Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site dciem.UUCP
Path: utzoo!dciem!mmt
From: mmt@dciem.UUCP (Martin Taylor)
Newsgroups: net.nlang
Subject: Re: Re: Credibility
Message-ID: <1610@dciem.UUCP>
Date: Mon, 1-Jul-85 14:25:56 EDT
Article-I.D.: dciem.1610
Posted: Mon Jul  1 14:25:56 1985
Date-Received: Mon, 1-Jul-85 16:16:32 EDT
References: <271@sri-arpa.ARPA> 
Reply-To: mmt@dciem.UUCP (Martin Taylor)
Organization: D.C.I.E.M., Toronto, Canada
Lines: 66
Summary: 


>I am constantly finding people who take arms against any change of the
>written language.  I have always felt that the written language is
>primarily a representation of the spoken one.  There is an easy test to
>tell you how you think about it.  Just decide which of the following
>sentences is "correct".
>
>        Type a "L".   (or)  Type an "L".
>
>I know people who argue strongly for the first form because "L" does
>not begin with a vowel.

"I have always felt that the written language is primarily a representation
of the spoken one."  Many respectable authorities would agree, but I
think it is a dangerous misconception.  Spoken language has primarily
evolved in the context of conversation, with frequent opportunities
for feedback and error correction, as well as continuous inclusion of
out-of-channel signals (e.g. body language).  Written language has evolved
to handle the storage and transmission of ideas to distant places and
times.  If one theory of the origin of writing is correct, there was
no initial connection with spoken language: writing carried the tallies
of goods being transported.  The carrier might be able to describe the
contents of the wagon by looking at the tokens, but he could equally
well do so by looking in the wagon.  Would one then say that the wagon
contents represented spoken language?

Many very long-lived writing systems have only a tenuous connection with
the sounds of language.  Chinese (~5-6000 yr) and hieroglyphics (~4000 yr)
both suggest the sounds of the desired words, sometimes, but they make
no provision for computing the sounds of unknown words.  One can read
aloud texts written in either, but this does not mean that such reading
is the primary function of the writing.  Quasi-phonetic writing is a
relatively new (~2-3000 yr) entry into the field of writing.  It is not
a "perfection" of writing methods that we can now, in some languages,
deduce the sound of an unknown word from its written form.  It is a
convenience to do so, since we can communicate with someone at a distance
and then later talk to them about the same novel topic.

The structure of written language differs from that of spoken language
in ways that are quite dramatic if you look/listen closely.  Listen not
to academics, but to everyday conversation (or even televised press
conferences).  How often do you hear clearly constructed sentences?
How often do you hear isolated phrases, or incomplete structures that
run into one another for a minute or two without pause?  Could you read
such stuff?  Probably, with difficulty, but in conversation there is
no problem because of the out-of-channel information and because of
the various opportunities for feedback.

Others have commented on the value of spelling homophones differently
when the etymology is thereby clarified.  As with phonetic spelling
it is a convenience, not a necessity.  There are doubtless many ways
English spelling could be improved, by which I mean made to convey
more rapidly and easily the writer's intentions, or to be easy to learn.
Phonetization is technically impossible because of horrendous dialect
variations, but internal consistency could be improved.  I'd hate to
try to do it, and I would be very concerned about the problems of reading
out-dated materials if many spellings changed.  But in the long run
it will happen, I suppose, and it will probably be for the good.

But don't make the mistake of thinking writing is just a transcription
of what might otherwise be speech.
-- 

Martin Taylor
{allegra,linus,ihnp4,floyd,ubc-vision}!utzoo!dciem!mmt
{uw-beaver,qucis,watmath}!utcsri!dciem!mmt