From lojban@cuvmb.bitnet Sun Mar 24 10:23:43 1996
Received: from punt4.demon.co.uk by stryx.demon.co.uk with SMTP
	id AA10950 ; Sun, 24 Mar 96 10:23:41 BST
Received: from punt-4.mail.demon.net by mailstore for ia@stryx.demon.co.uk
          id 827634670:08887:0; Sun, 24 Mar 96 02:31:10 GMT
Received: from cunyvm.cuny.edu ([128.228.1.2]) by punt-4.mail.demon.net
          id aa08525; 24 Mar 96 2:30 GMT
Received: from CUVMB.CC.COLUMBIA.EDU by CUNYVM.CUNY.EDU (IBM VM SMTP V2R3)
   with BSMTP id 7345; Sat, 23 Mar 96 21:29:43 EST
Received: from CUVMB.COLUMBIA.EDU by CUVMB.CC.COLUMBIA.EDU (Mailer R2.07) with
 BSMTP id 1944; Sat, 23 Mar 96 21:30:17 EDT
Date:         Sat, 23 Mar 1996 18:16:02 -0800
Reply-To:     "John E. Clifford" <pcliffje@CRL.COM>
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@CUNYVM.CUNY.EDU>
From:         "John E. Clifford" <pcliffje@CRL.COM>
Subject:      Re: Opacity and "sisku"
To:         lojban list <lojban@cuvmb.cc.columbia.edu>
In-Reply-To:  <199603231750.AA29509@mail.crl.com>
Message-ID: <827634617.8525.0@cunyvm.cuny.edu>
Status: R


On Sat, 16 Mar 1996, ucleaar wrote:

> Djan:
> > There seems to have been some sort of confusion about the place
> > structure I proposed for "sisku", with a property in x2.  This
> > did not mean that "sisku" was a search for properties, but rather
> > a structure of the form:
> >    x1 searches for a thing/things with property x2 from field (set) x3
> [...]
> >        da poi bloti zo'u mi sisku le ka me da
> >        There exists a boat X such that I search for something with
> >                the property of being X.
> >        mi sisku le ka [ce'u zo'u] ce'u bloti
> >        I search-for-something-with the property-that {x : x is-a-boat}.
>
> I don't object to this structure in itself, but I don't see why {sisku}
> should be different from {nitcu}, {djica}, {kaltu}, etc etc. Are you
> able to persuade Lojbab to change the place structures of all opacifying
> gismu, so they match {sisku}?
> coo; mie and
>
Well, {sisku} and {kaltu} have the predicate of their nu-clauses already
built into their underlying semantics, so that a nu-clause argument will
not make sense, as it will with {nitcu} and {djica} and most (? I haven't
really counted, but many anyhow) of the opaquing predicates.  I have to
admit that, for most uses, {nitcu} could be taken to embody "to
possess/have access to something possessing the property...," on the model
of {sisku}.  But not always, and {djica} does not suggest any similar
deep structure at all.  By the way, I still do not like the {sisku}
solution much, but it is better than ignoring the problem or trying to
foist it off on the descriptor involved.
pc>|83