Date: Tue, 16 Aug 1994 04:22:19 -0400
From: Logical Language Group <lojbab>
Message-Id: <199408160822.AA02079@access1.digex.net>
To: jorge@PHYAST.PITT.EDU
Subject: Re: xruti
Cc: lojbab@access.digex.net, lojban@cuvmb.cc.columbia.edu,
        nsn@krang.vis.mu.oz.au
Status: RO

Those may be the ones you have, but Nick has been working on a new list
with 800 more added already, and a queue of additional ones left to be done.

I can't do much with the old list anyway, since I find Nick's abbreviations
unusable.

It ran away from me today, but for a brief moment I had a good reason
against the change.  Ah, yes!

krefu is used to talk about events that recur, in which all sumti return to a
previous value forming a relationship that existed in the past.  No agent
is implied, and there is no focus on any of the sumti.

xruti puts focus on ONE of the sumti, which becomes such that the relationship is restored to the previous value.  If the only thing about the relationship
is this focus, though, I think that there are ways of indicating that focus.

On the other hand, if the need for the focus is compounded with the need for
an added agent who causes the return to previosu state by acting on the
one sumti that is the focus - then we have the xruti that is in the gismu list.

So there are two features that distinguish trhe current xruti from krefu.
You seem to propose eliminating one of them.

Now the question becomes whether a -gau lujvo adds an agent while preserving 
the focus on the x1 place.  I find this doubtful.

Somewhere in all this, my mind is drawn to stika/cenba and galfi/binxo as
being more relevant than gasnu tot his question.  These words have not seen
much use in the language, especially in lujvo, and I think that may be a
shortsightedness on the part of the people who have been making ad-hoc lujvo.

At any rate, I seem to have drifted, and I am not sure whether my unease at
making a change to xruti has come across along with any cogent reasons for same.
Of course I was having similar unhappiness today over agentless muvdu, which
seems rather too much like klama these days - I think there should be some
clear semantic distinction but am not sure whether the additional x5 place in
klama provides one which is meaningful.

So I will stop now and reiterate that Nick needs to comment based on his
understandings on the semantics of the gismu as they seem to be being used in
lujvo, as well as from his experience as the most prolific and fluent Lojbanist.
Others can also comment, but a strong opinion from Nick has particular
signififcance especially since we have nominally frozen the lists for the
duration of dictionary writing (I can still make the change but it is
starting to get REAL hard to do so, and will shortly become almost impossible
without causing a lot of data-thrashing.  (We'll have to deal with such
anyway once the dictionary is done and places change, but I won't then be
still trying to WRITE the damned thing.)

I still want opinions from the rest of the community too - so far only Jorge 
has spoken, and that is not enough to make a place struture change.

lojbab