Re: mine, thine, hisn, hern, itsn ourn, yourn and theirn (was[lojban] si'o)

[Nick Nicholas <nicholas@uci.edu>]

cu'u la xorxes:

>>I mean, honestly, if you ignore the gismu list's definition of {moi},
>>"*ordered* by rule x3", then there is in principle no reason why you can't
>>use any 3-place gismu for any other.

>I don't ignore it, I generalize it. If we order all hands by
>who they belong to, which hand do you think will correspond to
>Alice?

...

>{be fi le ka ce'u xance makau}, the sequence defined by whose hand
>the hand is, indexed by behanded beings.

OK, *now* I see what you're doing. I still loathe it, but it's an
interpretation that can be argued for: where you don't have numerals as
ordinals, treat the 'ordinal' sumti value as an index. In Perl terms,
associative array instead of numerical array. Associative arrays are not
ordered, per se, but they are indexed.

Hm. Associative array-like concepts have not been much explored in Lojban
(John?); you can say hand[3]==Alice's hand, but it's harder to say
hand{Alice}==Alice's hand. Of course, that's because we've used {pe} for
that instead.

Well then. I still think it's butt-ugly, but it's not as insane* as I first
thought. (For 'insane', read 'ill-motivated, capricious, and
unformalisable'.)

(Do not construe this as approval of {memimoi} in general, btw. And Lojbab
and John, if you'd meant VEI... MOI, I wish to God you'd actually used
VEI... MOI. In fact, can't a VEI expression stand in for a numeral
everywhere anyway? {li vei pa su'i re}? If so, you didn't need to do this
at all.)

Oh, one more thing? If this is how you do it, then I don't think sumti are
indexes, but quoted words are. I.e. not {me mi moi}, but {me zo mi moi}.
The 'mi' is being used as a {tcita}. But {mi} am not a {tcita}; {zo mi} is
a {tcita}. Just like it's not {lo cimei} that's doing the labelling in {ci
moi}, but {li ci}, the abstraction. We do say {ny.moi}, not {me'o ny.moi};
but because {ny.} really is being dereferenced here (value for which n
stands), I don't think that's a valid counterexample.

Oh, I know {zo} will be judged as terribly formalistic, etc. etc. etc.,
because it doesn't say what you want to say. Whatever.

So:

la .oslos. pamoi le'i trutca leni berti

*Winces, scowls, kicks himself because he knows xod's going to love what
he's about to say*

la .oslos. me [zo/le] gugdrnorge moi le'i trutca
leka ganai ce'u ny.moi vo'e vo'i gi ce'u trutca [la'eny./ny.]

I don't know if that {leka ce'u se tructa} will work for "'ordinal' label
of this capital is the country of which this is the capital" ---
particularly with that nasty recursion I slipped in there with {vo'i}. If
it doesn't, xorxes, I think the responsibility is yours to find a
{ce'u}-expression that will; and only then can I accept this as a
legitimate expansion of {moi} (which I still won't use. :-) ). Because you
are not inimical to formalisation, I think you will...

>i xu zu'i pe le gencukta cu du le finti be gy

.i mi jinvi ledu'i le vamji be zo zu'i cu jalge le steci ke selbacru jibni
(to zoigy. context gy.)

.i ku'i kakne fa lenu mi srera .eni'ibo lenu le vamji be zo zu'i na se
jitro pa steci ke selbacru jibni

.i ganai go'i gi mi stidi zo zo'e po'o gi'e tugni do tu'a zo zo'e

Nick Nicholas,  TLG, UCI, USA.   nicholas@uci.edu    www.opoudjis.net
"Most Byzantine historians felt they knew enough to use the optatives
 correctly; some of them were right." --- Harry Turtledove.