[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
Krawn:
> --- In ceqli@yahoogroups.com, "And Rosta" <a.rosta@v...> wrote:
> > [Making good on my pledge to reply....]
> >
> > Garrett to Ceqli list:
> > > A while back on the conlang list I compiled a list
> > > of meanings for "the" and "a". It might be useful for you to have
> > > this list if you're intending on defining articles precisely for
> Ceqli.
>
> --SNIP--
>
> > This seems pretty reasonable. XS makes do with two articles, which,
> > with the addition of quantifiers, covers all meanings we know of.
> > That'd be a minimal system, which could be augmented for
> > demonstrativeness (kra) or definiteness (te, sao).
> >
>
> And, I skimmed through your post and was most interested in your
> summary. Two articles with quantifiers seems like a nice-sized
> system :-) .
>
> As I recall, we were thinking for Ceqli of: a definite article; a
> noun phrase marker that serves as an indefinite article when one is
> desired; a proper name marker; and an article for something that is
> definite to the speaker but not to the listener. (I hope I remembered
> them all.) What do you think of such a system?
I'm not sure what "indefinite article" means here. I'd also
suggest that a zero article should be neutral between all the
other articles; that is, it is left to the hearer to guess
which article the speaker intends.
It seems good to have an article for descriptive names like The
Daily Record, but for other names I'm not so sure. Take names of
species: you might want to be able to say "is a Homo Sapiens" (i.e.
a verbal use), or "a Homo Sapiens", or "the Homo Sapiens", so marking
something as a name doesn't necessarily mean it has to function as an
articleless noun.
As for the rest of the system, I'm not sure it covers the logical
bases, and the options for definiteness could be elaborated if
you chose.
For comparison, let me list the articles in my own conlang:
1. the set of all X
2. X-kind (viewed as an individual X)
3. descriptive name
4. demonstrative (pointed to)
5. unspecified (hearer guesses which other article is most
appropriate; speaker uses it out of laziness) -- I suggest
this as the function of Ceqli zero article.
6. ordinary existential quantifier [not syntactically an article
in my conlang]
7. special existential quantifier that has scope over the whole
text ("Once upon a time there was *a poor woodcutter*")
8. three articles inspired by empathetic use of reflexives in
English -- I won't elaborate on these here
9. specific ("a certain X")
Specific is then subclassified according to varieties of
definiteness:
9a. specific (equivalent to Lojban/XS "le")
9b. specific, not necessarily speaker-identifiable
("there's *some guy* on the phone asking for you")
9c. specific, speaker-identifiable, not necessarily known
to speaker (= one use of English indefinites)
9d. specific, speaker-identifiable, hearer-known, not
necessarily hearer-identifiable (= one use of English
indefinites). Used when the hearer has knowledge of
the referent, but it is not relevant for the speaker
to provide sufficient information for the hearer to
identify the referent.
9e. specific, speaker-identifiable, hearer-identifiable.
(= one use of English definites)
9f. specific, speaker-identifiable, hearer-identifiable,
discourse-new. (= one use of English definites)
9g. specific, speaker-identifiable, hearer-identifiable,
discourse-old. (= one use of English definites)
'Hearer-known' means the hearer has a mental file on the
referent. 'Hearer-identifiable' means the hearer can
identify which file to access. 'Discourse-new' means
the file is not already active/open. 'Discourse-old'
means the file is already active/open.
I'd have thought that Ceqli ought to have ways of expressing
the following, whether or not the job is done by articles:
1. the set of all X
2. X-kind (viewed as an individual X)
3. descriptive name
4. demonstrative (pointed to)
6. ordinary existential quantifier [not syntactically an article
in my conlang]
7. special existential quantifier that has scope over the whole
text ("Once upon a time there was *a poor woodcutter*")
9. specific ("a certain X")
As for definiteness, that's a useful optional extra. A compromise
might be to have four:
9a +/-s-identifiable +/-h-identifiable
9b -s-identifiable +/-h-identifiable
9c/9d +s-identifiable +/-h-identifiable
9e/9f/9g +s-identifiable +h-identifiable
--And.