From cbmvax!uunet!math.ucla.edu!jimc Mon Apr 6 17:37:27 1992 Return-Path: From: cbmvax!uunet!math.ucla.edu!jimc Return-Path: Message-Id: <9204061658.AA03245@julia.math.ucla.edu> To: cowan@snark.thyrsus.com (John Cowan) Subject: Re: Operator precedence in Lojban mathematics In-Reply-To: Your message of "Thu, 02 Apr 92 10:27:54 EST." Date: Mon, 06 Apr 92 09:58:33 -0700 Status: RO > > Despite PC's fulminations, Lojban is ideally suited to the concept of a > > "number" as "an equivalence class of sets all of which have the same > > number of elements". Then lo'i cimei becomes "the number three". > > Yes, but what if you like Von Neumann numbers? This is analogous to what PC says: Lojban has to be culturally neutral. My response: it doesn't bother me that one model of numbers automatically fits one of the grammar constructs :-) The real question is, should "li ci" be made illegal? In other words, once one recognizes that one model of numbers is supported in the grammar, should the language designer say "OK, that's enough, I'm not going to put in any more features to support numbers"? A principle of Lojban (not loved by all designers) is "kill one bird with many stones", and so clearly the Lojban designers will retain "li ci". But how does "li ci" represent Von Neumann numbers? Only in that it represents whatever number model that pleases the speaker. A big plus for "lo'i cimei" is that its meaning is unambiguous. I would hope that MEX grammar would accept "lo'i cimei" without a lot of gyrations, not being restricted only to "li ci". -- jimc