From pycyn@aol.com Tue Nov 13 07:43:08 2001 Return-Path: X-Sender: Pycyn@aol.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_0_0_1); 13 Nov 2001 15:43:08 -0000 Received: (qmail 99159 invoked from network); 13 Nov 2001 15:43:07 -0000 Received: from unknown (216.115.97.172) by m3.grp.snv.yahoo.com with QMQP; 13 Nov 2001 15:43:07 -0000 Received: from unknown (HELO imo-r02.mx.aol.com) (152.163.225.98) by mta2.grp.snv.yahoo.com with SMTP; 13 Nov 2001 15:43:07 -0000 Received: from Pycyn@aol.com by imo-r02.mx.aol.com (mail_out_v31_r1.8.) id r.aa.19ef501 (4403) for ; Tue, 13 Nov 2001 10:42:54 -0500 (EST) Message-ID: Date: Tue, 13 Nov 2001 10:42:54 EST Subject: Re: [lojban] Why is there so much irregularity in cmavo/gismu? To: lojban@yahoogroups.com MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="part1_aa.19ef501.2922997e_boundary" X-Mailer: AOL 6.0 for Windows US sub 10535 From: pycyn@aol.com X-Yahoo-Profile: kaliputra X-Yahoo-Message-Num: 12083 --part1_aa.19ef501.2922997e_boundary Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit In a message dated 11/13/2001 1:26:11 AM Central Standard Time, thinkit8@lycos.com writes: > It's one thing for English to have its quirks, but one competing to > replace it should be as regular as possible. Here's an obvious > example, with modals. > > mukti=mu'i > mupli=mu'u > > Why the irregularity? Maybe because someone decided to make a Hindi > word thousands of years ago that didn't jive with a Chinese one made > even earlier. > > As far as I'm concerned, all the cmavo, gismu, and rafsi should be > redone so they are much more systematic. For true cultural > nutrality, make them more or less random within a systematic > framework. > Well, it depends on what system you have in mind. I assume that your problem is not (as the example suggests) that cmavo don't have very clear relation to the "corresponding" gismu, since the phonology explains that. So the problem is why are similar looking words so unrelated in meaning (and conversely). In this case the system is not a semantic one but a heuristic one: Lojban is supposed to be easy to learn and gismu are meant to be easier to learn if they have bits and pieces of the "corresponding" word in your native language. There was a formula for this and the gismu list is one kind of optimization under that formula (which includes a few other factors, to be sure). Now, the plausible theory about learning has not turned out too well, but it is a part of Lojban and so we are stuck with the results. If you want a language where words that mean similar things have similar forms, there are many around (try AUI, which is about the purest semantic prime language I know of -- it has no other virtues that I know of and is even unchangeably decimal). --part1_aa.19ef501.2922997e_boundary Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: 7bit In a message dated 11/13/2001 1:26:11 AM Central Standard Time, thinkit8@lycos.com writes:


It's one thing for English to have its quirks, but one competing to
replace it should be as regular as possible.  Here's an obvious
example, with modals.

mukti=mu'i
mupli=mu'u

Why the irregularity?  Maybe because someone decided to make a Hindi
word thousands of years ago that didn't jive with a Chinese one made
even earlier.

As far as I'm concerned, all the cmavo, gismu, and rafsi should be
redone so they are much more systematic.  For true cultural
nutrality, make them more or less random within a systematic
framework.


Well, it depends on what system you have in mind.  I assume that your problem is not (as the example suggests) that cmavo don't have very clear relation to the "corresponding" gismu, since the phonology explains that.  So the problem is why are similar looking words so unrelated in meaning (and conversely).
In this case the system is not a semantic one but a heuristic one: Lojban is supposed to be easy to learn and gismu are meant to be easier to learn if they have bits and pieces of the "corresponding" word in your native language.  There was a formula for this and the gismu list is one kind of optimization under that formula (which includes a few other factors, to be sure).  Now, the plausible theory about learning has not turned out too well, but it is a part of Lojban and so we are stuck with the results.  If you want a language where words that mean similar things have similar forms, there are many around (try AUI, which is about the purest semantic prime language I know of -- it has no other virtues that I know of and is even unchangeably decimal).    --part1_aa.19ef501.2922997e_boundary--