misc forwarded mail

[Logical Language Group <lojbab@access.digex.net>]

I don't know how often you are over here on my account, but I saved a message
for you the other day, and it iss still here, so I will forward it to you.

lojbab

>From arktech@clark.net Fri Jul 28 11:09:28 1995
Received: from clark.net (arktech@clark.net [168.143.0.7]) by mail1.access.digex.net (8.6.12/8.6.12) with ESMTP id LAA09376;  for <lojbab@access.digex.net> ; Fri, 28 Jul 1995 11:09:24 -0400
Received: (arktech@localhost) by clark.net (8.6.12/8.6.5) id LAA20808; Fri, 28 Jul 1995 11:09:21 -0400
Date: Fri, 28 Jul 1995 11:09:20 -0400 (EDT)
From: "Jonathan Crawford @ ArkTech" <arktech@clark.net>
To: Logical Language Group <lojbab@access.digex.net>
Subject: Re: Divided by a common language (was: WORD ORIGINS AND USAGE ...)
In-Reply-To: <199507281427.KAA18831@access5.digex.net>
Message-ID: <Pine.SOL.3.91.950728110835.20486A-100000@clark.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: R



On Fri, 28 Jul 1995, Logical Language Group wrote:

> > -- Jonathan Crawford
> >    Britain and the United States: Two countries divided by a common language.
> >      (Oscar Wilde?)
> 
> George Bernard Shaw, actually.
> 
> -- 
> John Cowan		sharing account <lojbab@access.digex.net> for now
> 		e'osai ko sarji la lojban.


Thank you!  I've used that quote a few times, and I've always been 
frustrated by not knowing who to properly attribute it to!

-- Jonathan Crawford


>From PDoudna@aol.com Sat Jul 29 18:11:57 1995
Received: from mail04.mail.aol.com (mail04.mail.aol.com [152.163.172.53]) by mail1.access.digex.net (8.6.12/8.6.12) with ESMTP id SAA08993;  for <lojbab@access.digex.net> ; Sat, 29 Jul 1995 18:11:54 -0400
From: PDoudna@aol.com
Received: by mail04.mail.aol.com
	(1.37.109.16/16.2) id AA150395883; Sat, 29 Jul 1995 18:11:23 -0400
Date: Sat, 29 Jul 1995 18:11:23 -0400
Message-Id: <950729181122_43597879@aol.com>
To: lojbab@access.digex.net
Subject: Re: Lojban
Status: R

This is in response to e-mail from July 25.  I was assuming that you had
access to Lognet 95/1.  That is where the first part of this letter was
printed.  The second part must be interpreted in that context.  The previous
letters in Lognet are not especially relevant.

<<<quote>>>
Your statement about "your" deep structure, suggests that you see deep
structure as personal to individuals, and I don't think that is Chomsky's
claim.  Rather, he claims that each grammatical expression in a language has
*A* deep structure which is more or less invariant with respect to context.
<<<end quote>>>

The terms "surface structure" and "deep structure" are metaphors, which may
be subject to some difference in interpretation.  I see deep structure as
being subjective to some extent.  Basically it is an analysis using a
structure that is not clearly evident in the language that one sees or hears
directly without analysis.  Deep structure is an intellectual tool which can
be useful in interpreting a language string.  Obviously people have been
interpreting language correctly long before they ever heard of deep
structure.  There probably is no unique deep structure (as I would use the
term) any more than there is a unique correct grammatical description of a
given language.

JCB apparently sees "cinta kicmu" as merely a juxtaposition of two words with
a structure of AB.  I see a distinctly different "deep structure" of

(x who-is a-doctor) having-as-a-patient (y who-is a-baby)
(x / B ) R (y / A)

Another possibility is

(x who-is a-doctor) who-is a-patient)
(x / B) / A

Thus I see a surface structure "AB" and two possible deep structures, "(x /
B) R (y / A)", or "(x / B) / A".  I don't mean to imply that my analysis is
the only correct analysis, merely that it is an analysis that makes clear a
type of distinction that we may wish to make.

<<<quote>>>
I am not sure what you mean by "more than one interpretation".  Obviously "le
blanu" (the blue-thing(s)) has many possible interpretations depending on
which blue thing(s) is/are being referred to.  But other than that
distinction, there should be only *one* interpretation for the set of symbols
that is a sentence about "le blanu".
<<<end quote>>>

The preceding examples illustrate what I mean by "more than one
interpretation".  I am not referring to ambiguity of reference.  I am not
referring to lexical ambiguity (an attribute of one word) or to structural
ambiguity (an attribute of at least 3 words).  I am referring to word-pair
ambiguity (an attribute of two words with an unspecified relationship).

<<<quote>>>
"Metaphor", or Lojban "tanru" refers in Loglan discussion to a specific type
of semantic ambiguity, that of the variety of interpretations that can arise
when two words are combined in a modifier/modificand pair.  Thus "blue house"
could mean a house that is entirely blue, inside and out, or one that is
white with a few blue highlights, or (stretching) a green-painted house that
is filled with blue objects, as distinguished from another green-painted
house filled with yellow objects.
<<<end quote>>>

What I called word-pair ambiguity is apparently what you refer to as
"'metaphor' . . .  a type of semantic ambiguity, that of the variety of
interpretations that can arise when two words are combined in a
modifier/modificand pair".

In spite of the variations of precisely what is meant by "blue" when applied
to "house", a blue house is always a house which is blue (in some sense).  On
the other hand, a brick house (except as far-fetched metaphor) is _not_ a
house which is a brick.  Thus in the first case the "deep structure" (as I
use that term) is (x / B) / A and in the second case is (x / B) R (y / A).
 The first case may be imprecise (in the sense that all language is more or
less imprecise) but the second case is strictly ambiguous, in that "R" is not
specified in the surface structure.

<<<quote>>>
. . . Lojban in addition is more explicit about marking what is called
"object raising", perhaps the one area where we recognize that the surface
structure only approximates the deep structure.  . . .
<<<end quote>>>

At least you recognize a possible difference between surface structure and
deep structure.  You just see it's application as more restrictive.

<<<quote>>>
The correct term is grameme, and JCB has accepted that the term has been
misused.  But the actual difference to a non-technical person is rather small
- it is a jargon word that has its own special connotations and denotation
within the Loglan project.

"-eme" as a suffix, means "most basic analytical subunit".  Hence "grameme"
is clearly "most basic analytical subunit of grammar". "lexeme" is only
incorrect when Loglan computer analysis goes beyond assigning word-type to a
word in the language.
<<<end quote>>>

The term "grameme" makes sense in place of JCB's use of "lexeme".  I prefer
it to "form class" or "part of speech".  But I have never seen this term
used.  Does it occur in linguistic literature or is it just a Lojban coinage?
 I don't think the question of jargon usage in unimportant -- unless Loglan
and Lojban people are expected to avoid dictionaries or linguistics books.

      Paul Doudna