Re: Mad Proposals II: The watered down version.

la xorxes. cusku di'e

> I never claimed {gijoi} is more meaningful than {.ijoi}, but it's not less
> meaningful either. When we discover what {.ijoi} means, we'll know what {gijoi}
> means (it's not just an expansion, but their meanings are related).

Related, yes, but not necessarily in an obvious way.  In particular, all the
JOIs are defined for sumti.

> Or are we going to eliminate all constructions for which we don't know the
> meaning yet?

Actually, one of the reasons the papers have been going so slowly is that
I don't know the meaning of every construct yet!  But we have eliminated
some (NAhE NUhI, NAhE KI, e.g.) because no discernible meaning existed.

> > The only ijoik explained in my reference grammar is ".ice'o", which separates
> > the elements of an ordered list of bridi.
> 
> "gice'o" would have a very similar meaning (from the same example):
> 
> {mi ba kanji lo ni cteki kei gice'o lumci le karce gice'o dzukansa le gerku}
> 
> is just as meaningful as

[example omitted]

This looks like explanation-by-expansion, but we know that this does not
work for non-logicals.  Why should non-logical bridi-tail connection be
explained by non-logical bridi connection?  After all, non-logical sumti
connection and non-logical tanru connection
are known to be independent in meaning (though intuitively related).

> They have the same natural semantics that non-logical bridi connectives have,
> since bridi-tails are just a type of bridi. If they don't have a place in the
> language, neither do non-logical bridi connectives.

Actually, I'm dubious about even the ".ice'o" usage, but Bob suggested it,
so in it went.  I think that jek/joik interchangeability is probably a
mistake, adopted for simplicity but not really semantically sound.

> And here's a possible example with {gijo'u}:
> 
> mi zgana le se tivni gijo'u citka le cidja
> "I watch the TV program along-with eat the food"
> 
> Any doubt what that means? I think {gijo'u} is what is meant in many cases
> that {gi'e} is now used, because there's no other option. {gi'e} makes the
> two claims without establishing any connection (other than the logical one)
> between them, while {gijoi} and company make a single claim, composed of
> subclaims that are not claimed separately.

I don't see it.  This seems as clear-cut a logical connection as any:
you watch-and-eat just in case you watch and (.ije) you eat.

> Here's another one:
> 
> mi'a cinba vo'a gijoi dasgau vo'a noda
> 
> I'll let you all figure out what that one means.

Means bugger-all to me.

	Part-of-the-mass-of-us-(excluding-you) kisses part-of-the-mass-etc.
		[forming a mass claim with?]
	part-of-the-mass-etc. is-an-agent-in-the-wearing-by part-of-the-mass-etc.
		of-zero-things.

Note that the morphology rules demand "dasygau".

This gives a general impression of kissing while undressing, but I fail to see
the precise significance of the "gijoi"/".ijoi" here.  Looks more like ".ije" or
".ica".

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.