Roger Mills wrote: > Stephen Mulraney wrote: > Very interesting. Since I've just been working on Gwr, a language of similar > type (except I know nothing of Mandarin), I'm struck by the more than > occasional resemblances. Ah, that's what kind of a thing Gwr is; I wasn't sure. Actually, if you're not familiar with Mandarin, I expect Gwr will be quite different (ie. more interesting?) from Tiemish, simply because Tiemish shamelessly lifts everything from Mandarin, except in the rare cases where I have an original idea (ie. one lifted from a language other than Mandarin). Tiemish is a really desperate attempt to finally produce a halfway complete conlang :). > Tonogenesis is a problem; I know some of the principles, but need to find > good sources for some of the subtleties. I note that in Tiemish, tone seems > to be determined by the _coda_; do the initials have no affect? No, they don't. That in itself doesn't seem too strange to me, but outside of certain rules of thumb (eg. voiceless codas tend to give rise to high tones) and sanity checks (e.g. the contemporary codas -r -l -m -n -N and their syllabic equivilants, transcribed -re -le -me -ne -nge, come from historic -r -l -m -n -N, respectively), the details of the tonogenesis process are more or less arbitrary, and guided by the principle of having roughly the desired proportions of each coda in the contemporary lang. So, I haven't been worried about subtleties: this language mainly exists in order to become fodder for sound-changes. My past experience with only-slightly-elaborated conlangs (Tetelgen, especially, though Ligutniis`at was even less elaborated) suggests that I need more than just vocab - I need extensive texts - in order to produce interesting children. With only vocab, I can produce reflexes of individual words, but can't necessarily develop the syntax in any way short of reinventing it. > The proto sound system was: p t k q, b d g, m n N, w r l y s h, Vowels i a > u. All C could occur as initial, medial, final. > This developed to modern BD: p t ts tr tS k q, b d dz dl dZ g, f s S x, w l > y s h, Vowels i e E/& 1 a u o O and vocalic r [3^]. The only codas are ? N h > w y Very nice - I like the glottal stops, and the [3^] (very Mandarin! Or at least, Beijing). > Sample (easy!) derivations: > > *pátV > **pat > pa? high tone > *pVtá > **(p)ta > ta high > > *bátV > **bat > ba? low > *bVtá > **(b)ta > ta low (alternative? da high?) Another similar with Tiemish, here (though it's only briefly mentioned on the page). At least, many compounds will give clusters like 'pt' etc. I guess this will happen when a modifier-modified compound undergoes syncope of the first syllable, while non-syncoping groups will occur when things get suffixed onto the head. Needs lots of thought, though. In particular, I expect the common enough use of classifiers *after* the noun (with indefinite force) will give rise to common noun-endings in the daughter langs. > Well, I need to get this organized a lot more before creating a web page. I look forward to reading more :). The web page for Tiemish is more or less a copy of my working document (hence the struck-though paragraphs, etc). I usually get inspired on paper, and work out details on computer. s. -- Stephen Mulraney [log in to unmask] http://ataltane.net In 1869 the waffle iron was invented for people who had wrinkled waffles.