Message-Id: <m0rCWBI-00007FC@xiron.pc.helsinki.fi>
Date:         Mon, 28 Nov 1994 09:10:09 EST
Reply-To:     bob@gnu.ai.mit.edu
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         bob@GNU.AI.MIT.EDU
Subject:      Re: Some thoughts on Lojban gadri
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
In-Reply-To:  <199411240006.TAA16328@albert.gnu.ai.mit.edu> (message from Jorge
              Llambias on Wed, 23 Nov 1994 18:52:59 EST)
Content-Length: 5375
Lines: 123

Jorge explained his understanding of the Lojban gadri of selma'o LE.
It looks to me that much of what he says is at variance with how
Lojban has been defined hitherto.


    lo broda        At least one of all things that broda.

Yes, but bear in mind that the {su'o} and {ro} are merely *default*
values.  Unless explicitly specified as {su'o} and {ro}, an unadorned
{lo broda} utterance may (but usually does not) imply some other value
for them.
                    It doesn't say which one(s), but the question
                    is pertinent and has an answer in principle
                    (which doesn't mean that the speaker has to know it).

It is only in this recent thread that anyone has suggested that `which
one' is of any relevance to {le} or {lo} other than as a help in
making translations into English.  (Russian is more like Lojban since
it lacks an equivalent of `a' and `the'.)

Neither {lo} nor {le} are categorization operators that work by
specifying `which one'; they do something else.  They are not like
English categorizers at all.

If the context is such that there is one `for real' broda and one
other thing I might `designate as' a broda, {lo broda} does
distinguish one from the other.

As a general rule, you should not bother to talk about `which one' in
a brief statement about {lo} or {le}; `which one' is a consequence of
the utterance and its context, not of the grammar.  `Which one' is not
very pertinent to the act of categorizing whether something is
`designated as' or is `for real'.  It is only when translating into
short English phrases that `which one' becomes pertinent, since you
want to use `a' or `the' rather than the long phrase, `one or more of
those that really is/are in the context'.

The primary statement about {lo} is that it is a categorization
operator that tells you that this sumti is `for real' in the context
of the communication.  That is what is really pertinent.


    le broda        Each of the thing(s) that the speaker has in mind
                    that broda.

                    It does say which one(s), even if the
                    listener can't tell (in which case, the comunication
                    hasn't been all that effective).

The primary statement about {le} is that it is a categorization
operator that puts the following sumti_tail into the category of
`designated as'.  In itself, {le} is not a categorization operator
that specifies which one the way `the' and `a' do in English. (Note
that {le} is *not* like the French `le' either).

My niece draws pictures of umbrellas, red, green, purple, and blue.  I
also have a `real' umbrella.  In this context, {lo santa}
distinguishes among the possible umbrellas and specifies just one, but
{le santa} does not.

In this case, {le} fails to tell you which one.  The issue is more
than ineffective communication.  The listener is unable to tell which
picture I mean when I say {le santa} because the categorization
operation is not concerned with `which one' but with a different form
of categorization, namely that these are the umbrellas that belong to
the category of entities being designated as umbrellas.

Sometimes you can figure out `which one' and sometimes not;
`which one' is an orthogonal category to `designated as'.

However, as a default presumption, you can often *gloss* {le} with
`the' when you translate contemporary Lojban, since people have been
using it in a manner that permits this.


    lo'e broda      It doesn't say which one(s), and the question
                    is not pertinent.

Your comment should be phrased as a heuristic. Sometimes, {lo'e} can
say which one, although I agree that it usually does not and usually
the question is not pertinent.

Suppose I look out my window, as I am doing now, and see several pine
trees.  One of them, and only one of them, looks `typical'.  You are
part of the group with me looking out the window.  In this case, I
could say

    .i mi ca ca'a            zgana lo'e ckunu
       We/I now in actuality behold a/many/the typical pine tree.

If you know what a typical pine tree is supposed to look like, then my
utterance does say which one.  However, I agree that the most common
use for {lo'e} is, and should be, in the context of

    .i lo'e cinfo cu xabju lo fi'ortu'a
     The/A typical lion lives in what is really African terrain.

Also, of course, there are entities, such as the `typical family with
1.9 children' that cannot be pantomimed or pointed to.


    le'e broda      Like {lo'e broda}, but restricted to the
                    kind of broda the speaker has in mind (which
                    may include not-really broda as well).
OK.

    loi broda       Like {lo broda}, but instead of taken idividually
                    a group of broda is taken as a single entity.

An ambiguous comment: do you mean `the mass of individuals taken
together, like a pile of Christmas trees in the back of a truck', or
do you mean `a manifestation of the notion of Christmas tree, by
default (but not necessarily) a single Christmas tree'?


    lei broda       Like {le broda}, but instead of each separately,
                    all of them as a single entity.

Yes.

    Robert J. Chassell               bob@grackle.stockbridge.ma.us
    25 Rattlesnake Mountain Road     bob@gnu.ai.mit.edu
    Stockbridge, MA 01262-0693 USA   (413) 298-4725