WikiDiscuss

On Tue, Nov 23, 2004 at 09:05:35AM -0800, John E Clifford wrote:
> While I don't suppose there is any good
> systematic way to say what can occupy a place in
> a predicate there does seem to be a general
> notion that sumti in places stand for components
> of event being described. Some sumtcita
> expressions seem not to meet this condition:
> while the cause of a event may be seen as a
> component of the event, who knows about it or who
> has described it does not.

I can see that one either way, myself.

> These seem rather to
> adverbial to the main bridi, adjectival to one of
> the other places, or to suggest that the surface
> structure of the claim is inside out the logical
> claim (the brivla of the added place is the main
> brivla and the apparent main clause is shoved
> into some subordinate role).

Yeah, I don't like that solution in general myself.

> The epistemological
> sumtcita seem particularly to fall into this
> category , so that I wonder whether incorporating
> the "added place" locution is really all that
> informative. Perhaps just saying that they add
> information to the basic claim and then selling
> out the nature of that information in each case
> would be more elegant (and accurate).

That's a pretty massive change, though. Also, they *do* take sumti,
which makes them sumti tags, like FA.

> --- Robin Lee Powell
> <rlpowell@digitalkingdom.org> wrote:
>
> > Trimming would have been nice.
> >
> > I happen to be working on the examples right
> > now.
> >
> > On Mon, Nov 22, 2004 at 04:31:04PM -0800, John
> > E Clifford wrote:
> > >
> > > > ;du'o (BAI): According to... Tags a
> > sumti
> > snip
> > > > se du'o, te du'o, ve du'o.
> > > > ** Keywords: According to, known by.
> > >
> > > Since {djuno}, like English "know," intails
> > that
> > > what is known is true, the reading "according
> > to"
> > > is misleading, since it has no such
> > implication
> > > and often (with a certain tone of voice to be
> > > sure) implicates that the information is
> > false or
> > > at best unknown. And, once it is established
> > --
> > > as anybody knowing it would do — the knower
> > is
> > > not important, the truth is established. I
> > > suppose that what is intended is something
> > "is
> > > vouched for by" citing the presumed reliable
> > > source of the information.
> >
> > Actually, I'm using it for exactly what I said:
> >
> > .oi sai mi cliva du'o la patfu
> > Dad knew I left!
>
> Note the apparent inversion in the translation.

It's a non-literalistic translation.

> Is this basically a claim about what Dad knew?
> Is it a joint claim "I left and Dad knew it"?

The latter more than the former, I would say.

But really, neither. It's a claim about a four-place predicate,
call it broda, with the place structure:

x1 leaves/goes away/departs/parts/separates from x2 via route x3
with knower of departure x4

where we tag the x4 place with du'o.

About "se zau":
> Yes, though this is a little obscure: "We left,
> our passport having been approved" or some such.
> If so, we get causal notions or the like again.

More permissive than causal, which is what we want here.

> > > {cusku} seems to deal with conceptual and
> > > propositional expressions, not performances
> > per
> > > se.
> >
> > Erm, no. cusku takes a se du'u, text, or lu'e.
> > All of these are
> > actual expressions, not concepts. If it was
> > conceptual, it would be
> > du'u, not se du'u.
>
> Yes, {cusku2} is text in some form. {cusku3} is
> ideational, however — certainly none of them is
> a performance.

Yeah, you seem to be correct in the letter of the law, but it seems
to me that cusku was intended for expression *in* *general*.

> > > Which does raise the question, "if the tagged
> > > item occupies the nth place of the underlying
> > > predicate, what place does the sentence to
> > which
> > > it is attached occupy?"
> >
> > A newly created, un-numbered place with the
> > semantics of the nth
> > place of the predicate underlying the BAI tag.

I'm sorry, I was anwering a completely different question. I don't
know the answer to yours.

I'm inclined to say "None of them; the transformation is not
reversible in that fashion.", but I confess to not having thoroughly
thought it through.

> > > > se'o verba selsanga secu'u le du'u lo za'i
> > > > jmive cu selsenva po'o
> > > > ''I know culturally that children's songs
> > > > express that life is only a dream.''
> > >
> > > Sentence is first place when tagged is
> > second.
> >
> > What?
>
> "Sentence" is wrong here, there isn't one.
> {verba selsanga} occupies the first place of
> {cusku} because (?) {le du'u...po'o} occupies the
> second (as flagged). Looking for a pattern here,
> but not finding it.

Again, I don't necessarily accept the transformation you are trying
to impose as valid.

> > > Reasonable but... . Why is this an extra
> > place on
> > > {selsanga} rather than directly expressed:
> > {lo
> > > verba selsanga cu cusku le du'u le za'i jmive
> > cu
> > > selsenvi}?
> >
> > You'd have to ask the original author of the
> > sentence; it's from
> > IRC.
>
> Putting this out to all and sundry is meant to
> get input from anyone who lnows anything about
> it. Hopefully including the author.

Indeed, but my point was that that is a stylistic issue, and not one
we need to address here.

-Robin