Suggested stress pattern: Word-initial, or near the beginning of the word (maybe there's derivational prefixes that never take stress?).
Hmm... what is -s? Some sort of habitual-aspect marker for only auxiliary verbs?
soŝn-a ga-s wek'ne
Soŝna-NOM NEG-HAB catch.fish
'Soshna doesn't catch fish.'
soŝn-a ga moci wek'ne
Soŝna-NOM NEG want catch.fish
'Soshna doesn't want to catch fish (at the moment).'
soŝn-a ga moci-s wek'ne
Soŝna-NOM NEG want-HAB catch.fish
'Soshna never wants (or doesn't usually want) to catch fish.'
Some suggested morphological distinctions:
- determiners: number (singular, plural), case
- - probably with some syncretism
- - number: singular, plural (with an affix), supraplural (meaning "many many Xs", and let's make it easy using full reduplication of the plural: "x.PL x.PL", or maybe "x.PL from x.PL")
- - - - something similar to the supraplural is attested for some collective nouns in Arabic, where collectives can take plural morphology
- - cases: NOM, GEN, INST (instrumental)
- - - - this is inspired from Georgian, which doesn't have an accusative
- - - - for direct objects we could use NOM in positive verbs or sentences, and GEN in negative ones, cf. de in French j'ai un problème 'I have problems' vs. j'ai pas de problème 'I don't a(ny) problem'.
- - - - a paradigm of 2 numbers x 3 cases seems good and fairly easy to learn to me, even with morphophonological complications
- verbs: no idea yet, but maybe distinguish habitual, present, past, irrealis at least?
Doesn't look much like Kabyle, but it looks good. Just let me Kabyle-ize it a little bit, adding [θ ð qʼ qʰ ħ ʕ].
/pʼ pʰ b tʼ tʰ d tʃʼ tʃʰ dʒ kʼ kʰ ɡ qʼ qʰ/ <p' p b t' t d c' c j k' k g q' q>
/f θ ð s ʃ x ħ ʕ h/ <f þ ð s S x H 3 h>
/w j/ <w y>
/m n ŋ/ <m n N>
The use of <H> is from Klingon and Arabic, <3> /ʕ/ is from Kabyle ɛ and Arabic, <N> is sometimes used for /ŋ/ when romanizing languages from India. Regarding /ʃ/ <S>... do we allow /sh/ to happen at all? If we don't we could simply use <sh>. I can type ŝ or ś or š easily, but I think of our mates on Windows 10...
Alternatively, /ʃ ħ ʕ ŋ/ <S H 3 N> can be written with the IPA symbols. We should agree on only using one set or the other when making the Annie dictionary though.
To Kabyle-ize this I'll just take out a bunch of vowels. What do you think of cutting the length distinction down to the Persian system?
The phonemes, with suggested romanization:
/i e æ ɑ o u/ <i e æ a o u>
We could kind of keep your idea about /ɛ/ by saying /æ/ is always unchecked (can only appear in open syllables), and if it gets morphologically checked then it becomes /e/ <e>. So if we have a singular-plural paradigm like sg. */kæbi/, pl. */kæbnu/, then we get /kæbi/ and /kebnu/.
Some random words:
<Slereah> Oh god
<Slereah> I just noticed that I sparked the collablang
<Slereah> What have I done
I don't particularly care about the typology of the conlang. I'd be okay with something very analytic or something very synthetic, something with many affixes or something with a lot of root compounding, something with European-style standard negation (a single invariable particle before or after the verb) or something with a more exotic kind of standard negation, say, negative inflectional affixes that have a different tense system from that of positive (non-negative) verbs (cf. Mandarin bu4 and mei2, which don't match directly with positive zai4, zhe and le).
Just don't make the vocabulary have too many unpredictable stems per word. Learning another Ancient Greek with 6 stems per verb, rather unpredictable for most roots, will make me unhappy. 1-2 stems, and maybe 3 for a few words, is as far as I'll go...