From pycyn@aol.com Thu Jun 22 11:04:36 2000 Return-Path: Received: (qmail 27116 invoked from network); 22 Jun 2000 18:04:34 -0000 Received: from unknown (10.1.10.26) by m2.onelist.org with QMQP; 22 Jun 2000 18:04:34 -0000 Received: from unknown (HELO imo-r18.mx.aol.com) (152.163.225.72) by mta1 with SMTP; 22 Jun 2000 18:04:33 -0000 Received: from Pycyn@aol.com by imo-r18.mx.aol.com (mail_out_v27.10.) id a.bb.47b1ee3 (657) for ; Thu, 22 Jun 2000 14:04:18 -0400 (EDT) Message-ID: Date: Thu, 22 Jun 2000 14:04:17 EDT Subject: Re: [lojban] on Lojban pronunciation To: lojban@egroups.com MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: AOL 3.0 16-bit for Windows sub 41 From: pycyn@aol.com X-Yahoo-Message-Num: 3192 In a message dated 00-06-22 10:31:21 EDT, aulun writes: << the Lojban /x/ was to pronounce like German 'ch' in 'Bach' or Spanish 'jinete' etc.. This for sure is the rule defined. (Meanwhile I have changed the audio file.) Since in German the pronunciation of the 'ch'-fricative changes from guttural (Bach=creek, Rache=vengence) to about central- palatal (ich=I, nicht=not) over somewhat in between (echt=genuine, schlecht=bad) depending on the accompaning vowel respective, I'd - for me - tend to pronounce it different for /xance/, /xekri/ and /xirma/. According to Lojban rules, the different pronunciations for /r/ (and I think - quite unexpressed and informal - /l/ too) is free for personal/national usage. This should be the same with /x/, because its variations indeed having no *morphemic* functions! (French - and maybe Chinese - speaking Lojbanists excluded for ambiguous /x/=/c/ (French) or /x/=/s/ (pinyin 'x'/W.-G. 'hs'). What is your opinion on this?>> It is certainly the case that /x/ tends to be pronounced differently in different environments and that is OK. It is even the case that it has different native language forms (English is often less gutteral, closer to /h/, German and Slavic presumably happier with the genuine gutteral fricative). What cannot be allowed to happen is that it falls over into the area of another Lojban sound, either /c/ or /s/ or whatever is used for /'/ (one of the reasons for suggesting theta for /'/). It is ultimately the patterning of the sounds, rather than the sounds themselves, that counts (as we remarked about transliterating Chinese). If your pattern for /x/ overlaps mine for /c/, say, I may have trouble understanding you for a few moments, but most of us (hopefully) adjust pretty rapidly, identifying your pattern. Maybe we should be sure there is a good vvariety of version up, when we get around to the serious job of providing models. <> No, just a practical one (disguised, as often with JCB, as an empirical discovery). We needed another sound, we were misrepresenting a lot of language contributions by lacking an /h/-ish sound, we needed a sound that would be distinct in usual channels (as ordinary /h/ is not) and, lo, we found that most languages had a /x/ but not an actual /h/. And so, /x/ it was.