Robin Lee Powell wrote:
On Thu, Nov 09, 2006 at 09:17:49AM -0500, Bob LeChevalier wrote:When a computer starts parsing your speech stream and makes mistakes because you stuck a la in there, ***then and only then*** will we find out whether people can overcome the bad habit of slipping up on the la rule.If you truly believe that, then you've just admitted that Lojban is not a language that can be spoken by humans, to humans, or at least not solely. If that's the case, we might as well give up now.
No. All I have admitted is that certain kinds of ad hoc word formation in the process of translation are prone to error. Word making in translation is a major part of the language now. In the long run, it will be very minor. Ad-hoc word making is entirely orthogonal to speaking the language correctly.
Making Type IV fu'ivla is probably equally prone to serious errors that violate the rules of the language, and which even skilled Lojbanists who would detect "la" in names wouldn't see the violations in fu'ivla. But we managed to come up with Type III fu'ivla so that people wouldn't have to do so, and we discourage Type IV fu'ivla making (without a computer tool to check the resulting wordform).
So maybe we'd need a computer tool to check newly Lojbanized cmene, but once the cmene is made, the language can be spoken with no problem.
lojbab