[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
--- Martin Bays <mbays@hidden.email> wrote:
> On Tue, 14 Oct 2003, Jorge Llamb�as wrote:
> > Why is {co'a na'o broda} "starting to typically broda" allowed, but
> > not {na'o co'a broda} "typically starting to broda"? How does one
> > reflect semantic structure more than the other?
>
> Sorry? They're both grammatical in the current grammar, but {co'a na'o
> broda} falls foul of what CLL says about ZAhOs coming last. Is that
> what you meant?
That too, but I meant to give a different example, sorry.
Why is {ze'u co'a broda} "a long start of broda" allowed, but
not {co'a ze'u broda} "start of a long broda"?
{co'a ze'u broda} will be accepted by the parser, but it will
parse as {co'aku ze'u broda}.
> And what that reflects is the tense model which I thought was the
> official one (though I'm less sure now) and which is at least a
> plausible one - you define (considering just time for simplicity) an
> interval, then a subset of that interval with TAhE and PAroi, and then
> how that subset relates to the event of the bridi(/seltcita sumti,
> preferably, though apparently that does contradict CLL) with ZAhO and
> PAre'u.
But there is a contradiction there. You want {ba ba'o broda}
to locate the aftermath of broda in the future (which I agree
with) but {ba ba'o ko'a broda} will not take you to some future
interval where you locate the aftermath of ko'a. How do you
consiliate {ba ko'a broda} and {ba'o ko'a broda} with
{ba ba'o ko'a broda}, compared to {ba broda}, {ba'o broda}
and {ba ba'o broda}?
> If you want to use ZAhO in a different way (as in your translation of
> {co'a na'o broda}), then that's cool - as long as it's part of a general
> scheme which gives meaning to this class of tenses ({ZAhO TAhE}, or
> preferably {ZAhO (tense as above)}). That's all I'm saying, and that's
> what I don't see happening, and is what I was clumsily attempting with my
> JOI1 thingy.
The general scheme is this: each tag unit acts on (has scope over)
everything that follows. This includes the imaginary journey for
PUs: {pu pu broda} means that {le nu le nu broda cu purci cu purci}.
{pu ba broda} means that {le nu le nu broda cu balvi cu purci}.
{pu co'a broda} means that {le nu le nu broda cu cfari cu purci},
and {co'a pu broda} _should_ mean that {le nu le nu broda cu purci
cu cfari}. Every tag unit corresponds to a binary relationship, that
relationship is all we need to interpret them and their combinations.
> I guess this is basically the prescriptivist-naturalist (is that what you
> guys call it?) debate again - the alternative to prescribing an
> understandable model for construction and interpretation of tenses being a
> lawless breeding-ground for confusion and malrarbau, with people just
> using the keywords to translate to and from lojban - rather than
> translating directly to and from the spatio-temporal locations of events.
> IMO. I just think we can and should do better.
I certainly agree. But the CLL prescription is not always complete,
and it has odd restrictions.
> Now whether we actually declare phrases which don't fit our models to be
> unparsable or just meaningless, I don't see much matters. But having a
> the set of meaningful words being easily decidable (by brains as well as
> machines), such as is assured by making it part of the formal grammar,
> would be particularly nice.
I agree. Ideally every phrase that parses should be meaningful,
and every meaningful phrase should parse.
> > > So how does this work with a full tense rather than just a fragment of
> > > one? What would
> > > {loi snima cu carvi pu zi ze'a ba'o le ca dunra}
> > > mean, for instance?
> >
> > First you'd have to tell me what {loi snima pu zi ze'a ba'o carvi}
> > means. Your sentence is very similar, but with the current winter
> > as reference, instead of an implicit one. I would take it to mean
> > that a short time ago, for a medium interval of time, it had snowed
>
> Hmm, same here. But how does that fit in with {carvi ze'a le ca dunra}
> *not* being the same as {ze'a carvi}-with-the-winter-as-reference?
{ze'a} relates the bridi event to its duration. Each tag relates the
bridi event to something else, which can be made explicit with a sumti
or left implicit when the tag is used directly on the selbri.
> I
> don't see how you can reasonably make a special case out of ZAhO, with
> it having a different meaning when part of an explicitly expressed tense
> like the above. Whatever CLL says.
I don't want ZAhO to be a special case. CLL makes it a special case.
I want {ba'o} to always indicate the aftermath of the bridi event,
not sometimes that and sometimes the aftermath of some other event.
{ba'o} should relate the bridi event to something else which specifies
the aftermath of the bridi event. Always.
> > (then I suppose it must have started to snow again, otherwise how
> > would you know that the aftermath lasted a medium interval and will
> > not a last for a long one?).
>
> I think the implications of this lead to a contradiction of CLL and
> common usage, but I can't seem to find the words to express it right
> now. Sorry. I might try again later.
There isn't that much usage of compound tenses though.
mu'o mi'e xorxes
__________________________________
Do you Yahoo!?
The New Yahoo! Shopping - with improved product search
http://shopping.yahoo.com