Re: [lojban] Re: mi kakne lo bajra

["John E. Clifford" <kali9putra@yahoo.com>]

Sent from my iPad

On Oct 30, 2010, at 17:48, Lindar <lindarthebard@yahoo.com> wrote:

Other examples of potentially raising selbri are djuno, cilre, facki,
jimpe, and all the others with a place structure involving "fact x2
about x3". The argument x3 can be raised from the subordinate clause
in x2. This rarely happens however, because the raised argument is
inconveniently located.

That's not what I'm talking about. That's a feature of the language.
When x1 of a subordinate clause is elided, it's assumed to be the x1
of the main bridi. That's not what we're discussing, so forgive me if
I've applied the wrong terminology.

The case cited is not a case where x1 in the subordinate is elided because it is the same as the x1 in the main clause, but rather of some term in the subordinate clause being moved from the subordinate clause to another place in the main clause.  As noted, this is rarely done when  the subordinate clause comes before the place where the raising would place the term (although shifting that place is not that uncommon).

   Nobody really says:

    mi facki lo du'u sralo kei ko'a
    "I found out being Australian about her."
    "I found out about her being Australian."

We really should, though. >_>

Why? it's awkward and it can lead to problems.

Instead of:

    mi facki lo du'u ko'a sralo
    "I found out that she is Australian."

Raising is just not convenient in Lojban for these propositional
attitude selbri.

(Also, it is not clear why some of them have a raising place and
others, like for example "birti", don't. Either all should have it or
none, but Lojban place structures are so full of exceptions. But
that's just an aside.)

I think that's an error. I've already semi-addressed it to the BPFK.

Which? Some having and some not having raising places (we can actually provide those places -- grammatically -- for any word; they just don't make any sense in many cases) or that there are general exceptions in Lojban place structure?  The latter is simply an observable fact, the former is more easy to dispute (one may argue that there are significant differences behind the different place structures -- and that there are not).  And has been.

Wanting an apple for the purpose of eating
it is still sumti raising, because it's adding an implied concept of -
having-. That's what sumti raising is. =/

It is, in a sense, sumti raising, but not for the reason you give.
Consider these:

      mi pilno lo mapku lo nu dasni
      mi nitcu lo mapku lo nu dasni
      mi djica lo mapku lo nu dasni

Two of these are wrong. =\

Careful!  In the rush to avoid one (or two) sort of error, you do not want to create a new sort of error.  There are surely occasions when I want or maybe even need a particular hat (in this case) and then these three sentences present no problems, even if they do involve raising (which it is not perfectly clear they do -- as opposed to elision of repeated information).  The problem comes when there is not such a particular hat and we do raise it out of the event description anyhow.  There is no way to tell in Lojban which of these histories the given sentence comes from (notice this is a problem even for 'pilno' if we pull the term across tense or  modality borders).  So, we can find out what hat he uses by looking and what hat he neds or wants by asking or some such thing.  But when we can't do that, even in principle, the raising ought to be illegitimate (and marked somehow to prevent it -- putting it into some abstraction phrase: nu, ka, and the like
 will work).   

"pilno", "nitcu" and "djica" all have basically the same place
structure. (There may be others like them, for example "sazri.)

Now, we could say that in those three examples, there is a double
sumti raising, since the x1 of dasni is raised to the x1 of the main
clause, and the x2 of dasni is raised to the x2 of the main clause.
But there is nothing wrong with any of them! sumti raising is a normal
part of the Lojban grammar. Some selbri just happen to have argument
places for raised arguments. So what? Why this witch-hunt about the x2
of djica? Why doesn't anyone ever worry about the tens or maybe
hundreds of other sumti raising places that the gismu list provides?

Because these are grammatical changes which do not correspond to valid argument, contra "the Logical Language"  (one version, anyhow), It does seem that the racial matter is cases where human intentions are involved and only for these is there a problem.  And, with varying likelihoods: "I gave an apple" has almost zero chance of being a problem (or of being a raising, for that matter), "That is a picture of a horse" has a significant but not major chance of being an improper raising,  "I need a hammer" is virtually certain to be improper.

{nitcu} and {djica} both have (or should have) an abstracted second
place.

Here again, you have overshot the mark.  These predicates don't *require* abstract second arguments, but they usually need them to be true,  But not always (see above).

What about the apple do you want/need?You've expressed reason
and the target, but not what to do with it. -THAT- is my problem here.
{pilno} is a bad example, because there's nothing else implied.
{nitcu} ... do you need possession? Do you need to throw it? The''''
problem is that the definition does not include -having-. So do we
assume when it's an object it's having, and something else in all
other cases? That's not what Lojban is about. That's just plain bad
practice.

I see your general point and it is somewhat right: we need to insulate the term here from the domain of discourse and putting it in an event description, say, does that very nicely (though that gets ignored as well).  But it is the insulation, not the additional information about why one wants or needs an object, what one is going to do with it, and so on, that is required.

And you didn't say what you think about "dunda lo plise". Do you
object to that too, or do you wisely ignore the gismu list comment in
that case?

{dunda} implies no transfer of ownership. It's a simple physical
transfer of an object from one person to another. It's like borrowing
a pen. It -could- mean a transfer of ownership, but no such sense is
implied by the word in of itself.

xorxes' point (though not how we would put it) is that, if it does involve a change of ownership, then it is likely that any raising involved (I forget what the structure of dunda' is, so I am not clear  where the raising come in), it may be illegitimate (in the semantic/logical sense -- it is always grammatically allowed).  

There are other ways around this problem, notably taking 'lo broda' to refer not to brodas directly but to brodaness or broda-type, but that creates problems in the ordinary, otherwise nonproblematic  uses of the term, which seems a net loss. (What is the type of three girls walking down the street and how is involved in causing a car wreck?).




      

-- 
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.