From jorge@PHYAST.PITT.EDU  Sat Mar  6 22:44:56 2010
Date:         Mon, 29 May 1995 21:16:21 EDT
From: jorge@PHYAST.PITT.EDU
Subject:      Re: quantifiers on sumti - late response
To: Bob LeChevalier <lojbab@ACCESS.DIGEX.NET>
X-From-Space-Date: 
X-From-Space-Address: LOJBAN%CUVMB.BITNET@uga.cc.uga.edu
Message-ID: <rZnYJd5s8kK.A._lE.ot0kLB@chain.digitalkingdom.org>

la dilyn cusku di'e
> The inner quantifier is {ro} unless it's changed, no?  By putting in
> {su'o}, you explicitly say it's not "the one and only set", but some
> piece of that set.

For {lo}, the inner quantifier is always {ro}.  If you say {lo ci
broda}, you are saying that three broda are all the broda there are.
(See example 7.7 of the sumti paper.)

I think it is the same for {lo'i}, so {lo'i su'o broda} is the set of
all broda, and you are saying that "all" are at least one, i.e. it is
not an empty set.

So {lo'i [ro] broda} and {lo'i su'o broda} refer to the same set, but in
the second case you are also saying that it is not the empty set.

>  > But you never encounter {lo'e remna}. Or rather, you can't conclude
>  > anything about {lo'e} remna from properties of the one you encounter.
>  > ...
>
> Yes, this is quite true, but not relevant.  And's point is that the
> properties of {lo'e remna}, unlike the properties of {lo'i} or {loi},
> are of the same type as the properties of {lo remna};

The properties of {loi remna} are of the same type as those of {lo
remna}.

> in particular,
> since practically all {remna} have exactly one {stedu}, it should be
> true that {pa da stedu lo'e remna}.  Remember, it's a myopic singular.
> (And yes, {da} would probably be {lo'e stedu}, but you don't need to
> specify it.)

If that were so, then you couldn't say {ti e ta stedu lo'e remna},
because only one thing could be a {stedu lo'e remna} and the usefulness
of {lo'e} decreases dramatically.  I think the best way to think about
it is that {ta stedu lo'e remna} is not at all a predication about {lo'e
remna}, but only about {ta}.  It is as if {lo'e remna} makes {stedu}
into a one-place predicate "x1 is a human head", and all you say is that
{ta} fits that predicate.

> Huh?  How else would you say "x1 has exactly one head"?

I didn't express myself clearly.  {ta se stedu pa da} means "that has
exactly one thing as head".  But there is another possible 1-place
predicate "x1 is one-headed" (or whatever) that is not a relationship
between two objects but only a property of one.  Say {pavselstedu} is
that predicate, then you can say {lo'e remna cu pavselstedu}, but you
can't say {lo'e remna cu se stedu pa da}, because there are more than
one thing that are in relationship {stedu} with {lo'e remna}.

>  >  How do you like {lo'e reno plise}?
>
> {ta tanxe lo'e reno plise} sounds to me like it's a box that can hold
> twenty different apples (not necessarily at the same time), rather
> than a typical mass of twenty apples.  Is this a silly interpretation?

Actually, I think you may be right.  (After holding twenty apples, the
box self-destructs.)

Ok, so it's {ta tanxe lo'e plise renomei}.

>
> mu'o mi'e. dilyn.
>
co'o mi'e xorxes