[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Re: The New Method
On Fri, 18 Dec 2009, Jorge Llambías wrote:
In
practice, this has pretty much been a non issue - negation and tags
are taught later, and they are described as being on the outside of
the "whole selbri" if you can kinda catch my drift. I wasn't able to
completely describe my style in the original post. I assure you I'm
doing the best to address the issue.
OK. Arguments can be made both ways as to what is better. Since you
can say things like "lo ca klama", if "ca" is not part of the selbri
you have to specify that it's not just a selbri that can be converted
into a sumti with "lo", but a selbri plus its tag.
I don't expect this to be much of a problem in informal teaching, but
when you write it down more formally you have to consider that it may
end up being used as a reference work in the future, so it's better to
keep things as consistent as possible.
I guess I hadn't thought of {ca klama} as a selbri. I had considered it as a
selbri tcita ({ca}) attached to a selbri ({klama}). But it's possible that
was I was thinking of as selbri is more what has been called "tanru unit", but
that seems even more complicated.
Perhaps the solution is to not claim that {lo ... ku} can convert only selbri
into sumti, but just claim that it can convert selbri into sumti, and leave
other constructs for later, when they've been introduced on their own.
I really haven't had a problem justifying its existence - though I do
admit I have to rely on the tried and true "Just accept that it's
there" argument to get it into people's heads.
In the wave there is a comment:
<Suprano> I don’t see why we need the {be} yet
which is unanswered. When I tried to think of an answer, I realized
that in fact we *don't* need it, except to make "ku" more elidable.
Yes, it would have been possible to make {lo ... ku} convert a bridi (with
empty x1, or maybe even {ke'a}) into a sumti. Under that model, {ku} would be
elidable far less frequently. We'd also get internal sumti in tanru as {mi
zdani do klama} instead of {mi zdani be do [be'o] klama}. I think lojban did
make a good trade off here, but it's hard to explain the reasoning without
mentioning the elidability.
If we had enough students, we could try a lot of variations. Teaching the
terminators as an actual part of the construct, which can in some cases be
omitted if doing so doesn't cause a different parse, does seem to produce much
better results than teaching them as something extra that needs to be added at
odd times.
It might be possible to teach elidability right after {lo ... ku}, but hold
off on {cu} until considerably later (after {NU ... kei}, for instance). It
would still encourage use of terminators when they actually matter, but would
explain why {be ... bei ... be'o} are needed. Something to ponder.
mi'e la xalbo mu'o
--
Adam Lopresto
http://cec.wustl.edu/~adam/
"I have a very firm grasp on reality! I can reach out and strangle it any time!"