There are some instances where you can say "cu" or a pile of terminators. For
example:
.i lo zbasu be le minji be lo nu lo remna cu vofli bu'u la .kiti'ok. cu du la
.uilbr. joi la .orvil.
.i lo zbasu be le minji be lo nu lo remna cu vofli bu'u la .kiti'ok. ku ku ku
du la .uilbr. joi la .orvil.
.i lo zbasu be le minji be lo nu lo remna cu vofli bu'u la .kiti'ok. be'o be'o
ku du la .uilbr. joi la .orvil.
There is also a natlang, namely Tok Pisin, which has a word like "cu". It's
called the predicate marker, though being the only word in its selma'o, it
might as well just be called a particle.
Man i mekim masin bilong man i flai long Kitty Hawk i Wilbur na Orville.
(I don't know how Tok Pisin handles names of people and places in English
outside PNG.)
Pierre
--
La sal en el mar es m�s que en la sangre.
Le sel dans la mer est plus que dans le sang.
--To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/CuzwB6IY5XEJ.
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To post to this group, send email to lojban-b...@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginne...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban-beginners?hl=en.
yes cu does serve a purpose. It's a shortcut for "close everything that needs closing so that I can say the selbri now".
So in "lo ctuca cu citka lo plise" "cu" is a shortcut for plain old "ku" (not much of an advantage here). But it can come in handy in situations like "lo nu mi pensi lo nu mi cusku lu zenba cu se kakne". Here "cu is standing in for "li'u kei kei". Sometimes it's easier for lazier folk like myself to be able to say "ugh, that convoluted sumti is done, now I want to do the selbri... shoot what all needs closing? Bah forget it. 'cu'. "
"le cevni" is "what I describe as god" rather than "your perception of
god". The latter would be more like "lo simlu be lo ka cevni bei do".
mu'o mi'e xorxes
I like the Lojban better than the English, which means, I think, that you're
thinking in Lojban. It's really hard to render the Lojban into any natlang I
know; the closest I can come offhand is Spanish "te amo y Dios cu�dete" (I
attend a Hispanic church, so much of my thinking about God is in Spanish).
Adding to what Lindar said, the fa'orma'o is, as far as I know, a part of
speech unique to the Loglanic languages, so you need lots of practice to get
it right. "cu" has an equivalent in Tok Pisin; attitudinals exist (though not
the same list) in some language spoken in the NE India/Burma area. Terminators
are what make it possible to say lots of grammatical constructs unambiguously.
Pierre
--
.i toljundi do .ibabo mi'afra tu'a do
.ibabo damba do .ibabo do jinga
.icu'u la ma'atman.
--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/Zhg_0FOinHEJ.
I actually said "iure'e le cevni ku kurji ko" when I read it aloud,
anticipating the "ku-" of "kurji". The short version is also made entirely of
amphibrachyes (if "r�'e" is so stressed), which is pe'i the most natural foot
for Lojban. I've come up with some sayings in amphibrachic n-meter, such as
"lo drata cu muvdu .ijo lo se drata cu klama fu zi'o".
Pierre
--
Jews use a lunisolar calendar; Muslims use a solely lunar calendar.
> Well, I wouldn't say that "ca" (or any of selma'o PU) can replace "cu"
> unambiguously. Take the following sentence:
>
> lo ca klama ca pinxe
>
> In this case, the first "ca" attaches to the sumti to make it mean
> something like "the one currently going", where the second "ca" attaches to
> the selbri of the entire bridi, for a total meaning of "the one currently
> going is currently drinking", possibly alluding to someone drinking while
> driving or biking.
>
How do you identify "pinxe" as a selbri in this case at all, i.e. what stops one from reading the whole "ca klama ca pinxe" as a tanru (regardless of how it would make sense in translation)? Is this really ambiguous, or just plain wrong (the selbri reading, I mean)?
Klaus
> Alright, I found it (I think). CLL C 5, S 13, last sentence: " Tense,
> modal, and negation cmavo can appear only at the beginning of the selbri.
> They cannot be embedded within it. " That means that we can have our
> descriptor of selma'o LE, followed by the slot for a PU cmavo (or,
> technically, NA or modal or other tense cmavo), followed by exactly on
> selbri, be it a tanru or just a single brivla. That leads to A) an
> unambiguous interpretation of "lo ca klama" as "the current goer", and B)
> of "lo klama ca pinxe" as "the goer is currently drinking", and C) of "lo
> ca klama ca pinxe" as "the current goer is currently drinking", since the
> "ca pinxe" cannot be absorbed as part of the tanru. Mystery solved.
>
Oh, I see! Learned something again, thanks. May the rest of your night hours be peaceful.
Haven't said a word in here for too long, so that I forgot to sign with my already known jbocme.
mu'o mi'e feliks.