[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Aorist



   Date:         Wed, 9 Oct 1991 11:40:18 EDT
   From: John Cowan <cbmvax!snark.thyrsus.com!cowan%UUNET.UU.NET@cuvmb.cc.columbia.edu>
   X-To:         lojban <lojban@cuvmb.cc.columbia.edu>

   [I have moved this post from conlang to Lojban List.]

   The Lojban evidentials are outside the grammar (can be inserted anywhere)
   and always refer to the speaker.

Welllll, this is the basis of the just-mentioned discussion among me,
Lojbab, and Nick.  I had said that "ko'a djuno lenu zo'ekau cu co'e"
(=he/she/it/they know(s) the-event: something-unspecified(known!) does
something-unspecified.) would mean that the *speaker* knows who or what did
something, which lent itself to a nice idiomatic construction in the
sentence I was working with (meaning "there was something in particular
that did it, and I know what it was").  Nick interpreted "djuno" as sort of a
"preprocessor instruction", shifting the attitudinals to ko'a, not the
speaker, even though no direct quotes were used, thus getting the meaning,
"he/she/ knew what did something."  Bob agreed with Nick, though agreed
that it wasn't cut-and-dry, and suggested that in cases which might be
confusing, one should use the "self-oriented" and "other-oriented"
attitudinals (se'i and se'inai), which specifically relate to the speaker,
to redirect the other attitudinal(s).  

           la sofis. djuno le nu za'a la .artr. klaku
           Sophy knows the event-of [I observe it!] Arthur weeps

So by the usage suggested by Bob and Nick, this observation would likely
apply to *Sophy*, not to the speaker.  To get the meaning you want without
ambiguity, one way is "la sofis. djuno lenu za'ase'i la .artr. klaku",
while expressing the other reading more carefully could be done with "la
sofis. djuno lenu za'ase'inai la .artr. klaku".

   To say what Sophy observed, it is necessary to use the Lojban metalinguistic
   indicator "sei".  This marks a subordinate claim that comments on the main
   claim, in the same way that the evidentials do, but with greater semantic
   generality:

           la sofis. djuno le nu la .artr. klaku sei ra zabna
           Sophy knows the event-of Arthur weeps (the former observes it)

Important point:  "zabna" means "favorable connotation".  "zgana" is the
predicate for "observes" (lojban have easily-confused words?  Nah.)  

   This may be used quite generally:

           la .and. rostas. sei la lojbab. .e mi na krici cu merko
           And Rosta (Lojbab and I don't believe it!) is an American.

   I would use this sentence if reporting someone else's opinion.

I think there's no problem with this sentence, but you should realize that
"sei" has a terminator "se'u", which may not be elidable.  I expect this
one to be present more often than many other elidable terminators.  For
reporting someone else's opinion, I'd probably use jinvi ("opine") rather
than krici ("believe"), but that's a fuzzy distinction.  An easier way, of
course, is to use the attitudinal "pe'i" (I opine), with "nai" (negator) or
better still "cu'i" (scalar neutral), thus "pe'icu'i" means "(I) don't
necessarily opine"--a very good standard disclaimer for someone else's
opinions.

~mark