[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] semantic parser - tersmu-0.1rc1




Sent from my iPad

On Nov 26, 2011, at 12:29 PM, Martin Bays <mbays@sdf.org> wrote:

> * Saturday, 2011-11-26 at 14:35 -0300 - Jorge Llambías <jjllambias@gmail.com>:
> 
>> On Sat, Nov 26, 2011 at 12:41 PM, Martin Bays <mbays@sdf.org> wrote:
>>> Hmm. I wonder if I now finally understand part of xorlo: would you say
>>> that {lo broda} is equivalent, under this side-clause interpretation of
>>> {noi} you've just set out, to {le broda noi broda}?
>> 
>> I can't think of any reason why not, but then I'm not completely
>> satisfied that I understand "le".
> 
> Ah! Maybe this is progress! For my personal understanding of xorlo, that
> is.
> 
> {le} to me seems pretty clear: {le broda} refers, wherever it appears,
> to some individuals which I have in mind or would have in mind
> if I thought about it (to steal pycyn's phrase), and which I hope you
> will be able to glork from a mixture of context and them being described
> as brodaing.
> 
> Whether I actually believe them to broda is beside the point; presumably
> I do expect that you believe them to broda, or that you expect me to
> expect you to believe them to broda, or etc.
> 
> Since there's a single intended referent-bunch, {le broda} is invariant
> under passing it through a negation.
> 
> Obviously it isn't wholly immune to scope, because of the {ro da le
> broda be da} issue.
> 
> I don't see why it should be even when the description doesn't
> explicitly mention bound variables; e.g. why {ro verba cu prami le
> mamta} shouldn't be a reasonable abbreviation of {ro verba cu prami le
> mamta be ri}, or why in {pu je ba ku mi'o jinga fi le bradi} we
> should have {le bradi} getting the same referents both times.

It is a linguistic precondition of the collapse of parallel sentences marked by {je}.
> 
> xorlo seems to declare that it is constant in this way - unless I'm
> misunderstanding again? (Just being hopeful...)
> 
> 
> Anyway, {lo broda} just adds to {le broda} the side-claim that the
> referents *actually* broda, rather than merely that I expect you to
> think that they do (or otherwise understand me when I describe them as
> brodaing). OK!

And subtracts the specificity that the in mind provision gives.
> 
>>> I don't know about {lo}, but presumably {le broda ku voi brode} is
>>> equivalent to {le broda je brode}.
>> 
>> Let's see.
>> 
>> Assuming
>> 
>> (1) le broda = zo'e noi mi ke'a do skicu lo ka broda
> 
> I wouldn't want to take that too literally... using {le broda} doesn't
> merely claim that I'm describing the referents to you as brodaing, it
> actually *does* describe them as brodaing.
> 
>> (2) ko'a voi broda = ko'a noi mi ke'a do skicu lo ka broda
>> 
>> Then we have:
>> 
>> (3) le broda ku voi brode = zo'e noi mi ke'a do skicu lo ka broda zi'e
>> noi mi ke'a do skicu lo ka brode
>> 
>> which I'm happy to reduce to:
>> 
>> (4) le broda ku voi brode = zo'e noi mi ke'a do skicu lo ka broda .e lo ka brode
>> 
>> I'm not so sure about the move to:
>> 
>> (5) le broda ku voi brode = zo'e noi mi ke'a do skicu lo ka broda gi'e brode
>> 
>> and then to:
>> 
>> (6) le broda ku voi brode = zo'e noi mi ke'a do skicu lo ka broda je brode
>> 
>> which would give us:
>> 
>> (7) le broda ku voi brode = le broda je brode
>> 
>> The move from (4) to (5) is the one I find most suspect. It has to do
>> with the semantics of "skicu", and of course also definitions (1) and
>> (2) in terms of "skicu" may or may not be right.
> 
> Yes, I agree that (4)->(5) is dodgy. But I don't see any corresponding
> dodginess in {le broda ku voi brode} -> {le broda je brode}.
> 
> Martin

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.