Message-Id: <m0qaoDw-00004xC@xiron.pc.helsinki.fi>
Date:         Wed, 17 Aug 1994 12:53:29 EDT
Reply-To:     jorge@PHYAST.PITT.EDU
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Jorge Llambias <jorge@PHYAST.PITT.EDU>
Subject:      Re: xruti
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 1221
Lines: 37

la lojbab cusku di'e

> I would suspect that if "-gau" is causing problems with x2, you may need a new
> convention that does not cause x2 patient.

I guess you meant that you want a convention that _does_ make x2 the patient.
I think this may cause more confusion than anything else. Since the problem
is solved using {jai}, I wouldn't make an exception for just a few cases.

> Either that, or change "jai" to something that can get a useful rafsi, which
> will in turn probably cause some other rafsi changes.

It can get a useful rafsi as it is: "jaz" (or "jax"). Of course, something like
jal, jar, jam or jan would be even better, but you'd have to take it away from
something else.

Then we'd have:

jazga'i (jai galfi)  x1 changes x2 into x3 with event x4

As opposed to:

gafygau (galfi gasnu)  x1 changes, with event x2, x3 into x4


> Remmebering that -gau and -zu'e (which I presume also has the same problem)
> are unique among gismu in being designed to cause raising,

I guess one could define:

gafygau: gas1 (gas2=gal1) gal2 gal3
         x1 is the agent in an event that changes x2 into x3

but this would deviate from the usual way -gau works, and doesn't really
offer anything better than jai.

Jorge