Date: Fri, 27 Jan 1995 00:49:35 +0000 (GMT)
From: Iain Alexander <ia@STRYX.DEMON.CO.UK>
Subject: Re: replies re. ka & mamta be ma
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Reply-to: ia@STRYX.DEMON.CO.UK
Message-id: <01HMBPNHYLK60000X3@FIPORT.FUNET.FI>
Content-transfer-encoding: 7BIT
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Length: 2427
Lines: 51

In message  <9501261944.aa07700@punt.demon.co.uk> ucleaar@ucl.ac.uk writes:
> > > Isn't {lo mamta be lo patfu be la .and. joi la xorxes} vaguer than
> > > {lo mamta be lo patfu be la and beho beho .e lo mamta be lo patfu
> > > be la xorxes}? ...
Goran:
> > Yes, you are quite right here. {joi} version is definitely much shorter,
> > but is also certainly semantically different than full one. But the
> > sentence you just gave *can* collapse into
> > lo mamta be lo patfu be la .and. .e la xorxes.
>
> Are you sure? Wouldn't your version instead expand into {lo mamta be
> lo patfu be la and beho .e  lo patfu be la xorxes} (meaning A & X have
> the same grandmother)?

{lo patfu be la .and. .e la xorxes.} is identical (at least in all
respects relevant to this discussion) to {da poi patfu la .and. gi'e
patfu la xorxes.}, which is a common father of A & X.

There's no way to get an exact equivalent, abbreviated in this form,
of the original, but {lo mamta be lo patfu be la .and. ba'e .a la xorxes.}
is close.  The difference is that it could refer to one or more
grandmothers of And alone, but this is unlikely to matter too much
in practice, at least with a reasonably cooperative listener.

(If the quantifier is {ro}, there is an exact equivalence:
{ro patfu be la .and. .a la xorxes. broda} is the same as
{ro patfu be la and. .e ro patfu be la xorxes. broda},
but the proof is rather long, and I don't have it handy.)

This came up previously in the context of the so-called
tanru connectives ({melbi je cmalu bo nixli ckule} etc.),
and John Cowan at one point proposed a "distributive" marker
cmavo, but I can't wrap my brain round whether it would work
here.  If it did (let's call it {xa'i}, and make it a UI),
then {lo mamta be lo patfu be la .and. .e xa'i la xorxes.}
would be equivalent to the original above.

As for non-logical connectives, it's not clear that any of
them do anything appropriate here.  {jo'u} again suggests
a common father, {joi} means something strange (And-cum-Jorge),
and most of them definitely don't distribute anyway.
(I.e. {lo patfu be la .and. jo'u la xorxes.} is not the same
as {lo patfu be la .and. ku jo'u lo patfu be la xorxes.})
I think {fa'u} is an exception, and it might just work here,
but at best it would probably depend on the context.

co'o mi'e .i,n.
--
Iain Alexander                    ia@stryx.demon.co.uk
                    I.Alexander@bra0125.wins.icl.co.uk