ever get off the ground? 
>>Surely the public  rules become established 
>>through successful attempts at  communication?


>We don't form rules and then communicate, 
>thanks to them. It is the publicity of the rules 
>rather than public "ratification" that is at 
>issue.  I think Reichebach "gets it. If 
>you think not, tell me why.
Well, what he says is, tersely:
   "A communicative function begins only when  
there are certain rules established for the use of the  terms."
One should check if he uses 'public' at all!
This was written in Istanbul in 1933, so we have to be charitable in the  
For example, 'certain'. This, I don't know, could be an ambiguous  
It has nothing to do with 'certain' alla Descartes of course.  It means 
_specific_, or _clearly defined_.
The 'establish' (as per 'established') is a fun one and as I noted in "The  
force of linguistic rules", the OED credits Reichenbach not for this quote but  
for the one on 'descripta' and _conventionalism_. So I would think his idea 
of  _convention_ could prove interesting.

I once wrote a paper on that. I studied Greek forms of 'convention'.  It's 
_thesei_, i.e. the dative of _thesis_, positio. Versus, notably _phusei_,  by 
nature. Some other texts have "nomoi", i.e. by law. Cfr. Poincare, truth by  
convention, etc. 
I notice that the word _rule_ is Romantic (well, Roman), for 'regula' (Die  
Regel in German -- so Reichanbach must be thinking in that). 
Note that he is careful to have it as they beings 'rule for the use',  rather 
than for the _meaning_! By terms, I would think he basically means  
_predicates_. For surely we can think that the 'structural devices' are _given_  to us. 
So basically he seems to be saying that there is some, I go with Baynes  
here, ... and would call it _procedure_. 
By 'communicative function' he must be meaning, as Baynes has it, the  second 
passenger in the bus: Language-As-Communication (Not  Language-as-Expression, 
she is so shy!). For surely we cannot analyse _meaning_  in terms of 
_communication_. That is, as Grice points out ('Meaning' this list's  webpage), 'not 
helpful' (but Stevenson 1944 commits the mistake). 
If _function_ is taken mathematically, I would say that for any predicate  P,
there is a function which gives its meaning.
We can even be extensionalist here. Consider the predicate "... is one of  
the seven wonders of the world". So to know the _meaning_ of that, for practical 
 purposes, is to know the extension.
Now we could use the expression freely,
       "Now _dthat_ should be a wonder of the  world to add to the list of 
If the procedure is 'public' it means that between Addressee and Utterer,  
they _share_ the procedure. It does not mean that who's an addressee can become  
an utterer and vice versa. There may be an agreement that the established  
procedure runs only in one direction ("When she says 'seven wonders of the  
world' she means ..."). 
What a public procedure amounts to then is
-- the 'ratification' as Baynes has it, that the Utterer is _abiding_ by  it
-- the 'ratification' that the Addressee is relying on it to _interpret_  the 
D. Frederick:

>>If that were so, how could communication  
>>ever get off the ground? 
>>Surely the public rules become  established 
>>through successful attempts at communication?

What I found useful here is S. W. Blackburn (at Pembroke then) in  "Spreading 
the word: groundings in the philosophy of language". Not terribly  deep (it 
covers too much grounding!) but a good account of what he calls, I  think, 
one-off communication. I was reminded of this by the 'off' (the ground)  as used 
by Danny Frederick.
But then what _is_ a procedure (or 'rule' if you must)? How _general_ has  it 
got to be.
I once wrote an essay for Rabossi on this (for one of his seminars). I  
called it "Aunt Matilda", for I was having in mind Grice's example:
           "runt"   =  means, 'undersized person'.
This _runt_ does figuratively. Literally it means _undersized_ 'anything'  
('runt of the litter'). 
Grice wants to say that Aunt Matilda _understands_ 'runt' but she has no  
procedure whatsoever to use it _communicatively_. She will 'understand' the  
expression, but the procedure (or rule) at play cannot be understood  
behaviouristically as 'dispositions to utter 'runt' to mean 'undersized person',  because 
she _lacks_ the disposition.
So, when it comes to _figurative_ uses of 'runt', surely we do not need any  
procedure or rule. Basic 'implicature' will do, i.e. Matilda's knowing how to  
build a working-out schema for the metaphor, as it were.
When it comes to the _literal_ use, the 'establishing' is pretty  _arbitrary_ 
if that's the word. If I like anything of David Kellog Lewis's PhD  
"Conventions of Language" (Harvard -- under Quine) is his appeal to  _arbitrary_:
It is _arbitrary_ that 'runt' means RUNT.
That's as far as 'non-natural' (to use Grice's wording) goes. Surely for  
many many many vehicles of meaning, the link is _not_ arbitrary.
Now, I would think that it is in the ability on the part of an addressee to  
grasp the _arbitrariness_ of the linguistic sign (to echo almost literally F. 
de  Saussure), we need a 'public' ratification.
Hence Reichenbach's "Ouch". 
"Natural" sign, no arbitrary procedure needed. 
The transfer is not so much an issue of evolution but 'semantic' freedom.  As 
Grice remarks in his 1-page theory of representation (in WOW, Retrospective  
Epilogue), not all representations need to be _eikonic_; that would make  
conversation pretty _rough_ (not subtle, or, as he'd prefeer, 'genteel').

J. L. Speranza
**************A Good Credit Score is 700 or Above. See yours in just 2 easy 

More information about the hist-analytic mailing list