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

Re: 'nai' and .UI and ambiguity



At 04:35 PM 02/05/2000 -0800, Jorge Llambias wrote:
>la adam cusku di'e
> >Since 'nai' has more uses than just within an indicator phrase,
> >allowing it anywhere would cause ambiguity, since in some cases
> >we wouldn't know whether it's negating an unspecified .UI or serving
> >some other role.
>
>Could you give an example? I think the reason {nai} was
>not put in UI is because it was felt that the parser had
>to identify things like {jenai} as a word different from
>{je} and the parser doesn't "see" any UI, but not because
>there could be any ambiguity. How could there be?

Those compounds that are recognized in the lexer (grammar rules above 800) 
cannot have any non lexer structures embedded.  jenai is one of these lexer 
compounds. I would not go so far as to call them "single words" - the 
construct "no fiho pumod" comprising a complete tense can be so complex 
that no one would ever consider it a single compound.  But originally you 
could not put a UI anywhere in such a lexer compound at all.  Now you can, 
because Cowan wrote the parser to filter out UI in advance of the main 
lexer rules.  So if you put ui in between je and nai, the nai will be taken 
as part of the attitudinal.

> >Speaking of which, how are we to interpret the pattern
> >
> >       broda je ge'e nai brode
> >
> >Does the 'nai' polarly negate the indicator, or does it pick up
> >after the indicator ends and contradictorily negate brode?
>
>It negates the indicator. Always the immediately preceding word.

Correct.

lojbab
----
lojbab                                lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA               703-385-0273
Artificial language Loglan/Lojban:  http://www.lojban.org