[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
> la and cusku di'e
>
> >According the mahoste -- the gloss "most", and things like "so'e roi", it
> >does
> >mean a majority. so'a and so'e make no sense at all unless they are defined
> >in terms of ro. But the issue is whether so'o, so'i and so'u are also
> >defined in
> >terms of ro. If they are, then what do they all mean, and how do we say
> >"many" and "few", given that they aren't defined in terms of ro?
>
> I think we can at least say that they all have import
> {so'V broda cu brode} entails {su'o broda cu brode} for the
> five V. (In fact so'i, so'o and so'u probably all entail
> su'ore. "One" can't be "few", "several" or "many", and probably
> it can't be "most", but it could eventually be "almost all"
> when "all" is "two". Also, {so'a broda cu brode} and {so'e broda
> cu brode} each entails {me'iro broda cu brode}
I had thought that the fractional ones are probably nonimporting,
since most of the contexts you and Adam gave for where we might
plausibly want nonimporting ro apply equally well to so'e and so'a.
I discussed this point in the summary I posted to the wiki and the
list.
But actually, now I come to think of it, the same argument applies
to every quantifier:
Q messages I post to Lojban list each month are garbage
-- we might legitmately want this to be true even when there are
some months when I post no messages, and the argument holds for any
Q.
So it is rather silly to say that all quantifiers are nonimporting;
the argument must be invalid.
So yes, on reflection, all the so'V are importing, but the
truthconditions of importingness are somewhat unclear (to me
at least).
--And.