[lojban] Re: semantic primes can define anything

[John E Clifford <lojban-out@lojban.org>]

--- Jorge Llambías <jjllambias@gmail.com> wrote:

> On 3/27/06, John E Clifford
> <clifford-j@sbcglobal.net> wrote:
> > > X often thinks about Y
> > > X thinks good things about Y
> > >
> > > X so'i roi pensi Y
> > > X jinvi lo zabna Y
> > >
> > > The first one does not say that X often has
> > > opinions about Y,
> > > but that Y is often on X's mind.
> >
> > But can one have someone on one's mind
> without
> > some proprositional content?
> 
> Yes, definitely.

Could you give a case (I suspect this is another
old disagreement here, related to the objects of
desire).
> 
> But even if one couldn't (which I don't see why
> not), one could still
> have some propositional content on one's mind
> without holding it
> as an opinion.

I have always thought the "opinion" part of
{jinvi} was peripheral in a sense; that is, it
was there only to distinguish {jinvi} from
{djuno} and maybe even from {krici}.  If this is
wrong then {pensi} is doubly defective: it lacks
a place for the content of the thought and it has
an unmarked opaque place for the object of
thought.
 
> > As far as I can see
> > -- from a logical point of view, mind you,
> not
> > NSM  -- {jinvi} should not have a third place
> (it
> > creates all kinds of problems and is an
> artifact
> > of English idiom) and {pensi} is then just
> {jinvi
> > tu'a}.  NSM seems to want {jinvi} with the
> common
> > case of dropping the x2.
> 
> I disagree that {pensi} is {jinvi tu'a}. I can
> think (pensi) of pigs flying
> without opining (jinvi) that pigs fly.

But apparently you can't pensi of pigs flying;
there is no place for propositional content.  Or
do you want that {pensi2} is for propositional
content (which would eliminate a raft of
problems)?
 
> > > So the sense of {fasnu} would not be prime?
> >
> > Apparently not.  I don't have cases and the
> talk
> > suggests that it is always "Something happens
> to
> > something," never "Such and such happens"
> > (indeed, the grammar for English NSM
> sentences
> > doesn't allow an event noun phrase in place
> of
> > "such and such"; the best one could get would
> be
> > "This happens: such-and-such [as a full
> > sentence]" Events tend to be unspecified with
> > "happens" anyhow.
> 
> So a paraphrase like this would not be allowed:
> 
>  Something happens.
>  X thinks that this is a good thing.
> 

Yes; I think it has to be "something happens to
something."  But again, in the absence of more
cases it is hard to be sure.

> > > So part of the problem is that they don't
> > > manage to explain
> > > their primes very well.
> >
> > Well, of course they are primes so they can't
> > officially be explained at all.
> 
> They can't be paraphrased in terms of primes,
> but surely they
> can be explained better.

I agree.  Minimally, we should be given the
paradigm formats and these distinguished from
others that are like them.  Even Durst's comments
about MOVE ar not as clear as he seems to think
they are.
 
> > Still, some more
> > examples would be useful.  The only one I
> have
> > actually seen that directly deals with this
> > problem is a comment by Uwe Durst in his
> rebuttal
> > article in Theoretical Linguistics 29.3,
> where he
> > selects "He moves (a part of) his body" as
> giving
> > the sense of MOVE in English (presumably the
> > volitional twitch, not the "cause to be
> > displaced" reading even here).
> 
> Ah, I thought MOVE was not meant to be
> agentive! I thought
> it was more like {muvdu}, as in "the ball moves
> from here to there",
> but also (unlike {muvdu}) including movement in
> place.

It is unclear -- to me, from this example --
whether it is agentive or reflexive only or even
just your {muvdu} (except there does not seem to
be any inherent place shifts).  Again, I have no
cases.  It is also unclear why it is prime, since
it seems to be definable in terms of PLACE and
OTHER and DO by some such pattern as one I laid
out for tokipona "tawa" elsewhere.


To unsubscribe from this list, send mail to lojban-list-request@lojban.org
with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
you're really stuck, send mail to secretary@lojban.org for help.