Another development thread. I have one over on the CBB, so I may as well put it here also. Starting with what I already have, and I'll keep this thread updated with the development process from there. Also, I'm writing most things in [brackets] because this is very hard to analyze.
Consonants:
Code: Select all
pʰ tʰ cʰ kʰ
p t ts ʈʂ c k q (ʔ)
s ʂ ç x xʷ χ (h)
ⱱ l ɣ
w (ð̞) j
r̥
r ʀ
h only appears word-initially (en [hen˩]) and word-finally from earlier */β ð ɣ/ (zigg [ˈt͡sɛh˩]).
ð̞ is an allophone of /t/ in unstressed position. The other approximants, /w j ɣ/, also appear as allophones of stops, but they can also appear in stressed position. (No examples in vocabulary yet, but hypothetical examples are easy to come up with: ebvonn [hɪˈwɞn˩], echdon [hɪˈjøn˩], echdonn [hɪˈɣɞn˩].)
Also, ɣ palatalizes to [j] before nonlow front vowels. Not sure whether this will happen with other consonants. t palatalizes to ts.
There will probably be some sort of allophony with r̥ since I don't like it.
Vowels:
Code: Select all
iy ʊ u
eø
ɛ ɞʌɔ
a
aɔ ʊi ɛa ɛɔ ie yø uo ʌi øy ɔu
* short vowels, [ɛ ɞ ʌ ɔ ø ʊ], are written as followed by double consonants. I'm probably going to explain this as some sort of areal feature where weird things happen to geminates, since a neighboring lang had geminates > preaspirates. Maybe preglottalization, whatever.
* long vowels, [a aɔ e ø ʊi i y u], are the default.
* overlong vowels, [ɛa ɛu ie yø uo ʌi øy ɔu], are... well, that's a bit complicated. Basically, they came from diphthongization of vowels given compensatory lengthening from the loss of a resonant in a consonant cluster. It's safe to assume that, when you see a consonant cluster, it's probably pronounced as only one consonant, and if the cluster starts with a resonant (<l r v j m n> and maybe <b d g>; I'm not sure about them yet), the preceding vowel is overlong.
Also, note that [ø] can be both a short vowel and a long vowel.
[ɛɔ] is [ɛu] before labials and velars.
Now, how do you write those vowels, you ask? Well, that's pretty simple:
Code: Select all
short long overlong
ä ʌ a ɛa
a ɔ aɔ ɛu
e ʌ e ie
ö ʊ ʊi yø
o ɞ ø uo
i ɛ i ʌi
y/ü ø y øy
u ʊ u ɔu
The mid/high diphthongs, [ie yø uo ʌi øy ɔu], can also be written <ie yö uo ei öy ou>. This is the only case where <y> and <ü> aren't interchangeable. My excuse here is that they originally represented two different vowels, which eventually merged.
There are also four unstressed vowels, [ɐ ɪ ʏ ʊ], written <a e y u>. (The <e> is probably a result of orthographic influence from Serhes Kettw, which only has two vowels, written <a e>.) [ɐ] can also come from <er el> (satteger [ˈsʌ̤i̯˧˥jɐ]). The last three take glides when word-final: Arve [ˈhɛu̯wɪj], not *[ˈhɛu̯wɪ].
Prosody:
Arve doesn't allow two adjacent syllables in a word to both be unstressed. When that happens, one of the syllables drops out, but it leaves its tone behind. I'll work this out more later.
Tone:
Every word has an underlying sequence of tones. That sequence is mostly predictable from the orthography: if a syllable ends in a voiced consonant, it has a low tone; if not, it has a high tone. But it's much more complicated than that, because only stressed syllables can carry phonemic tone (realized unstressed syllables can carry their tone, but it's always predictable), and stressed syllables can only carry two tones. Two-tone sequences are realized as contours; there are four tone contours: low rising, low falling, high rising, and high falling. I'm not sure how to write this up, so here are some examples, drawn from the copula conjugation chart that I've been writing up: (Also, tone is assigned from left to right, stress is marked here with an underline, and dropped syllables are italicized, as are dropped tones in the tone syllabification. There are more dropped syllables here than there would be in most other places.)
en tegnas
Syllabification: en tegnas
Tone sequence: LLH
Tone syllabification: (LL)H
Realization: [ˈr̥ɛŋs˩˧]
die tegnan
Syllabification: die tegnan
Tone sequence: HLL
Tone syllabification: (HL)L
Realization: [ˈt͡sɛŋ˧˩ɐ̃n]
satte
Syllabification: satte
Tone sequence: HH
Tone syllabification: (HH)
Realization: [ˈsʌ̤i̯˧˥]
en sattes
Syllabification: en sattes
Tone sequence: LHH
Tone syllabification: (LH)H
Realization: [ˈr̥ʌ̤i̯s˧˥]
Orthography:
I'll get around to this later. It's complicated, and I haven't worked it out that much yet. Just remember the consonant cluster rule that I mentioned earlier. Also, it's probably safe to guess that anything with <s> is either [ʂ] or [ç], anything with two alveolars is a trill, anything that's obviously palatal is palatal, and anything else is some sort of uvular.
Copula:
Arve uses periphrastic constructions with the copula for most things. What I have so far is here, although that's out of date. Here's the perfective active conjugated for person and number:
Code: Select all
SINGULAR
positive past tegna ˈt͡sɛŋ˩˧ en tegnas ˈr̥ɛŋs˩˧ die tegnan ˈt͡sɛŋ˧˩ɐn kes tegnar ˈçɛŋ˧˩ɐχ
positive present satte ˈsʌ̤i̯˧˥ en sattes ˈr̥ʌ̤i̯s˧˥ die satten ˈsʌ̤i̯n˥˧ kes satter ˈçʌ̤i̯˥˧jɐ
positive future terch ˈt͡sʌχ˥ en terches ˈr̥ʌχ˧˥ɪs die terchen ˈt͡sʌχɪn˥˧ kes tercher ˈçʌχ˥˧ɐ
negative past san ˈsaɔ̯n˩ en sanz ˈr̥aɔ̯nt͡s˧˥ die sann ˈsʌn˧˩ kes sander ˈçɛɔ̯r˧˩ɐ
negative present zigg ˈt͡sɛh˩ en ziggs ˈr̥ɛç˩˧ die ziggen ˈt͡sɛj˧˩ɪn kes zigger ˈçɛj˧˩ɐ
negative future sambe ˈsɛɔ̯f˩˧ en sambes ˈr̥ɛɔ̯vɪs die samben ˈsɛɔ̯v˧˩ɪn kes samber ˈçɛɔ̯v˧˩ɐ
PLURAL
positive past tegnes ˈt͡sɛŋks˩˧ vön tegnes vʏˈr̥ɛŋs˩˧ den tegnes ð̞ɪˈr̥ɛŋs˩˧ kors tegnes ˈʂɛŋ˧˩ɪs
positive present sattes ˈsʌ̤i̯s˧˥ vön sattes vʏˈr̥ʌ̤i̯s˧˥ den sattes ð̞ɪˈr̥ʌ̤i̯s˧˥ kors sattes ˈʂʌ̤i̯s˥˧
positive future terches ˈt͡sʌχ˥ɪs vön terches vʏˈr̥ʌχ˧˥ɪs den terches ð̞ɪˈr̥ʌχ˧˥ɪs kors terches ˈʂʌχ˥˧ɪs
negative past sanz ˈsaɔ̯nt͡s˩ vön sanz vʏˈr̥aɔ̯nt͡s˧˥ den sanz ð̞ɪˈr̥aɔ̯nt͡s˧˥ kors sanz ˈʂɛɔ̯nt͡s˧˩
negative present ziggs ˈt͡sɛç˥ vön ziggs vʏˈr̥ɛç˩˧ den ziggs ð̞ɪˈr̥ɛç˩˧ kors ziggs ˈʂɛç˧˩
negative future sambes ˈsɛɔ̯v˩ɪs vön sambes vʏˈr̥ɛɔ̯vɪs den sambes ð̞ɪˈr̥ɛɔ̯vɪs kors sambes ˈʂɛɔ̯v˧˩ɪs