Message-Id: <199309071626.AA18148@eli.CS.YALE.EDU>
Date:         Tue, 7 Sep 1993 12:23:45 EDT
Reply-To: Jorge LLambias <jorge@PHYAST.PITT.EDU>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Jorge LLambias <jorge@PHYAST.PITT.EDU>
Subject:      Re: TEXT: The Terrifying Adventure of the Windmills
To: Erik Rauch <rauch>
Status: RO

la veion cusku di'e to ki'e doi veion toi

>
> I have gone through part of your translation and checked it with
> John Cowan's parser. The parser found two sentences which didn't
> parse.
>

That gives a good average !! :)

> Here are my comments:
>
> i ca le nu ko'a fo'a co'a viska cu cusku fi le ko'a santyse'u fe lu
>
> ***   cusku-observative. I assume '..viska ku ko'a cusku..'

Yes, I did it on purpose. "ko'a" appears in almost every sentence, so
I elided it as often as the context makes it clear that it refers to him.
What do you think about this? In general, I prefer to avoid excessive
precision, unless it is required to avoid confusion.


> ***   santyse'u = umbrella servant, using 'santa' for 'shield'.
>       How about 'xacysanta'? (PS. I noticed further down that you
>       got the same idea :-)
>
Again, a strive for conciseness. But in this case {xacysantyse'u} may be
required. What I wanted was a word for "squire".


>         be'ucu'i le su'u le selfunca ca'o gidva le mi'o cuntu cu zabna
>         zmadu le su'u mi'o pu'i pacna
>         i va le nu facki fi ji'iza'uso'u cilce brabracrida kei do viska
>         doi pendo po'u la santcos le pu'o se damba be mi
>         ije go'i le te lebna be le ka jmive bei mi
>         ije go'i le te vimcu be fo le mi'o co'a se ricfu
>         i vu'e le nu jamna le go'i cu vrude
>         i xamgu selfu le cevni le nu vimcu le palci tsiju fo le terdi
>           sefta
>
> *** 'leaving as remainder the surface of the earth'

{fo} should be {fi}, which is not required, so just remove {fo}

>
>         li'u
> i ko'e cusku lu
>         ma brabracrida
>         li'u
> i le ko'e selse'u ko'e spuda lu
>         do viska le go'i noi birka clani
>
> *** took me a while to catch 'which are long of arms'

Would {clani se birka} be better? I liked {birka clani}.

>
>         i ca'a se cafne le nu le ri so'o birka cu ki'otre piso'a li ci
>
> *** must be 'li piso'aci'

As you can see, I still have to look more carefully into the quantifier
business.

>
>         li'u
> i ko'e ko'a spuda lu
>         ju'i doi nobli le va simla goi fo'a brabracrida na.e bifymlo
>
> *** ?simla = simsa
>     'le va simsa' = 'those who at a medium distance seem to be similar
>                      to something'

I said it on purpose in this strange way, for various reasons. The first
is that the original uses a strange phrase, at least from the modern
Spanish point of view. I think the meaning of the lojban is still right,
(changing na.e to nagi'e). Sancho doesn't say what they seem to be similar
to (actually, he doesn't talk about similarities at all), and this is the
appropriate attitude, given the relationship he has with Don Quixote.
He refers to them as something that "seems", so that Don Q's confusion is
understandable. (Of course it isn't, but Sancho says it as if it were.)
"Those that over there look like something, are not giants, they are
windmills" in English sounds strange, but in lojban doesn't, because the
"something" is not mentioned explicitly. I think it's great to be able to
use "transitive verbs" without an object.

The second reason may be sillier. In this phrase Sancho defines what fo'a
refers to. Later Don Quixote uses fo'a, but he doesn't mean "those who
seem to be giants", he means "those giants", so not saying what those
seem to be, simply that they are over there seeming, permits each to
comfortably use fo'a for the same referent.

>     I think you mean
>
>      le goi fo'a va simsa be le brabracrida cu brabracrida nagi'e
>      bifymlo
>
>      or reversing the bridi-tails
>
>           ... bifymlo gi'enai brabracrida
>
> *** ?na.e  => nagi'e  for bridi-tails : 'aren't giants but windmills'
>
>         i le fo'a simla be lo birka cu balre le molki
>
> *** ?simla = simsa
> *** ?balre = cutting blade? In Finnish we would use 'nalci'

I couldn't find anything better, and it hardly seemed to merit a lujvo.
I can change it to 'nalci', but its metaphorical meaning doesn't fit with
this extension. 'nalci' seems to be wings that are on the sides of the
winged object. I did not know that the "cutting" was part of {balre}, if it
is then it doesn't work very well. I'll give it some more thought.

>
>         i le go'i ca le nu ca'o se catke le brife cu gasnu le nu le
>         molki rokci cu carna
>         li'u
> i ko'a ko'e spuda lu
>         za'a do na certu le ve capli'u
>         i fo'a brabracrida ju'o
>         i i'ecu'i xu do terpa paunai
>         i ai ko sezmu'u fi ta
>
> *** ? 'do so as to make you moved there by something'

sezmu'u : m1 (m2=s1) m3 m4 m5  x1 moves (self) to x2 from x3 via x4

(I guess you read selmu'u)

>
>         i ko jdasku ca le nu mi fo'a damba co cilce je na'e simsa
>
> *** ? 'fight wildly and other then similar to something'

Yes, simsa is lousy. It's a "fierce and uneven battle". Any suggestions?
My brain went dead on this one.

>
>         li'u
> i ca le nu ca'o cusku di'u kei ko'a jafti'e minde le xirma po'u la
>   rosinantes
> i na'e jundi le ko'e krixa kajde be fi le nu le pu'o se gunta cu
>   bifymlo enai brabragrida
                      ---
Veijo pointed out in a previous private communication that I changed
from brabracrida to brabragrida halfway down the road.

>
> *** ? observative 'na'e jundi'

It can only refer to ko'a, or not?

> *** ? enai => gi'enai  for bridi-tails 'were windmills not giants'
>
Right.

> i ko'a birti le du'u fo'a brabragrida
> i seja'ebo na'e tirna le ko'e voksa gi'e na'e viska le ca'a me fo'a
>   ki'unai le nu vi zvati fo'a
>
> *** ? again observative 'na'e tirna ... gi'e na'e viska'
>
Again who else could it refer to? Also, this sentence is strongly linked
to the previous one.

> i ca'o cladu cusku lu
>
> *** ? observative
>
ko'a again.

>         le'o ko na rivbi fi'inai tolvri je palci danlu
>         i za'upa le nobli do na gunta
>         li'u
> i co'a brife ca le nu co'u go'i
> i le barda balre co'a carna
> i ko'a seri'i le nu viska le nu go'i cu cusku lu
>         mi do pu'o sfasa ki'unai le nu do muvdu le brika se klani be li
>         le brabracrida po'u la briareos brika
>
> *** ? brika => birka
> *** ? this doesn't parse due to 'li', you cannot have 'li le'.

I think this sentence is the one that took me the most time, and I could not
find a good solution. I definitely have to look more into the MEX.

>     I have no good alternative, but the following is the nearest
>     to your idea
>
>     le birka se klani be le le brabracrida po'u la briareos birka klani
>
>     which takes a few moments to figure out :-)
>
>     A more Lojbanic way is to use MEX:
>
>      le nu muvdu vei mo'e le le brabracrida po'e la briareos birka ve'o
>      birka
>
>     or reversing the possessive
>
>      le nu muvdu vei mo'e le birka pe le brabracrida po'e la briareos
>      ve'o birka

I hate terminators/brackets, but I'll use this one until I have a better
grasp on how to make comparisons. (Maybe I'll learn to love them.)

>
>         li'u
> i ca le nu ba'o cusku di'u kei ko'a sezpevdu'a le ko'a fetno'i po'u la
>   dulsineas gi'e pikci ri le nu sidju sera'a le ca ckape
>
> *** ? ko'a sezpevdu'a le go'a fetno'i =
>          he self-figurative-gives his female-noble called Dulcinea
>
>     OK. You have defined a lujvo
>
>        s1 sezpevdu'a s2 = s1 pe'a dunda le s1 sevzi ku s2
>
Something like commend oneself. Any better suggestions?

> i le xacysanta ca'o marbi ko'a
> ije le claxa'i cu bredi
>
> *** claxa'i = long weapon = lance, another possibility would be
>     kilga'axa'i = kinli grana xarci = sharp rod weapon
>
The only argument I can give to defend {claxa'i} is 3 vs 5 syllables.

> i ko'a fo'a co'a klama
> i ko'a gunta le pamoi ca'u molki
>
> *** ? This doesn't parse because 'pamoi' is a selbri. You get the
>     intended meaning by
>
>        le pamoi pe le ca'u molki
>
>     or
>
>        le le ca'u molki ku pamoi

Numbers again! I'll pick the first suggestion for the time being.

>
> i ko'a darxi le molki le xarci le balre noi tsali se muvdu le brife
> ja'e le nu porpi gasnu le xarci gi'e farlu gasnu le xirma e le
> xirselma'erno'i noi carna re'o le foldi
>
> *** ? 'porpi gasnu le xarci', 'farlu gasnu le xirma e ...' will not
>     work as a tanru has the place structure of the last component,
>     i.e. gasnu in this case. You'll have to change the tanru structure
>     (le nu gasnu co porpi le xarci) or use lujvo (po'irgau and falgau).

The lujvo are good.

>
> ********************
>
> I got this far last night. I'll continue later.
>
>    co'o mi'e veion
> ------------------------------------------------------------------
>
>  Veijo Vilva       vilva@viikki21.helsinki.fi


I'll post a provisional improved version. I have to learn more about numbers
and comparisons. (You stopped just before the one with the donkey, which was
a tough one, but at least it parsed :)

co'o mi'e xorxes