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

Re: [lojban] Re: "la" rule



Robert McIvor wrote:
On 13 Nov, 2006, at 19:49, Robin Lee Powell wrote:

On Mon, Nov 13, 2006 at 07:45:21PM -0500, Robert McIvor wrote:


On 13 Nov, 2006, at 17:36, Mark E. Shoulson wrote:

Robert McIvor wrote:


And the rule was altered to cover the LaPlace situation. I
believe  the current solution allows unambiguous parsing of
names (in  Loglan), and has only the restriction that names end
with a  consonant and a pause.

I'm still not fully understanding how the new Loglan rule covers
all cases. What about someone named, say {parmalagan}? Is {la
parmalagan} a single person, or is it {la parma la gan}, two
sumti,  the first being LA on a hypothetical predicate {parma}?


la parmalagan would be la Parmalagan, because la parma would need
to  be followed by a pause to be la parma la gan


Why is that, exactly?  That wasn't mentioned in your full
description.


There has to be a pause after a name, whether a predicate used as a name or a proper name. This signals the end of the name. The pause is before the second la. I presume when a predicate is used with la, it is intended to specify a particular instance of that predicate.

I believe that this is an example of what JCB called a "lexemic pause", where the pause has grammatical import as well as lexical import (in this case telling you that la parma la gan is two names. But of course "parma" would be detected by a lexer as a brivla and the pause is thus not part of "parma" but a separate lexeme. In Lojban names, the pause at the end IS part of the cmene.

This then causes a different sort of ambiguity, between a lexemic pause, and a hesitation or breath-pause that is otherwise legal between words.

lojbab