Message-Id: <199310300813.AA22653@eli.CS.YALE.EDU>
Date:         Sat, 30 Oct 1993 04:11:15 -0400
Reply-To: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
To: Erik Rauch <rauch>
Status: RO
Subject: Lojban

UC> So I guess one part of the Lojban learning observation is to see which
UC> kind of deep structure the users would define.  This choice would
UC> reflect the actual human internal deep structure, right?

Basically so.  I think rather that we will see if the language is
learnable as it is defined.  If so, then we know that there exists a
human-internal set of transformations from any deep structure that is
ingrained into our heads to the Lojban surface structures.  This then
would suggest some type of isomorphism with the Lojban
defined-deep-structure.

UC> The case tag system has some advantages, so I wonder if there are
UC> case tags for prepositions like "to" and "from", which are usually
UC> already expressed by mandatory arguments.  How do you say "What did
UC> you do to her?" in Lojban?

We don't have anything corresponding to the whole of English preposition
"to" or "from", especiallly including the example you give.  There is a
case tag for destination, and another for origin, but neither applies to
the example, nor to most uses of "from" and "to", only those with a
specific motion.  I would do the example as one of two versions
depending on emphasis.

Assuming ko'a represents "her":
do gasnu ma ri'i ko'a
You are agent in what-action, with-this-agency experienced by her.
or
do gasnu ma poi se lifri ko'a
You are agent in what action that is experienced by her
(which uses no case tags at all)

or
gau do ko'a lifri ma
With-agent-you, she experiences what?

ri'i is a generic experiencer case tag
gau is a generic agent case tag (the do-er of the action)

UC> >There is no exact equivalence of "of".  The words be/bei are just gram
UC> >separators to attach any arguments onto the'sheriff' argument.
UC>
UC> I thought "pe" is quite similar to "of".  So are there any differences
UC> between "le pulji pe la NATinxem" and "le pulji be la NATinxea bei"?

"pe" is similar to the generic "of" which is often the "genitive case"
in those languages that have such a thing.  But that is equating apples
and oranges - they are not grammatically similar and you cannot therfore
use them in nearly the same way (pe is not in any way a preposition, but
rather a binding word between two arguments).

le pulji be le NATinxem  (no "bei")
the police of-organization Nottingham
(be is serving to label the following argument as the 2nd place of pulji
which is the organization the police are associated with.

le pulji pe la NATinxem  (that should have been a "la" in the last one too)
The police that are in some way associated with Nottingham; i.e.
the police of Nottingham

Both versions can be translated to English "of", but that is not the
"meaning" of either, just the closest English colloquial approximation.

UC> >didn;t
UC> >have a good dictionary, or failed to check himself, because there are
UC> >of strange word choices fro his Chinese (I say this since I did the Ch
UC> >dictionary work for Lojban myself.
UC>
UC> Well, the problem of choosing between synonyms when making
UC> Loglan/Lojban words exists for not only Chinese.  I recall that for
UC> "see/view" the word is vinci/viska.  I guess "view" instead of "see"
UC> was chosen because it would be more consistent with words from other
UC> (especially European) languages.

Not quite that frame of mind.  We used several English synonyms and let
the computer decide which one matched up with phonemes from other
languages best.  In this case, you are correct that "view" was the word
chosen, but that may or may not have been because of similarity with
European language words.  We tried "see" as well.

Your guess on blood was correct - it is "ciblu".  Another word is a bit
harder, but highly recognizable once you think about it "blue" is
"blanu" bLANu and BLanU.

But I have heard that "brown" is not a color commonly used in Chinese.
I can't remember the word I obtained from the dictionary, but the result
was "bunre".  This may be a little less recognizable, especially if the
Chinese root we used is not well-known.

Also the algorithm might give a strange result because the word I got in
Chinese was a digraph, and the word making algorithm might have taken
part of each component resulting in something with no real recognition
at all.

I don't know any examples of this off hand, but I have the old Loglan
dictionary in front of me.  For the concept regular/cyclical/periodic,
Dr.  Brown used the Chinese "i lyu" by some Romanization, taking the
"il" to score as part of "rilri".  The Lojban word for the concept is
"diknu", by the way, and I know that the English component in the "ik"
from "cyclical".  Dr.  Brown's word for "powerful" is "lilpa" from
Chinese "li liang".  The Lojban word is "vlipa".  Dr.  Brown's word for
"screw/bolt" is "skori" from Chinese "lo so ding"; the Lojban word is
"klupe" where I do not know any of the etymology.  In the latter case,
the Chinese contribution appears to be the s and 2 vowels.  There are
probably comparable Lojban examples.

lojbab