Re: [lojban] Re: tersmu 0.2

On Sun, Nov 23, 2014 at 8:25 PM, Martin Bays <mbays@sdf.org> wrote:

* Sunday, 2014-11-23 at 11:11 -0300 - Jorge Llambías <jjllambias@gmail.com>:

> On Sat, Nov 22, 2014 at 10:01 PM, Martin Bays <mbays@sdf.org> wrote:
>
> > So for example, if yesterday morning it snowed out of a clear sky, but
> > in the afternoon it snowed while the sky was cloudy, and both of these
> > events are salient, what could you say about the truth value of
> > {ca lo prulamdei zo'u lo snime cu carvi .i je nai ca bo lo dilnu cu
> > gapru}?

For clarity, I should really have made that example
{ca lo prulamdei zo'u lo snime cu carvi .i je ca bo lo dilnu na gapru}.
That doesn't change anything, right?.

Let's see. For the second connectand the nai-version gives:

naku ca lo nu lo snime cu carvi kei lo dilnu cu gapru

~ lo nu lo dilnu cu gapru cu na cabna lo nu lo snime cu carvi

and the na-version gives:

ca lo nu lo snime cu carvi kei lo dilnu na gapru

~ lo nu lo dilnu na gapru cu cabna lo nu lo snime cu carvi

I think they say more or less the same but may not be logically equivalent, unless there's some cabna-logic that says that if an event doesn't cabna another, then the corresponding negated event must cabna it, which I think is reasonable semantics for "cabna".

> I would say its truth value is the same as the truth value of "ca lo
> prulamdei zo'u ca lo nu lo snime cu carvi kei lo dilnu cu gapru".

And to be clear, do you actually only mean \iota here, or do you really
mean that it has the truth value of that sentence when the {lo nu} are
taken to refer to the corresponding event-kinds?

I don't know enough about \iota and kind-theory to decide which one is a better match for "lo", but I doubt it's \iota.

>

> So you would expect both of these to be true:
>
> ca lo prulamdei zo'u lo snime cu carvi .i je nai ca bo lo dilnu cu gapru
>
> ca lo prulamdei zo'u lo snime cu carvi .i je ca bo lo dilnu cu gapru
>
> ? They may seem contradictory but they are not because of the hidden "su'o"
> you are assuming.

I don't think quantification being involved in the connective is key
to their being non-contradictory; that's just a matter of {ge ca ko'a
broda gi ca ko'a na broda} not being contradictory.

Is that then not equivalent to "ca ko'a ge broda gi na broda"?

But yes, I would expect both to be unambiguously true. I would say they
can be fairly accurately translated as "yesterday it snowed while it was
(not) cloudy".

Shouldn't the snowing be the reference, though, and the being cludy the tensed statement. Isn't the second connectand more like "yesterday it was (not) cloudy while it snowed"?

The event-kind semantics would have the translation be something like
"yesterday, during the snowing it was (not) cloudy"? And the problem
we're seeing here corresponds to the failure of the maximality condition
presupposed by that "the"?

The problem is indeed that the snowing occurred twice, so it's not very easy to figure out what it means to say that something occurred during it.

But that isn't really accurate, because you're going via kinds and
a temporally dependent notion of cabna, which seems a fundamentally
different route from that english sentence.

Actually, there's something I'm not understanding here.

I understood you as having cabna(ko'a,ko'e) holding at a given time t,
where ko'a and ko'e are event-kinds, meaning that instances of ko'a and
ko'e occur simultaneously at (or near?) t. There will always be other
instances at other times, but those are ignored.

If you insist on bringing instances in, you will have to quantify over them.

Then assuming we agree
that {ca lo prulamdei ku broda} only implies that broda occurs some time
yesterday, not that it occurs throughout yesterday, wouldn't you have
{ca lo prulamdei zo'u lo snime cu carvi .i je ca bo lo dilnu cu gapru}
being true because the two kinds were cabna (some time) yesterday?

I think it means they were cabna yesterday, not some time yesterday. In our universe of discourse yesterday counts as one time, we don't have it divided into many times such that cabna holds at some of them and not at others.

Or do you not believe in even this trace of quantification in the
semantics of {ca}? In that case, would you have {ca lo prulamdei ku mi
citka lo tamca} true only if it took up the whole day?

No, that would be "ze'a lo prulamdei". "ca lo prulamdei" just says that lo prulamdei is the time at which "mi citka lo tamca" takes place, not its duration. Similarly when I say "mi zutse lo stizu" I don't mean to say that there are some parts of the chair on which I sit, and some parts of the chair on which I don't sit, it's just a relationship between me and the chair. You can give explanations if you want that sitting involves the body of the sitter being in contact with some fraction of the surface of the chair and not necessarily with the whole surface, but that's part of understanding what "zutse" means, not any hidden quantifications in "mi zutse lo stizu", there are no hidden quantifiers over parts of the surface of the chair in that claim, and similarly there are no hidden quantifiers over parts of yesterday in "ca lo prulamdei".

> I'd prefer to go with:
>
> ca lo prulamdei zo'u lo snime cu carvi .i je nai ro roi bo lo dilnu cu
> gapru
>
> ca lo prulamdei zo'u lo snime cu carvi .i je su'o roi bo lo dilnu cu gapru
>
> (where I'm using "PA roi bo" with tense-like semantics).

I don't see how these are working, for the same reason.

Since the premise was that lo nu lo snime cu carvi happened twice yeterday, I would read those as:

Yesterday, it snowed (twice) and not every time that it did it was cloudy.

Yesterday, it snowed (twice) and at least one of the times that it did it was cloudy.

The "(twice)" is not stated, but it was a part of the set-up.

I'm taking "ro roi lo nu PA roi broda" to mean "each of the PA times that it brodas", which I think is a reasonable reading of ro roi ko'a" when ko'a is something happens multiple times. So "PA1 roi lo nu PA2 roi broda" ~ "ca PA1 lo PA2 nu broda".

mu'o mi'e xorxes