From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!think.com!ames!elroy.jpl.nasa.gov!swrinde!gatech!darwin.sura.net!jvnc.net!nuscc!mango!john Tue Nov 19 11:09:59 EST 1991
Article 1305 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!think.com!ames!elroy.jpl.nasa.gov!swrinde!gatech!darwin.sura.net!jvnc.net!nuscc!mango!john
>From: john@mango.iss.nus.sg (John Waterworth)
Newsgroups: comp.ai.philosophy
Subject: Re: Is semiotics an "informal logic"?
Message-ID: <1991Nov14.065924.29076@nuscc.nus.sg>
Date: 14 Nov 91 06:59:24 GMT
References: <1991Nov11.024611.12312@nuscc.nus.sg> <LmVaBB1w164w@depsych.Gwinnett.COM>
Sender: usenet@nuscc.nus.sg
Organization: Institute of Systems Science, NUS, Singapore
Lines: 113

In article <LmVaBB1w164w@depsych.Gwinnett.COM> rc@depsych.Gwinnett.COM (Richard Carlson) writes:
>(Michael Moriarty writes:)
>>The most important thing I have learned about semiotics is that the
>>lexeme /semiotic/ has encyclopedic denotations and connotations.  I'm
>>puzzling my way through the labor of building a tree to illustrate
>>this. At this point in the discussion, I believe that a variety of
>>semiotics are in play.  Mathematical semiotics, logic semiotics,
>>linguistic semiotics, and the rest of it.
>>
>>A healthy semiosis makes me move to another part of the discussion.  I
>>wonder what various discussants mean by the following lexemes:
>>
>>1. /rhetoric/ = ?               4. /monotonic/ = ?
>>
>>2. /logic/ = ?                  5. /code-decode/ = ?
>>
>>3. /informal logic/ = ?         6. /inference/ = ?
>
>I agree.  Especially for column 2 (items 4,5 and 6).  Even a
>reference to an easily available text would help since, for all I
>know, these terms and their associated notions may be well-known
>in some discipline with which the ordinary educated layperson with
>an interest in AI should be familiar.
>
>Although I've perused it carefully, I can't really understand Mr.
>Waterworth's post:
>
Please, call me John.

>john@mango.iss.nus.sg (John Waterworth) writes:
    >> Semiotics seems now like a hangover from a bygone age, since it is
>> basically a code-decode approach elevated to the level of a general, if
>> vaguely stated, theory of communication. These days most linguists would
>> take the line that code-decoding processes are subservient to
>> inferential processes. The latter are autonomous and don't need, though
>> they work better with, the former. (not that I can speak for most
>> linguists, or indeed any linguists, but this is the trend in several
>> recent and paragmatically-oriented works). The code-decode and, hence,
>> the semiotic model depend on the idea of shared knowledge, usually
>> called mutual knowledge. But in reality, there is little, sometimes no,
>> mutual knowledge. Communication is risky and error-prone, often largely
>> erroneous. Semiotics is intellectually bankrupt, it simply cannot say
>> anything significant about language use, which is why there are no
>> semiotic laws.             
>
>Since I view each word in an utterance as a bundle of semes and
>therefore inherently a fuzzy construct, I believe it is an
>"empirical" or observational fact rather than a necessary
>postulate that there are usually enough shared semes in P's and
>O's lexicon to make communication approximately accurate.  But I
>think usually there are.  What I don't understand is how the
>inference can _replace_ the coding and decoding (if these terms
>mean what they seem to mean on the surface) when "logically" (or
>common-sensically) an inference could be made or communicated only
>_after_ the "meanings" of the terms have been "decoded" -- or
>whatever.
>
>Perhaps thinking of the "signified" for any word ("signifier") as
>a collection of semantic elements of greater or lesser
>articulation and completeness depending on the sophistication of
>the speaker is so different from expecting each "word" to have
>some definite and delimited meaning for everyone (like a kind of
>phenomenological essence, except that linguists and logical
>analytical philosophers in the Anglo-American tradition wouldn't
>be familiar with Husserl's theories and so wouldn't be aware that
>they had exemplified some of his notions in their unconscious
>"pre-understanding"), that I just literally can't understand what
>the Anglo-American language theorists are doing or saying.
>
>--
>Richard Carlson        |    rc@depsych.gwinnett.com
>Midtown Medical Center |    gatech!emory!gwinnett!depsych!rc
>Atlanta, Georgia       |
>(404) 881-6877         |

The code-decode model is simply the idea that, you know, I have this
notion in my mind which I code into some linguistic representation which
I convey to you, in speech or writing, and which you then decode to get
at the notion I had in the first place. There was a lot of work on the
kinds of mutual knowledge needed for this to happen. But then it seemed
that the adjustments to the model became so complex that it was better
abandoned for a more parsimonious account. My point was that semiotics
is the original simple code-decode model taken to extremes, when the
model doesn't even work for simple things like me making a point about
one colleague to another by raising my eyebrows and moving my head
slightly. I would have thought that a few hours (minutes) reading on
this network would be enough to show how little knowledge is shared that
can be used to infer the meaning of the language used. Of course, the
more technical the writing, the more careful we are to reduce ambiguity,
but that's not normal communication. Note also thast semiotics does not
seem to strive for such clarity, because it doesn't recognise this key
problem.

Inference can be applied to all kinds of cues (like the height of my
eyebrows). The inference-ostention approach emphasises that our
inferencing is always guided by considerations of relevance and context.
This is partially motivated by a desire to account for irony, sarcasm,
etc.. where the coded form deliberately belies the communicated meaning.
The best reference to this, with lots of good examples, is 'Relevance'
by Dan Sperber and Dierdre Wilson (Oxford: Blackwell: 1986?). It's a key
work for anyone interested in communication, NLP, AI etc.. Read it and
you will lose your interest in semiotics.

The last paragraph os my mail did not make sense (which may mean that
people couldn't understand it) because for the first time ever I managed
to use this ^&*(ing editor without leaving a pile of junk under the
stuff I meant to send.

-- 
   John Waterworth (john@iss.nus.sg)
   Institute of Systems Science                           Ph: +65 772-3111 
   National University of Singapore                      FAX: +65 778-2571
   Singapore 0511                               


