From pycyn@aol.com Tue Jun 06 09:57:12 2000 Return-Path: Received: (qmail 14489 invoked from network); 6 Jun 2000 16:57:11 -0000 Received: from unknown (10.1.10.27) by m1.onelist.org with QMQP; 6 Jun 2000 16:57:11 -0000 Received: from unknown (HELO imo-d03.mx.aol.com) (205.188.157.35) by mta2 with SMTP; 6 Jun 2000 16:57:11 -0000 Received: from Pycyn@aol.com by imo-d03.mx.aol.com (mail_out_v27.9.) id a.9.65ef34c (4233) for ; Tue, 6 Jun 2000 12:56:56 -0400 (EDT) Message-ID: <9.65ef34c.266e8758@aol.com> Date: Tue, 6 Jun 2000 12:56:56 EDT Subject: Chinese cmene still 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 Yuen Ren Chao, master linguist and egg-scrambler, wrote a paper on the non-uniqueness of phonemic solutions which drove a major nail in the coffin of positivist linguistics many years ago. In it he presented the phonetic facts of his dialect of Mandarin (his wife is from Anwhei, I'm not sure about him) and then demonstrated that these data equally justified a goodly number of different phonemic solutions, some of which were further justified by other facts (e.g., Chinese Pig Latin, historical information, and so on). The point is that none of these is "the right way to describe Chinese." But all work if used correctly. Of course, each of them was a system devised specifically for that Chinese dialect. I am not sure (not having looked at Chinese data since about the time I read Chao, roughly 1962) whether Pinyin is actually a phonemic solution for any dialect or whether it incorporates a variety of elements, perhaps even inconsistent ones, in the interest of some political goals. But it is designed for Chinese and so does at least a fair job of representing Chinese sounds, when properly used. Lojban spelling, on the other hand, is designed for Lojban and, barring some open questions about a few sounds, does a good job for that language. But that language is different top to bottom from Mandarin Chinese. It is short a number of vowels and the ones it has are distributed very differently, its occlusives are voiced and voiceless rather than lax and tense voiceless (or unaspirated and aspirated), its affricates are palatal rather than often retroflex, its final nasals are labial and dental, rather than dental and velar, and Lojban has no tones.. Representing Chinese in this language is going to be very approximate at best and usually a clean miss in terms of actual sounds. The most that we can hope for is distinctive similarity and maybe some over all patterning. By and large we have that with the consonants as noted (the collapse of en and eng to /n/ aside -- that final fully form /g/ is even less appealing). The vowels remain a problem, especially a range of high vowels in front of /u/, back of /i/, above /y/ and with aberrant lip positions for the tongue positions, and the syllabic spirant(s), however described. By and large, the last of these can be dealt with simply as spirants, at the cost of a syllable: /djuandz/, /sma/ (plus a final C to be chosen later). The break in the Chinese pattern warns the aware that something odd is going on here and can be interpreted back to Chinese appropriately (the /ng/ could be as well, were it not ugly in its own right). That leaves the yeri and the umlaut, which can't be /ui/ or /iu/, since both of those are already used in the usual way. The other vowel combinations that might be suggestive are forbidden in a single syllable; writing them forces a /'/ between. The question then is how best to distribute these sounds out in a way that does the lest damage to the original (so that a knowledgeable person would have a reasonable chance of getting back to the original). Putting them all into /y/ is one possibility, another is spreading them into /i/, /u/, and /y/ according to some rule. Of course, putting things this way assumes a certain phonemic solution for the Chinese langauge involved -- there are other solutions which might solve the vowel problem nicely, but I recall them as creating matching problems then with the consonants, especially in the absence of tones.