Message-Id: <m0r8y9L-00005XC@xiron.pc.helsinki.fi>
Date:         Sat, 19 Nov 1994 14:14:12 -0800
Reply-To:     Gerald Koenig <jlk@NETCOM.COM>
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Gerald Koenig <jlk@NETCOM.COM>
Subject:      Re: Cowan's sum:opaque
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 1335
Lines: 32

lojbab said:

We can solve the "look for object" problem most simply by just creating a lujvo
for this meaning.  studji or stujundji or faktoi all could be appropriate,
perhaps with slightly different place structures according to the Nick
formulary.

lojbab
--------------

After reading the above I was hit by a brainstorm which I share.

Suppose we had a language shift cmavo similar to, for example, the
Greek lerfu shift, "ge'o", which would shift the following expression
prior to its terminator into the language of first order predicate
logic.  But predications about predications, i.e, all kinds of complex
sumti, would be illegal then as expressing second order predicate
logic.  It would force all following sumpti to be objects. Example:

mi cu sisku *ge'x* le mi cukta *[ge'z]*

ge'x initiates the shift to 1st order predicate calculus and ge'z
terminates it. Again, pardon my ungrammatical temp words. My last ones
seem already extinct so I'm not too worried about it.

In effect the x2 of sisku would be redefined for  the scope of ge'x as
an object. That would go a long way toward making it transparent.
This is a more general solution than creating new lujvo on a case by
case basis. It would give a choice on sumpti places that do not take
objects but could, and make definite those where a choice is offered.

djer