Paul Dennett wrote:
> Don Blaheta wrote:
> >
> > Quoth Paul Dennett:
> > > Kapitano Eglefino wrote:
> > > >           Bai hund nigr du cxas rapid na kat blank
> > >
> > > You have obviously made an assumption somewhere that adjectives should
> > > follow the head noun.
> >
> > Yes, I prefer that this assumption be made (or the reverse, that
> > adjectives precede the nouns).  It makes it more readable (ok, to _me_
> > it does anyway) to be able to expect one or the other...
> I like the assumption, too.  I only drew attention to it because with
> Esperanto as a base it didn't have to be that way.  There could have been one
> NP with ADJ N and the other with N ADJ.  Once you get rid of p.o.s. markers
> you are storing up trouble by allowing both.

Another point:  If adjectives follow nouns, they are more
parallel to their sentence equivalents, that is:

Hund nigr > Hund est nigr
> >
> > >         Bai hund nigr du cxas rapid na mult kat blank jam.
> >
> > Eh?  You appear to be placing the adverb at the end of the sentence
> > here...
> Yep.  It could have been 'hieraw' or 'antaw kvin jar', and it didn't have to
> be at the end.  I took it out of the way of the verb to show that it wasn't
> any more or less connected to the VP than an adverb of place would have been,
> e.g.:

Fooling around with Ceqli I came to realize that the
adj/noun relationship is not the same as the adv/verb.
I choose to regard the adverb as a _sentence_ or
_phrase_ modifier.  As such, it can logically move

>          'Bai hund nigr du cxas rapid na mult kat blank en gxarden'.
> > I _really like_ the idea of grouping all modifiers with their
> > head, it makes it oh-so-much-easier to parse it.
> That's fine too, in this scheme.
> >  Adverbs which modify
> > adjectives would, of course, follow those adjectives, so this wouldn't
> > solve the PLGS (pretty little girls' school--read the FAQ) problem,
> > but it _would_ solve the ambiguity in e.g. the sentence "Edward fought
> > the knight with the sword": one version would translate as (pseudo)
> >         Bai Edward du did fight with sword na knight
> > and the other as
> >         Bai Edward du did fight na knight with sword
> I see what you mean, but that's hardly a fatal flaw. Doesn't this ambiguity
> exist in other languages, too?  My point was that tense should be indicated by
> adverb phrases, not by little auxiliary markers parked in front of or behind
> the verb, where WENSA types like us are likely to use them like ordinary
> flexions.

Another way of handling this would be to prohibit 'knight
with sword' expressions and replace them with the equivalent
of 'knight who has (uses) sword'

> Oh, yeah.  I forgot to say that one of the reasons I like the idea of an
> isolating grammar is it should be possible to import all manner of roots from
> all kind of languages without distorting them.

Yes.  I've found that you can find just about any word in any shape
to borrow if you look long enough.  I'd say go Z one better.  To avoid
homonyms he reached for roots from Germanic and Slavic.  No reason not
to reach further.

> This may be wishful thinking
> on my part, but there you go.  But you have to start somewhere, and it makes
> sense to me, anyway, to start with what I know.  I have yet to see a
> convincing way of constructing a vocabulary in which all language groups are
> proportionately represented.  What is the point of bringing in, say, three
> Zulu words, eleven Thai words or whatever?  And how is anyone going to decide
> what they should be?  Until someone comes up with a better idea I prefer to
> stick with a European based vocab and let usage dictate what is imported from
> other language groups.
> >

Yes. Proportionalism is pretty silly, really.  You can come up with
any number of schemes which yield very different results.  I thought
once about doing it by _family_.  That is, choose the biggest languages
from Indo-European, Dravidian, Bantu, Afro-Asiatic, Sino-Tibetan, etc.
An idea no better and no worse than all the others.

> WENSA isn't the target, it's the starting point.  And I thought we weren't
> having articles.  Actually the more I think about number in the NP, it is only
> in *definite* NPs that it matters anyway.  If you have one child and someone
> asks you 'Do you have children?' they'll think you're a bit odd if you say
> 'No, I've only got one child.'  The use of the numeral or some other
> determiner makes the number, well, determinable.

Yes.  I believe 'hund' should be indeterminate in number, as it
would be in Chinese.

Another thing to think about is how you might easily
incorporate self-isolating morphemes.  It is important
to distinguish sukero and suk-ero.

Rex F. May
"That government is best which governs somebody else."
Visit my 'Tceqli' artificial language site at: