Re: TECH: PROPOSED GRAMMAR CHANGE 38: lambda via new selma'o CEhU

[ucleaar <ucleaar@UCL.AC.UK>]

> > >> It is NOT a "nice" idea because it changes the meaning of a fairly
> >> stable word of the language (ke'a) into something potentially ambiguous
> >> in meaning.
> >There is no potential ambiguity. Or, if there is, noone has yet spotted
> >it and pointed it out. The meaning of kea is changed only in that it
> >would become usable in places it wasn't usable in before.
> I am still unconvinced.  ke'a is still a relative pronoun, and if it
> appears ANYWHERE in a relative clause then it refers to the relativized
> sumti.

No, this is wrong. kea "refers to" the sumti relativized by the relative
clause in the outermost bridi of which kea is in or exportable to the
prenex of. That way there can be relative clauses within relative clauses
and relative clauses within ka clauses and ka clauses within relative
clauses and ka clauses within ka clauses. If you think ambiguity is
still possible, try giving an ambiguous example.

> That is NOT what it means if the usage is in an abstraction, and I see
> absolutely no similarity in meaning other than both are KOhA variables
> that refer to something specific (hm maybe I should be careful about
> that "specific" even.

Relative clauses, sumti tails and ka with one L-variable are (or
can reasonably be seen as) 1 place predicates, with the function
of kea being keeping one argument unfilled.

> >> The only advantage cited for combining the two meanings that I have seen
> >> is to save a cmavo.
> >This is a big advantage.
> No it isn't. One cmavo more or less in several hundred is not a big
> savings.

It is big in absolute terms, but not in relative terms. A saving that
is big in relative terms would be ginormous in absolute terms.

> >I see no problem to entering CVVV space.
> I do.

That's not a good reason for not entering it. The less that
design decisions are subject to the whims and delusions of
individuals, the better the decisions will end up being.

> >The appeal of Chris's suggestion
> >is that it satisfies both camps - Jorge & I cd use kea & ignore
> >keae & keaa, while you & John cd use keae and keaa & ignore kea.
> That is not a "solution".  We are simply using two different words
> to mean exactly the same thing, or in this case, 3 words to cover
> 2 meanings. If you are concerned about wasting an extra cmavo, why
> aren't you concerned about the extra one here?

Because in effect, Chris's solution is "let's have two different
dialects, one with kea, and the other with keae and keaa, and by
choosing backwater cmavo for the latter dialect, we indicate that
the former is more sensible".

> >You cannot legitimately object that this adds 2 extra cmavo, since
> >you're opposed to the solution that adds no extra cmavo.
> I can and do.  Because my objection to the original proposal is that it
> unnecessarily packs meanings that I see as unrelated.  Adding cmavo
> does NOT remove the criticism of your usage of ke'a.  Furthermore it
> ruins MY usages of ke'a, since they might be taken to mean something
> other than what I intened them to.

I doubt that it ruins your usages of kea; did you have any examples
in mind?

> I'm sorry, but I will not consider packing the two meanings into ke'a.

Okay; it's not something to make a fuss about. Since there's no risk
of ambiguity, anyone who wants to can use kea as the lambda-variable,
and the result will be grammatical and intelligible. Those who want
to use cue (or whatever the official cmavo is) can too.

I also propose, following a suggestion by you, that kea be usable
within to...toi as a reference to, by default, the outermost bridi
of the sentence containing to...toi:

  Sophy, as I'm sure you know, is married.
  la sofi n to kea zou mi birti kuau do djuno kea toi cu ca speni

  Sophy is, I fervently believe, unsurpassedly beautiful
  la sofi n cu to mi carmi krici kea toi traji leka kea melbi

coo, mie And