Re: [lojban] a construal of lo'e & le'e

[And Rosta <arosta@uclan.ac.uk>]

>>> Rob Speer <rob@twcny.rr.com> 10/29/01 09:54pm >>>
#On Mon, Oct 29, 2001 at 04:32:25AM -0500, pycyn@aol.com wrote:
#> To be safe, say {ije} instead of 
#> {i} the second time (why the first, by the way?). 
#
#In writing multiple sentences, I am in the habit of using some separator
#to begin the first sentence. The reason is that otherwise the first word
#would be at the "start of text" which does weird things to the scope of
#attitudinals. (People didn't believe me the first time I mentioned this
#- it was in the heat of the attitudinal debate and they thought it was
#my own proposal - but the Book says that an attitudinal at the start
#of text applies to the entire text, so if you want it to be an
#ordinary attitudinal which applies to the sentence, you have to put .i
#before it.)
#
#Even though there was no attitudinal in this case, I find it useful to
#simply be in the habit of beginning the text with some sort of
#separator.

Interesting. I had tended to think of .i more as an end-of-sentence
marker like a full stop (e.g. does it go before or after a paragraph boundary), 
precisely because it's not required at the start of the first sentence. But
your remarks show me to be mistaken.

#Anyway, I'm a bit unclear on why .ije would make a difference. Does {.i}
#remove the assignments of {da}-cmavo? If so, why do people think {da'o}
#needs to be improved?

Because the first {da} = {su'o da}, and (pace pc) the quantifier goes to 
the prenex of the bridi it occurs in . The su'o can only bind variables within
its scope (= elements following it in the bridi it occurs in), so it cannot 
bind variables in following sentences.

If {ije} does indeed allow binding to cross sentence boundaries, then this
would require some special rule to get the su'o to have scope over the
je. On reflection, I think the default position is that absent any such special
rules, variable binding can never cross sentence boundaries.

--And.