Date: Fri, 23 Dec 1994 03:25:43 +0000
From: ucleaar <ucleaar@UCL.AC.UK>
Subject: Re: reply: (1) veridicality; (2) plurality
In-reply-to: (Your message of Thu, 22 Dec 94 15:24:47 MST.)
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Reply-to: ucleaar <ucleaar@UCL.AC.UK>
Message-id: <01HKYVCM5SW20000X1@FIPORT.FUNET.FI>
Content-transfer-encoding: 7BIT
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Length: 6083
Lines: 130

Chris:
> "Where the sons raise meat" (when heard, of course, not read) is ambiguous,
> since it could be "Where the sun's rays meet".  However, given the right
> build-up to the phrase, the listener will only hear it one way or the other,
> and it seems funny only later when the ambiguity is pointed out.  In this
> case, the contextual disambiguation clearly happens, at least in part,
> before any grammatical derivation can occur.
> What is the grammatically determined meaning of [sUnzreizmit]?  The question
> is meaningless, out of context.

That phonetic event you notate above has no grammatically-determined
meaning. If it is a realization/representation of the sentence
"Where the sun's rays meet" its gram-det meaning will be "at the point
at which the rays of the sun meet" [understand the paraphrase to
constitute a representation of gram-det meaning, just for convenience],
while if it is a realization/representation of "Where the sons raise
meat" then the gram-det meaning is "at the place at which the sons
rear flesh".
The disambiguation you speak of is disambiguation of syntax, not of
meaning.

> >Now, I ask you: Where do you think the information "at some time
> >prior to the utterance 'kissed', there were some event(s) of kissing,
> >the kisser being Sophy and the kissee being Edgar" fits in? Does
> >this have anything to do with the meaning of the utterance?
>
> Where does "on a continuing basis, there is some process of 'raising' going
> on, the raisers being the sons and the raisee being meat" fit in?  Sometime
> after the point where we've determined from context that [reiz] is a verb,
> not a plural noun.  This requires analysis of the context.

I agree with this.

> >Next, where in the grammar are the rules that tell you "Given the
> >sentence 'Sophy kissed Edgar', derive the proposition that at some
> >time prior to the utterance 'kissed', there were some event(s) of
> >kissing, the kisser being Sophy and the kissee being Edgar'"? Had
> >I enough time and intelligence I'd write out the grammar of
> >English and point out the relevant rules to you.
>
> Where in *what* grammar?

Where in, say, the grammar of your idiolect.

> "English grammar"s are useful inventions of linguists, and they are
> not all identical.

Why are they inventions of linguists?  Certainly, they are not all
identical: the grammar of each idiolect differs from every other,
to a greater or lesser degree. That doesn't undermine the notion
of grammar.

> An English Grammar would not be
> useful if it contained a hard and fast derivation like that for "Sophy
> kicked the bucket", that didn't consider the pragmatics *before* reaching
> the final proposition of "Sophy died" or "Sophy hit a container with her
 foot".

The two propositions you mention are both grammatically-determined
meanings --- one of a sentence with an idiom, and one of a sentence
without an idiom. Each sentence has its GDM irrespective of context.
Claims about GDM are not claims about processing, or the sequence in
which things are considered. You are right that we need to consider
context before we work out what sentence someone has just uttered.
All I am saying is that is that part of the structure of the sentence
is a semantic structure.

> >What have I said to indicate that I hold semantics in especial
> >reverence (or pragmatics in contempt)?
>
> Maybe reverence isn't the word, but primacy.  I read you as claiming that
> semantics is prior to pragmatics, in some schematic of how the brain
> processes langauge, and I don't think that's necessarily so.  The two things
> are intertwined.

I don't think this, and doubt I said anything like it. I don't think
semantics is entirely prior. As for primacy, I give it primacy over
non-grammar in the context of Lojban, since Lojban is at the stage
of grammar-development.

> >> If the 'grammatical meaning' is not influenced by the context, then
> >> it's an artificial construct that bears little relation to real human
> >> communication.
> >
> >What are your arguments? I say the grammar says "'twenty' means 20",
> >and that a general theory of communication, such as Sperber & Wilson's
> >Relevance Theory, can explain how, if the grammar says what I say
> >it does, in a given context the utterance "twenty" can mean
> >'approximately 20', or 'give me another banana'.
>
> If Sperber and Wilson merely say that "twenty" can mean any of these things,
> then I have no argument with them.  But if they claim there's a fixed
> English grammar, with fixed rules like chess, that derive 20 from twenty,
> and THEN another process gets 'give me another banana' from 20 and the
> context, I disagree.  It doesn't explain the pun, or homonyms in general.

They are saying what you disagree with. You misunderstand the nature
of the pun and of homonymy: these are evidence of the role in disambiguating
a phonological structure which may represent a number of alternative
grammatical structures. [That shouldn't happen with Lojban, which has
no homonymy.]

> >What can I say? The only way to resolve the issue is to go and do
> >some linguistics based on those assumptions.
>
> Your assumption leads to the conclusion that people can't learn "le" vs.
> "lo" without resort to onerous social stigma.  My assumption leads to the
> conclusion that they can naturally become part of the rules of a language.
> What better experiment than Lojban?

We still haven't agreed on what the "le" vs "lo" rule is. I hold
that it should be a rule governing explicatures. As such, specificity
will turn out to be more important than veridicality, I would
predict, and in the absence of social stigma I would expect "lo"
to be used as -specific, but sometimes figuratively.

So yes, Lojban can be an experiment: does it get used non-literally,
in the absence of social stigma attached to non-literality. I predict
yes, and you apparently predict no.

> What is Crimbo?

A dialect word for Christmas. I like it because it feels apter.
I wasn't terribly keen on Jorge's xisnun[birth]. Maybe I'd use "lo krimbo".

Ko se zdile lo krimbo!

---
And