Message-Id: <m0pgFWB-0000R4C@xiron.pc.helsinki.fi>
Date:         Mon, 14 Mar 1994 10:40:40 -0500
Reply-To:     Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      Re: Even Madder (A no-go :-)
To:         lojban@cuvmb.cc.columbia.edu
Cc:         loj@access.digex.net
In-Reply-To:  <199403121450.AA09869@nfs1.digex.net> from "Veijo Vilva" at Mar
              12, 94 11:24:36 am
Content-Length: 2449
Lines: 49

la veion. cusku di'e

> I haven't really checked but I think we could get away with
> using the afterthought connectives as the infix part of the
> forethought construction. I don't think there is any conceptual
> reason to use a distinctive connective here. Actually, having
> the same connective would be logically clearer and mostly
> compensate for the use of/need for the operand type indicators.

Interestingly, this design is the original Loglan one, present in the 1975
version but changed before 1980 (TL3/4:60):

        If a speaker has genuinely thought enough ahead to know that
        the first sentence uttered is to be [logically] connected to
        another, he presumably has thought what the connection will
        be.  Thus, he can tell his hearer what that connection is to
        be immediately, rather than waiting until the first sentence
        is over to say what the connection is; although, with the
        [gi...ga] form, he said >that< it would be connected at the
        very beginning.  In short, the order of the "forethought
        connectives" is now to be reversed: the content item coming
        at the beginning and the [gi] section coming between the
        two connectands (much as the connective [.i], to which [gi]
        is clearly related, comes between two sentences).  Thus,
        instead of "[gi] S [ga] T", we now write "[ga] S [gi] T" with
        the same meaning.

                        -- pc

The passage goes on to explain that the [nai]s attached to each connectand
remain adjacent to that connectand, so that the old [gi...naga] now becomes
[ganai..gi], not [naga...gi].  In the 1975 version, [ginai] did not exist;
negation of the left connectand was expressed by prefixed [na], just as in
afterthought.  Further, it is clear from context that all uses of the
forethought connectives underwent this swap, although only bridi connection is
discussed.

(Of course, the actual text used Loglan cmavo forms, hence all the brackets.
For those who care, "nai" used to be "noi", "na" used to be "no", and the
cmavo beginning with "g" used to begin with "k".)

> BTW. is there a zo'e type and/or vague connective?

The vague bridi connective is of course just ".i"; despite Jorge's ingenious
use of "do'e", there are no clear-cut vague connectives at other levels.

--
John Cowan              sharing account <lojbab@access.digex.net> for now
                e'osai ko sarji la lojban.