Re: [lojban] Re: [humanmarkup] Lojban personal experience

["James F. Carter" <jimc@MATH.UCLA.EDU>]

On Fri, 23 Mar 2001, Pierre Abbat wrote:
> On Fri, 23 Mar 2001, John Cowan wrote:
> >But in general, *syntactic* ambiguities do not exist in Lojban...
> 
> It is possible to create a syntactically ambiguous sentence with {zo'u}. For
> instance: {la carlyt. zo'u mi klama} can mean "I'm going to Charlotte", "I'm
> coming from Charlotte", "I'm going by way of Charlotte", or "I'm taking the
> Charlotte" (if Charlotte is a vehicle). Anything that appears in the prenex can
> be put in any unfilled place in the bridi...

Is this really true? I'd interpret it idiomatically as ``speaking of
Charlotte, I'm coming'' (and the listener might pick up a clue for the
missing places of klama, but not by any rule of the language). With the
example of computer languages, it's like declaring a variable [*in such
language] and then not using it: the formal effect is the same as if the
variable (or prenex item) had never been mentioned.

That last sentence might be taken as a counterexample, but I'm taking it as
a reminder that there are steps beyond semantics in interpreting a
sentence, namely taking context into account, and unless the language
includes some pretty strong algorithms for interpreting context (and Lojban
doesn't), you're going to miss a lot of the meaning if you stop at
semantics. In this example I originally was going to say "it's like
declaring a variable and not using it"; then I said "the context is obvious
to me but maybe not to others, so I need a prenex to suggest the context";
and than I said "this duplicates Pierre's sentence; I'd better actually use
the prenex symbol within the main sentence."

James F. Carter Voice 310 825 2897 FAX 310 206 6673
UCLA-Mathnet; 6115 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA 90095-1555
Email: jimc@math.ucla.edu http://www.math.ucla.edu/~jimc (q.v. for PGP key)