Date: Wed, 21 Dec 1994 15:47:13 -0500 (EST)
From: jorge@PHYAST.PITT.EDU
Subject: Re: ni'i vs naja
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Reply-to: jorge@PHYAST.PITT.EDU
Message-id: <01HKX314EJWM00068X@FIPORT.FUNET.FI>
Content-transfer-encoding: 7BIT
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Length: 2351
Lines: 66

la kris cusku di'e

> "nu" seems to have become a general-purpose abstractor in common usage.  If
> that's not the intent, and the reference paper is wrong here, I'll have to
> be more careful, because I think I do that too.

nu is the general-purpose event abstractor, which can be made more
specific with mu'e, pu'u, zu'o and za'i. I don't think nu can stand for
du'u or ka, which are the other two general types of abstraction.

{jei} is not an abstraction. {si'o} I put together with {du'u}, not being too
sure of how they differ. {li'i} is looking more and more like {ka}, the
more I think about it, the difference could be that one is for "state"
type gismu and the other for "action" type, or maybe permanent and
non-permanent. {su'u} I don't know.

nu   (mu'e, pu'u, zu'o, za'i)
du'u (si'o)
ka   (li'i)

jei  (not really an abstraction)
su'u ?

I've seen {su'u} in use only a couple of times, and it seems to be used
for "how". This seems to be an indirect question, maybe {le su'u broda}
could be {le du'u taimakau broda}. But there isn't enough usage to tell.

I don't remember the grammar paper saying that nu is general purpose for
all types of abstractions.


> >[...]what do you use when you are really
> >talking about logical implications?
>
> If you mean complete syllogisms, like your example of the 4-legged dog, I've
> never yet written one in Lojban.  I would probably use ni'i.  Admitting
> ignorance, I question whether "logical implication" means a complete
> syllogism or just the last step in a logical argument.

Probably {ja'o} can be used for the last step.

I don't really know how logical {ni'i} is supposed to be either, I read
the same paper you did and got a different impression. I didn't even
think that {krinu} was a moral justification.

>  the correct table is:
>
>         X nibli Y
>         Y ni'i ledu'u X
>         Y .ini'ibo X
>
> I guess either one works just as well -- its a matter of remembering which
> way it goes.

Yes, but this convention has some funny consequences. The forethought
version of {Y .ini'ibo X}  is  {ni'igi X gi Y}, while for logical connectives
{Y iju X} goes to {gu Y gi X}. It would be better if things were more
consistent.

Also {Y pu lenu X}, goes to {X ipubo Y}, another contradiction.

The way you had it before is much more consistent with the rest of the
language.

Jorge