NetNews Usenet Archive 1992 #26/NN_1992_26.iso/spool/bit/listserv/csgl/1475

Text File | 1992-11-13 | 6.9 KB | 136 lines

Comments: Gated by [e-mail]
Path: sparky!uunet!paladin.american.edu!auvm!CCB.BBN.COM!BNEVIN
Message-ID: <[e-mail]>
Newsgroups: bit.listserv.csg-l
Date:         Fri, 13 Nov 1992 09:55:14 EST
Sender:       "Control Systems Group Network (CSGnet)" <[e-mail]>
From:         "Bruce E. Nevin" <[e-mail]>
Subject:      Penni: information
Lines: 125

[From: Bruce Nevin (Fri 921113 09:52:14)]

(penni sibun 921112.1600) --

Thanks for posting the quote from Alice.  It saved me typing it!  (You
must have an on-line copy.  Is that available someplace?)

>    [From: Bruce Nevin (Wed 921111 10:59:45)]

>    No, I am not saying that meaning is everything, I am saying that each
>    thing "has" specific meanings.  Understand that every "thing" is a
>    perception, and each meaning is another perception.  I am not saying
>    that a given thing/perception "has property x" i.e. has meaning in some
>    vague, generalized sense of "the property of meaning." I am saying that
>    each given thing-perception has associated with it specific other
>    thing/perceptions, differently for each

> as stated, this is either circular or leads to infinite regress.

It is not circular in the usual sense invoked in discussions of formal
semantics, because we have stepped out of the realm of words and
symbols.  It is circular only to the extent that the associativity of
perceptions one with another turns out to be circular, and that simply
is as it is.  Similarly, there is no infinite regress unless associative
memory is a recursive process with self reference, and that seems prima
facie doubtful.  Even if such recursion were found, possibly there are
inhibiting processes that come into play when some associatively evoked
perception turns out to be a controlled perception, and possibly failure
of that underlies some kinds of mental illness, but that's just off the
cuff speculation.

> i think perhaps you're trying to say that meaning inheres in the
> structure of relations bet. perceptions.  that's fine, but then i
> wonder why it's necessary to call it ``meaning'' rather than
> ``structure'' or ``assocations'' or something like that.

No.  I am saying that the ensemble of associated perceptions IS what we
call meaning.  To reify the word "meaning" is a mistake.  To say that
perceptions are meaningful to us suggests (because of the way English
works) that there is some substance or essence called "meaning" that
"inheres in" the perceptions.  But there are only the perceptions
themselves.  To say that a perception is significant for us is only to
say that we are controlling that perception--that is, that we maintain
in memory a reference (preferred amount) for that perception, that a
comparator function within us produces an error signal when the
perception and the reference differ, and that we do whatever it takes to
reduce that error (assuming that conflict does not conceal it, and
assuming that higher level control does not override it, resulting in
willing suspension of control despite error).

A structure of relations between perceptions creates information.
In particular, the structure of relations between elements of language
(language-perceptions) constitutes linguistic information in utterances.
For linguistic information, I refer you to:

Z. Harris.  1990.  _Language and Information_.  New York:  Columbia
        University Press.

-----.  1992.  _A Theory of Language and Information_.  Oxford:
        Clarendon Press.

-----, M. Gottfried, T. Ryckman, et al.  1989.  _The Form of Information
        in Science: analysis of an immunology sublanguage._  Boston
        Studies in the Philosophy of Science.  Dordrecht: D. Reidel.

We also talked about linguistic information here a year or more ago.
One could dig some stuff out of CSG archives given time and motivation.

>i think you are, roughly, equating words w/ verbal perceptions.

There is nothing but perceptions available to us.  The distinction I
make between verbal and nonverbal perceptions has to do with convention,
arbitrariness, and participation in constituting linguistic information.

>                            i don't see your proposal as any different
>from the venerable view of language as encoding/decoding meanings from
>my head to yours and vice versa

In a code, there is an external table of equivalences between elements
of the code and elements of language.  Linguistic information is in a
coded message, modulo substitutions through the table, precisely to the
extent that it is in the decoded message.  (Perhaps there is some
legitimate use of "code" that does not rest upon language, but I have
yet to encounter it.  I believe that any other use of the concept
of encoding is in fact only metaphor.)

There is no external table of equivalences for language.  Nor is there
anything with which the elements of language can be interconverted by
such a table, were it to exist.  The closest analog is the association
of perceptions with language-perceptions within a person.  This is not
stable and is many-many rather than one-one.  Salience in the associated
perceptions (meanings) depends upon control of those perceptions on the
occasion, not upon something like a grammar in which things are looked
up and correlated.  I should think this view would be very congenial
with yours, in at least some respects.

To be sure, there are constraints.  Language is a matter of learned
convention (language-particular) within the general characteristics
of perceptual control (language universals).  The discussion with Avery
about agreements, membership, consensual reality, and so on bears
directly on these constraints.  That discussion suggests how there can
be stabilities without tables for encoding.

Roy Harris has some useful things to say about de Saussure.  He seems
deeply vested in an "encoding" conception of language.  He rails a lot
at other linguists.  His review of a book by Zellig Harris a year or so
ago was exceptionally uncomprehending and self serving, merely a forum
for riding his hobby yet again.

In the two social situations that you described, you should be aware
that your perception of what you are doing is quite likely different
from the perceptions that other participants have.  In the first
situation in particular you are deliberately and probably effectively
dissembling.  In the second situation, you and your flirtation partner
are very much involved with imagined perceptions.  Is it necessary to
say that there are many misunderstandings in this process?  That it is
only a beginning, an assertion of availability and interest, and that it
goes nowhere without on subsequent occasions spending some quality time
(Ed's definition of the term) together?  We call this evocation of
perceptions and negotiation of agreements information, of course, but
that is an informal, metaphoric use of the word, not the technical sense
of linguistic information that I am using.

        Bruce
        [e-mail]